_ in ' ±: If) ^ CO

ygn LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHilWS SBIBVyan LIBRA
"Z . to Z > to 2 to Z ,

^ I

ITION NOIiniliSNI NVIN0SHims‘^S3iavyan^LIBRARIEs‘^SMITHS0NlAN INSTITUTION NOlini
to r; > to — in — in

yan LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHIIWS S3IUVaan LIBRA

JTION NOIiniliSNI NVINOSHimS SBIHVaan libraries SMITHSONIAN INSTITUTION NOlin.
to z to z to z V.- to

CO CO 1$^ S3 to in CO

> 5 > X^iixs^ 5 >

yan LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHlIWs'^Sa I y Vy a 11^ LI B RA

to Z CO — CO

..A/ H ^ =! g:

o 'v«\d5>^ “ o ^ O o

JTION^ NOlinillSNl’^NVINOSHimS S3iyvyan”^LIBRARI ES^ SMITHSONIAN"^ INSTITUTION ^NOlinj

Z ^ X, Z r- z

9 m 2 ° rr, z:

CD
33
>

33 V-t,\

cn '■ \^' ^ XjvAstj^ m ^ m

CO ' • ± CO ± CO ^

yan LIBRARIES SMITHSONIAN INSTITUTION NOIiniliSNI NVINOSHIIINS S3iyvyan LIBRA
’ z * to 2 to -

^ -

^ I 1 I XW I I It.. J!) i f X

CO Z ^ > z CO X

TION NOIinillSNI_NVINOSHllINS S3iyvyaiT LIBRARIES SMITHSONIAN INSTITUTION NOIIOJ

CO — \ to — CO ~ CO

to

o

-J X _j z _j 2

yan libraries smithsonian institution NoiiniiisNi nvinoshiiws saiyvyan libra

I” z

o

X z ■- \ z

jTiON NOIiniliSNI NviNOSHiiiNS S3iyvyaii libraries smithsonian institution NOlin.

CO z to z ,,y . to Z .V

- 2 < 2 , < *>N. 2

• S zKi22S?X 1

•A^

— CO _ c/> _ CO =

HSONIAN INSTITUTION NOliniliSNI NVINOSHillNS SBiyVyaiT LIBRARIES SMITHSONIAN INS!
^ ^ r ^ 2: ^ > ^ ^ ^ z: r,

i | /' •^■’ ^''' I |- ' i I 5

CO ■ •■■" Z CO Z <0 ■^' Z (O

josHilws SHiyvyan libraries smithsonian institution NoiiniiiSNi_NviNOSHims sai

— CO — CO ^ 't CO

CU i.i Z

..•?? C

/ ■' :j

O * O

Z _i Z -j

rHSONIAN_INSTITUTION NOIiniliSNI NVINOSHiliNS S3iyvy9l1 LIBRARIES SMITHSONIAN INS1

Z f- z r- Z r-

O

CO ? z D 2 m

josHilws S3iyvyan libraries smithsonian institution NoiiniiiSNi nvinoshiiws S3i

w Z W. to Z V- to z to

— ^ ')' ” ^^\'\ — ^

,yy./ /yy, ? z! z =i z

s i ^ I W i i ^

to »z CO "Z trt z

‘HSONIAN institution NOliniliSNI NVINOSHIIINS S3iyvyan libraries smithsonian INS'

5 . CO Z CO

o uv-^ NiOiir^JiX o " — ^C.UiiS]>- q

IOSHilWS_S3l yvy an^LI BRAR IES^SMITHS0NIANj;^INSTITUTI0N^N0linillSNI^NVIN0SHiIlNS^S3 I

TO / 2 = \v O

5^ ^ 2 fi g

> Ip 351 i_ jyy'i >

■ Z)

XI

mm. \ 't^^LAIkS^^Jr/ I *,. ^ \

m 2 m

CO _ to ^ to

rnsONIAN INSTITUTION NOlinillSNI NVINOSHIIWS S3iyvyail LIBRARIES SMITHSONIAN INS'

_ _ Z to Z to Z > CO z .

sj X .>c » o E 0'- E Xw 5

i I " i i -w- >

</i Z to Z CO ■'^‘ Z CO

40SH1IINS S3iyvyan libraries smithsonian institution NOlinillSNI NVINOSHIIINS S3I

to ^ — tf> — CO z

uJ xt^STiroX TZ ui ?r uj r: V. UJ

Z _JI Z -I z

THSONIAN INSTITUTION NOlinillSNI NVINOSHIIIAIS S3iyvy9l1 LIBRARIES SMITHSONIAN INS

z r- z •“ 2; r-

✓c^^^ovTX O . o >-<ucvAr\ CD

03
33
>

33

m '<<c;oc^ w m ^ to n r; m

to £ y, \ E CO — to

MOSHiiiNS saiyvyan libraries smithsonian institution NoiiniiiSNi nvinoshiiins S3i

tO Z,,.tO Zvy-tO Z to

^ ^ I -■

Palaeontology

VOLUME 16 PART 3 AUGUST 1973

Published by

The Palaeontological Association • London

Price £5

THE PALAEONTOLOGICAL ASSOCIATION

The Association publishes Palaeontology and Special Papers in Palaeontology. Details of member-
ship and subscription rates may be found inside the back cover.

PALAEONTOLOGY

The journal Palaeontology is devoted to the publication of papers on all aspects of palaeontology.
Review articles are particularly welcome, and short papers can often be published rapidly. A high
standard of illustration is a feature of the journal. Four parts are published each year and are sent
free to all members of the Association.

SPECIAL PAPERS IN PALAEONTOLOGY

This is a series of substantial separate works. Members may subscribe to the Series or obtain
individual copies at reduced rates. The following Special Papers are available:

1. (for 1967): Miospores in the Coal Seams of the Carboniferous of Great Britain, by a. h. v. smith and

M. A. BUTTERWORTH. 324 pp., 72 text-figs., 27 plates. Price £8 (U.S. $22.00), post free.

2. (for 1968): Evolution of the Shell Structure of Articulate Brachiopods, by a. williams. 55 pp., 27 text-
figs., 24 plates. Price £5 (U.S. $13.00).

3. (for 1968): Upper Maestrichtian Radiolaria of California, bv Helen p. foreman. 82 pp., 8 plates.
Price £3 (U.S. $8.00).

4. (for 1969): Lower Turonian Ammonites from Israel, by R. freund and m. raab. 83 pp., 15 text-figs.,
10 plates. Price £3 (U.S. $8.00).

5. (for 1969): Chitinozoa from the Ordovician Viola and Fernvale Limestones of the Arbuckle Moun-
tains, Oklahoma, by w. a. m. jenkins. 44 pp., 10 text-figs., 9 plates. Price £2 (U.S. $5.00).

6. (for 1969): Ammonoidea from the Mata Series (Santonian-Maastrichtian) of New Zealand, by
R. A. HENDERSON. 82 pp., 13 text-figs., 15 plates. Price £3 (U.S. $8.00).

7. (for 1970): Shell Structure of the Craniacea and other Calcareous Inarticulate Brachiopoda, by
A. WILLIAMS and K. D. WRIGHT. 51 pp., 17 text-figs., 15 plates. Price £1-50 (U.S. $4.00).

8. (for 1970); Cenomanian Ammonites from Southern England, by w. j. Kennedy. 272 pp., 5 tables,
64 plates. Price £8 (U.S. $22.00).

9. (for 1971): Fish from the Freshwater Lower Cretaceous of Victoria, Australia, with Comments on
the Palaeo-environment, by m. waldman. 130 pp., 37 text-figs., M plates. Price £5 (U.S. $13.00).

10. (for 1971): Upper Cretaceous Ostracoda from the Carnarvon Basin, Western Australia, by R, h. bate.
148 pp., 43 text-figs., 27 plates. Price £5 (U.S. $13.00).

11. (for 1972): Stromatolites and the Biostratigraphy of the Australian Precambrian and Cambrian, by
m. r. WALTER. 268 pp., 55 text-figs., 34 plates. Price £10 (U.S. $26.00).

12. (for 1973): Organisms and Continents through Time. A Symposium Volume of 23 papers edited by

N. F. HUGHES. 340 pp., 132 text-figs. Price £10 (U.S. $26.00) (published with the Systematics Asso-
ciation).

SUBMISSION OF PAPERS

Typescripts on all aspects of palaeontology and stratigraphical palaeontology are invited. They
should conform in style to those already published in this journal, and should be sent to The
Secretary, P.A. Publications Committee, Department of Geology, Sedgwick Museum, Downing
Street, Cambridge, CB2 3EQ, England, who will supply detailed instructions for authors on
request (these are published in Palaeontology, 15, pp. 676-681).

The Palaeontological Association, 1973

Cover illustration : Ancyrodella element (Conodont), Cashaqua Shale, Upper Devonian,

New York State, x70.

THE EYES OF ASAPHUS RANICEPS
DALMAN (TRILOBITA)

by E. N. K. CLARKSON

Abstract. The holochroal eyes of the Lower Ordovician trilobite Asaphus raniceps Dalman have been studied using
light and electron microscopy.

In these eyes the refractive elements are elongated calcite prisms underlying a cornea which is continuous with,
although structurally dissimilar to, the ‘outer cuticular layer’ described by Dalingwater. The prisms are orientated
with their c-axes normal to the surface. Some of the material studied (from Oland) showed the effects of at least two
phases of diagenesis, which in one case had resulted in the production of secondary prisms growing syntaxially on
the primary prisms, and confusingly similar to primary structures.

The visual surface approximates a segment of an almost perfect spheroid whose radii of curvature in vertical
and horizontal planes all converge to a single point, in a manner very similar to that of some superposition eyes
in modern arthropods, with which analogies are drawn.

Problems in the use of calcite as a primary refractive medium are discussed, and it is concluded that the effects
of birefringence could have been minimized by suitable pigment screens, like those of insects and crustaceans, under-
lying the prismatic layer.

The ‘sensory fossettes’ on the eye-socle are craters, each with a central perforation communicating with the internal
surface.

Trilobites of the family Asaphidae have distinctive compound eyes whose range
in form is quite well known from palaeontological literature. These eyes, often well
preserved, are usually rather large and prominent, and often rise above the glabella;
one remarkable species, Asaphus kovalevskii Lavrov, has eyes situated upon very
elongated bases resembling long stalks. Asaphid eyes were referred to by various
authors working in the nineteenth and early twentieth centuries; Schmidt’s (1904)
monograph, for instance, contains good illustrations. More recent authors have also
described and figured asaphids with intact eyes, and Whittington’s studies (1963,
1965) included many details of external eye morphology. Hupe (1953) has provided
an excellent figure of the eye of Asaphus cornutus (Pander), reproduced in the ‘Trea-
tise’ by Harrington (1959), and Rose (1967) has shown that in N ileus and Isotelus
growth of the visual surface is accomplished by the addition of new lenses round
the lower margin of the eye.

Although we possess a reasonably good understanding of the range in form and
external morphology of asaphid eyes, there have been only two serious attempts to
investigate their internal morphology; the first being that of Lindstrom (1901), who
described sections and fracture surfaces of the eyes of several Scandinavian asaphids.
He showed that the refractive elements were elongated prisms (rather than lenses)
underlying the cornea, and gave a good account of their anatomy. Balashova (1948)
confirmed that the eye had a prismatic structure, and further indicated that the
eye-socle (Lindstrom’s reticulate or spongious zone) was permeated with very fine
pore-canals and that there were ‘fossae’ on the socle opening downwards into fine
calcite-filled tubes, to which she imputed a sensory (tactile) function. The advent
of the scanning electron microscope stimulated further study of the eyes of asaphids,
and, as expected, revealed many details invisible to Lindstrom.

[Palaeontology, Vol. 16, Part 3, 1973, pp. 425-444, pis. 48-50.]

426

PALAEONTOLOGY, VOLUME 16

In 1967 Dr. John Dalingwater kindly sent me a number of finely preserved speci-
mens of Asaphus which he had collected from Bohlin’s (1949) locality where the
glauconitic 'raniceps' limestone is exposed at the cliff of Haget, northern Oland.
Tjernvik (1972, p. 305) gives the age of this limestone as lower Llanvirnian (bifidus
zone). Specific identification of these was somewhat difficult as most of the specimens
were fragmentary, but Dr. Dalingwater and I agree that they most closely approxi-
mate A. raniceps Dalman, sensu Angelin (1878, p. 53).

During the investigation of these eyes, using light and electron microscopy, it
became evident that different specimens had been variously affected by diagenesis.
This made the interpretation of the original structure difficult, for it was not imme-
diately apparent in all cases which structures were primary and which were the
results of secondary recrystallization. In one specimen, for instance, there were
radially arranged microstructures extending quite deep inside the eye, these were
so regularly formed that they could have been mistaken for primary structures, but
they proved, in fact, to be secondary, growing syntaxially upon primary elements
of the ‘refractive’ zone.

Part of this study has been therefore orientated towards an understanding of the
nature of primary structures and how they were affected by diagenesis; the rest is
more closely concerned with the organization of the eye as a functional visual organ.

The specimens were prepared for examination as follows. The prefix ‘Gr T refers
to the collections of the Grant Institute of Geology, University of Edinburgh.

External surface only. Gr I 5501.

Internal structure using thin sections, polished surfaces, and cellulose peels. Gr I 5502, 5503, 5510,

5511, 5512.

External and internal features (fracture surfaces and etched sections) using the Stereoscan. Gr I

5504, 5505, 5506, 5507, 5508, 5513.

THE CUTICLE OF ASAPHUS

Dalingwater (in press) in a study of the structure of trilobite cuticles has shown
that Asaphus raniceps (from the same locality as my material) has a cuticle of two
distinct layers. The outer layer, less than one-tenth of the total thickness, is com-
posed of fairly regular perpendicular crystallites which have a fibrous appearance.
The thick inner area is less distinctly structured, and no individual crystallites could
be seen. Neither layer extinguished uniformly in polarized light implying that the
calcite of which the bulk of the cuticle is composed does not occur in regularly
arranged crystallites. There was also some organic matter remaining, which could
be isolated by decalcifying the cuticle with EDTA.

These two cuticular layers have their direct counterparts in the eye. The thin
outer layer passes laterally into the cornea, losing its fibrous appearance at the peri-
phery of the eye-socle, and becoming thinner. The thick inner cuticular area is
directly equivalent to that part of the eye underlying the cornea, consisting of large
hexagonal prisms of calcite, which acted as refractive units, directing light to the
photoreceptive organs below. These large prisms unlike the inner cuticular area are
regularly structured and have their c-axes orientated near normal to the outer surface
of the eye.

CLARKSON: ASAPHUS EYES

427

VISIBLE STRUCTURES IN THE EYE OF ASAPHUS RANICEPS

External surface. The external form of the eye (text-fig. \a-c) closely approximates
that of Asaphus cornutus Pander, from the Ordovician of Estonia, figured by Hupe
(1953, p. 77, fig. 31) and reproduced by Harrington (1959, p. 0.88, fig. 64i). It is
large and strongly curved in plan, projecting well above the glabella. The visual
surface, which has a much higher profile curvature anteriorly, is situated upon a
vertical ‘eye-socle’ rising abruptly from the librigena {sensu Shaw and Ormiston
1964), upon which are shallow funnel- or basin-shaped cavities, irregularly distri-
buted and decreasing in size towards the base of the eye-socle. These were described
by Hupe as sensory fossettes. The facial suture is semicircular, separating the visual
surface from the palpebral lobe, which slopes sharply down to the glabella, and
carries terrace-line ornamentation.

There is a thin pellucid cornea covering the surface, merging laterally into the
outermost layer of the cuticle (text-fig. 4c; PI. 50, fig. 9). Through this cornea the
many quadrate or hexagonal lenses can be seen by translucence, especially if the speci-
men is immersed in a medium of high refractive index. Hupe’s specimen showed
patches of larger irregularly distributed lenses, which he thought had resulted from
damage during ecdysis.

Even with the Stereoscan the external surface of the best-preserved specimens
appears to be smooth and structureless; a microgranular effect is not evident until
magnifications of over x 500 is reached. Relative granularity, however, varies
according to the quality of preservation of the material.

Internal structure. Lindstr6m(1901, pp. 28, 37-43, pi. l,figs. 8-30) gave a good account
of the eye structure in asaphids as known to him, and noted the following main
points. Welded to the inner surface of the cornea are the refractive organs, appro-
priately termed prisms, which are closely packed hexagonal pillars, arranged radially.
(This zone is here termed the primary prismatic region.) If the eye is fractured the
prisms separate cleanly from one another, and can be seen under low magnification
like columns of basalt lying on their sides. Vertical sections show that the prisms
generally become longer and thinner near the margins of the eye, and especially
towards the lower rim of the visual surface.

Above and below the visual surface the prisms pass into a rather structureless
‘marginal zone’, twice as thick (in section) as the primary prismatic region. Lind-
strom referred to this also as the ‘spongious’ or ‘reticulate’ zone, in view of its spongy
appearance in slightly decomposed specimens. He noted the fossettes on the eye-
socle also but did not impute a sensory function to them.

The use of the Stereoscan has given more information upon the nature of most of
these regions, and most particularly of the primary prismatic layer.

In the best-preserved specimens, each prism is a single crystal of calcite with its
optic axis normal to the surface (text-fig. \d). The optic axes of neighbouring prisms
diverge slightly. Thin sections made horizontally through a complete eye and ex-
amined using crossed nicols show that the prisms undergo extinction in the NS.
and EW. positions; this pattern remains constant on rotation. Tests with a sensitive
tint plate show that the c-axes of the calcite crystals are normal to the surface. It is
difficult to imagine how such a system as this, with regularly diverging optic axes.

428

PALAEONTOLOGY, VOLUME 16

can be other than primary, for although further impregnation with calcite during
diagenesis might take place in optical continuity with the existing crystal lattices,
more extensive recrystallization would very likely destroy the regularity of the pattern,
as has actually happened to some extent in some of the less well-preserved specimens.
The eyes in living specimens of Asaphus raniceps must therefore have had a high

TEXT-FIG. 1. a-c. Left eye of A. raniceps Dalman. x6-5. Gr I 5501, in lateral, dorsal, and posterior views.
The black specks are adherent glauconite grains, d. Diagrammatic construction showing crystallography
of the primary and secondary (diagenetic) prisms underlying the cornea, e-f. Polarization of light rays
passing through a peripheral (e) and a central (/) primary prism. The o-ray (XX) is shown passing along
the long diagonal (XY). For full explanation see text, p. 438. g. Minimal visual field of A. raniceps, from
Gr 1 5501, orientated as in the small diagram (the latter x2l3).

CLARKSON: ASAPHUS EYES

429

proportion of calcite, each prism being a single calcite crystal, presumably associated
with protein and other organic material.

It is not surprising to find calcite used as a structural component in the eyes of
trilobites, for many modern arthropods which have cuticles reinforced with calcite
are found to have calcite in the eye as well. It is not, however, used in the same way,
for although the prisms of A. raniceps have an extremely regular arrangement, there
is no regularity of structure in the rest of the cuticular inner area. The arrangement
of calcite crystals in the eyes and cuticle of many modern arthropods is singularly
irregular. According to Richards (1951, pp. 103-105), crystallization of calcite in
very many modern arthropods begins independently at various loci and each crystal
simply continues growth until it contacts another crystal. This is true for the calcite
crystals within the eye as well, as Diidich (1931) clearly showed; for randomly
orientated calcite crystals cut across the ommatidia at all angles without any rela-
tionship at all to organic boundaries. In A. raniceps on the other hand each visual
unit, or at least the upper part, was individually calcified, and, as argued later,
calcite was probably the primary refractile material.

In thin sections the cleavages in the prisms often appear distinct, each prism having
its own set of cleavages, orientated slightly differently to its neighbours. Not in-
frequently, however, two or three neighbouring lenses in some sections can be seen
to have the same cleavages running through all of them, and the small block or
‘domain’ of lenses goes into extinction as a unit with sharply marked edges. Rota-
tion of the stage thus produces a stepwise rather than a regular extinction. It is likely
that such domains were secreted together and retained their optical continuity
throughout life. It has been suggested to me by C. Eccles that this may have resulted
from a crystallographic constraint, and that neighbouring prisms had to grow in
the same optical orientation as the angle of divergence of individual prisms was too
small to permit crystallographic separation. Only when there was a critical angle
of divergence could another domain grow at a different crystallographic orientation.
It is not clear whether new prisms were always secreted in domains but it is not
unlikely, for the individual prisms often seen in section with separate orientation
could belong to a small domain of three or four lenses of which only one was cut
in the plane of the section.

The tendency for small groups of neighbouring lenses to have the same crystallo-
graphic orientation is shown also in Stereoscan photographs, such as PI. 49, fig. 3
where two adjacent prisms lying below the cornea have cleavages running through
both of them without a break.

The sides of intact prisms as shown in Stereoscan photographs are remarkably
rough, with corrugated granular surfaces (PI. 48, figs. 1-3; PI. 49, fig. 5). These
corrugations are always parallel with the cornea, and are related to the underlying
cleavage directions (text-fig. \d). Usually, though not always, the prisms become
long and thin towards the top and particularly the bottom of the visual surface
(PI. 48, fig. 2). Here they may be up to twice the length in other parts of the eyes.
In some specimens, however, this tendency is far less evident. It is possible that the
former condition occurs only in immature specimens. It is clear that the new visual
units must, as in Phacopina (Beckmann 1951 ; Clarkson 1966a) have been produced
in a generative zone at the periphery of the eye and the work of Rose (1967) confirms

430

PALAEONTOLOGY, VOLUME 16

that they were added only at the base of the visual surface. In Asaphus the genera-
tive zone seems to have lain at the lower junction of the visual surface and the marginal
zone. New prisms were first of all the same width (in section) as the marginal zone
and very thin; they shortened and grew thicker as they became functional.

Some of Lindstrom’s specimens showed concentric layering within the lenses.
Such concentric structure was not visible in any of my material, and neither did
etching with dilute acid (PI. 48, fig. 5) reveal any structures other than the cleavages
which were picked out by the acid. The apparent absence of layering may have been
because Lindstrom’s material was less fresh than mine, and that mild weathering
might have revealed primary structures in his specimens which were not clear or
evident in mine though they may have been present.

The sensorial fossettes. Hupe (1953) noted a series of irregularly distributed, shallow
excavations on the eye-socle, which he described as sensorial fossettes. Lindstrom
(1901) had previously noted the presence of these little pits, but imagined them to
be the excavations of some boring organism. Some thin sections and polished sur-
faces made in the present study showed that each fossette is set at the summit of a
narrow canal (now calcite filled) which can be traced to the inner surface of the eye-
socle though little structure can be seen, even in the posterior part of the socle where
the best and largest examples are normally located (text-figs. 4a, b, c; PI. 50, fig. 9).
Stereoscan photographs of the external surface simply show the fossettes as shallow
rimless craters and contribute nothing further to our knowledge (PI. 48, fig. 6).

Hupe’s interpretation of the fossettes as sensory structures seems appropriate;
similar structures in the neighbourhood of some insect eyes are the sites of vibro-
sensory organs whose nerves are connected with the third optic lobe (Burtt and
Catton 1966a). Very many trilobite eye lobes are provided with pit-like structures,
narrow vertical grooves, small tubercles (often only properly visible with the Stereo-
scan), or other such organs, usually located on the eye-socle; these may all be the
sites of sensory organs of some kind, hence the fossettes of Asaphus, though excep-
tionally large, are not unusual amongst the trilobites.

EFFECTS OF DIAGENESIS

Recrystallization of the primary prisms. The course of diagenesis was followed in
thin sections of the eyes of different specimens following Friedman (1964). Some
sections, or parts of sections, showed the original form of the calcite prisms, either
as single crystals or as small domains. In these, the cleavages are always distinct

EXPLANATION OF PLATE 48

Asaphus raniceps Dalman (Stereoscan photographs all of fracture surfaces except figs. 5 and 6).

Fig. 1. Prisms near lower margin of the visual surface, x 110. Gr I 5505.

Fig. 2. Elongated prisms near the lower margin of the visual surface, x 440. Gr I 5504.

Fig. 3. Corrugated surfaces of prisms in the central part of the eye. x2300. Gr I 5507.

Fig. 4. Similar prisms showing cleavages parallel with the corrugations, x 1 100. Gr I 5507.

Fig. 5. Horizontal ground surface, cutting through prisms and etched with dilute HCl. x525. Gr I 5510.
Fig. 6. Surface of the eye-socle in the posterior region, showing sensory fossettes. x 1 10. Gr I 5507.

PLATE 48

6

CLARKSON, Asaphus eyes

432

PALAEONTOLOGY, VOLUME 16

(PI. 50, fig. 1), and extinction follows a stepwise pattern. The lower ends of the prisms
may be rounded or somewhat ragged in appearance, but never show crystal faces.

Two stages of diagenesis have been differentiated. The first stage involved some
degree of recrystallization of calcite in situ, into a microcrystalline form, resulting
in the crystal boundaries and cleavages becoming indistinct, and blurring the ex-
tinction pattern (PI. 50, fig. 6). This was seen in its early stages in certain parts of
otherwise unaltered eyes. The recrystallized areas (yellow-brown under crossed
nicols) are often darker in colour than the primary crystals.

In some sections diagenesis has been carried a stage further, with a second develop-
ment of microcrystalline calcite invading the already altered primary prismatic
region (PI. 50, figs. 7, 8). This second stage microcrystalline calcite is variable in
colour, but is usually a very light yellow, contrasting with the darker yellow-brown
of the first-stage diagenetic material. Sometimes it appears as randomly orientated
flecks or patches within the prisms. Often it is seen as a ‘front’ which has advanced
into the prismatic region from either surface. On occasion it has picked out the
boundaries of the altered prisms, which then appear as thin yellow lines, and it may
have invaded the interior of each prism from all its edges at once. In such cases all
that is left of the original prismatic layer is a series of elongated kernels (already
altered during the first stage in diagenesis) surrounded by lighter coloured micro-
crystalline calcite of the second stage. These kernels may be regular in appearance,
but are sometimes truncated by a ‘front’ of second-stage microcrystalline material,
where the latter has grown more rapidly in one direction than in others (PI. 50,
fig. 8).

Recrystallization of primary structures is less easy to recognize in fracture surfaces
or in etched sections using the Stereoscan. It has been observed, however, that
whereas certain calcite prisms have sharp, well-defined corrugations on the surface,
parallel with the cleavages, other calcite crystals in the same eye may have much
less regular corrugations, and only a rough granularity to the surface. As unaltered
primary crystals have well-defined cleavages, it would seem likely that the crystals
with roughly granular surfaces have undergone some measure of diagenesis.

Secondary growths below the primary prisms. In one specimen (Gr I 5508), both
fracture surfaces and thin sections revealed inward extensions of the primary prisms.

EXPLANATION OF PLATE 49

Asaphus raniceps Dalman (Stereoscan photographs, all of fracture surfaces).

Fig. 1. Slightly oblique view of primary prisms (corrugated), with smooth secondary (diagenetic) prisms
growing syntaxially upon them, x 180. Gr I 5508 (cf. text-fig. 2).

Fig. 2. Lower terminations of the same secondary prisms and subjacent area of recrystallized calcite.
X 130. Gr I 5508 (cf. text-fig. 2).

Fig. 3. Several prisms underlying the cornea, visible as an upstanding wall near the top of the photograph.
Two adjacent prisms are fractured showing a common cleavage direction running through both, x 325.
Gr I 5506.

Fig. 4. Junction between primary and secondary prisms (cf. PI. 49, fig. 1). x485. Gr I 5508.

Fig. 5. Enlarged surface of corrugations on the outer surface of a prism (cf PI. 48, fig. 2). x 2400. Gr I
5504.

Fig. 6. Prisms near the lower margin of the visual surface. x65. Gr I 5505.

PLATE 49

CLARKSON, Asaphus eyes

434

PALAEONTOLOGY, VOLUME 16

Stereoscan photographs showed these to be of extremely regular form, being syn-
taxial, pillar-like extensions of the outer prisms (PI. 49, figs. 1, 2; text-fig. 2a, b).
They are found only in certain parts of the eye, and in all the Stereoscan preparations
extend to about the same level. Between and around these areas lie micrite and finely
recrystallized sparite, in patches. These extensions, referred to as ‘inner’ or ‘secondary
prisms’, have very smooth outer surfaces, and the transition from the corrugated
outer zone is sharp, being marked usually by a line of somewhat irregular fractures
(PI. 49, fig. 4). One example showed the edges of a primary prism being met by the
sides of an inner prism, but normally the edges of a primary prism are continuous
with those of a secondary one.

Below the inner prisms (sub-prismatic region in text-fig. 2) are large, equant calcite
crystals with patches of micrite and sparite. Some of these are syntaxial with the
inner prisms. Several orientated crystals of dog-tooth spar cemented together by a
calcitic jacket were noted in one area; these had near-perfect crystal faces.

Though there would be little doubt that all the structures in this sub-prismatic

cornea

primary prismatic region

secondary prismatic region

sub-prismatic region

a

TEXT-FIG. 2. Drawings made from fracture surface of Gr I 5508 showing areas of primary and secondary
(diagenetic) calcite. a. Enlargement of area 1 x 120, b. Fragment of eye showing the two areas of secon-
dary prisms, one of which is enlarged in a. x 18.

CLARKSON: ASAPHUS EYES

435

region are secondary, the regular appearance of the inner prisms in the Stereoscan
photographs might suggest that at least these could be of primary origin, though
other Stereoscan evidence is equivocal. Thin sections, however, indicate that the
inner prisms are very probably secondary. The series illustrated in text-fig. 3u-c are
vertical sections made obliquely through the posterior region of the eye of Gr I 5508.

Text-fig. 3a (PI. 50, fig. 3) shows a cellulose acetate peel stained with methylene
blue (Dickson 1966). The primary layer has been differentially affected by first-
stage diagenesis, though crystal boundaries are clear in places. There is a pronounced

TEXT-FIG. 3. Effects of diagenesis shown in thin sections and cellulose peels of Gr I 5502. a. Oblique vertical
section near posterior margin stained with methylene blue, showing primary and secondary prisms.

b. Similar section, stained with alizarine red-S and acid fuchsine, with very irregular secondary prisms.

c. Thin section through the same area, showing complete recrystallization of the secondary and partial

alteration of the primary prisms. All x 20.

line of demarcation between the primary and the inner prisms. The latter, though
regular in places, are elsewhere of differing lengths, and sometimes overgrow one
another. They show clear evidence of secondary growth on the bases of the primary
prisms in that a record is left of the past position of the euhedral crystal faces. The
structure becomes very irregular at depth, but even here some crystals are still more
or less syntaxial with the primary prisms. Similar, though much smaller, secondary
growths were visible in other calcitic shells in the same section.

Cellulose peels stained with alizarine red-S and acid fuchsine (PI. 50, fig. 2; text-
fig. 3b), through the same general region but more posteriorly, showed much the
same kind of structural elements but picked out slightly different details. The section
illustrated was more strongly affected by second-stage diagenesis, and the boundary
between primary and secondary structures was indistinct. In general, there was far
less regularity, which emphasizes the secondary nature of the inner prisms. This
section also showed a fringe of small secondary calcite crystals, elongated and with
axes normal to the surface of the cuticle, growing on the inside of the eye-socle.
These are clearly analogous to the inner prisms.

In an optical thin-section (PI. 50, fig. 8; text-fig. 3c), there are very large elongated
euhedral crystals, growing below the primary layer, each encompassing the bases

436

PALAEONTOLOGY, VOLUME 16

of several (altered) primary prisms. They are in optical continuity with the latter,
though these are altered by first-stage diagenesis, but they are not in optical con-
tinuity with the light yellow material of the second diagenetic stage. Probably these
large euhedral crystals resulted from the coalescence of several smaller crystals
which originated in contact with the primary prisms.

It is clear from the foregoing observations that all the calcitic elements below the
primary prismatic layer are of secondary origin. The great regularity of their struc-
ture in some parts of the eye (which is not always maintained in other regions) is
merely a reflection of the regular arrangement of the large prisms above, which pro-
vided suitable foci for continued growth of calcite, provided that there was a void
below. It may have been that there were partially calcitized elements below the
primary prisms which could have controlled the direction of subsequent calcitiza-
tion, but there is no direct evidence of this.

VISION IN ASAPHUS RANICEPS

Visual field. A provisional determination of the visual field has been made, using
similar apparatus and techniques to those described formerly (Clarkson 1966a, b).
Since the prisms are normal to the surface it was possible to use a graticule and pro-
tractor to measure the inclination of the upper and lower margins of the visual sur-
face, every 10° of longitude from front to rear, in a manner comparable with the
measurement of the axial bearings of individual phacopid lenses. Plotting these
inclinations on a Lambert net gave an angular range of vision quite similar to that
of many phacopids (text-fig. Ig), overlapping at front and rear to give some degree
of binocular vision. The relatively narrow latitudinal extent of vision may be con-
trasted with that of many holochroal eyes, where latitudinal ranges of up to 120°

EXPLANATION OF PLATE 50

Asaphus raniceps Dalman (Photomicrographs of thin sections, except Figs. 2 and 3 which are acetate
peels).

Fig. 1. Horizontal section through eye showing the cleavages running through the prisms and the rounded
terminations of the latter. Plane polarized light, x 80. Gr I 5512.

Fig. 2. Oblique vertical section (acetate peel) stained with alizarine red-S and acid fuchsine, with very
irregular secondary prisms. x65. Gr I 5502 (cf. text-fig. 36).

Fig. 3. Similar section (acetate peel) stained with methylene blue, showing primary and secondary prisms.
The latter are rather irregular and exhibit growth lines, x 65. Gr I 5502 (cf. text-fig. 3a).

Fig. 4. Oblique vertical section showing primary prisms only, x 18. Gr I 551 1.

Fig. 5. The same under crossed nicols. x 18. Gr I 551 1.

Fig. 6. Part of vertical section through an eye somewhat altered by first-stage diagenesis, showing primary
prismatic layer, with indistinct prisms, and the thin cornea (lower part of this section in fig. 9). X 40.
Gr I 5503 (cf. text-fig. 4c).

Fig. 7. Part of a vertical section, showing advanced diagenesis, passing through the uppermost part of
the visual surface (upper marginal zone of Lindstrom). x80. Gr I 5503.

Fig. 8. Oblique vertical thin section showing complete recrystallization of the secondary prisms and
partial alteration of the primary prisms, x 37. Gr I 5502 (cf text-fig. 3c).

Fig. 9. Downward continuation of section in PI. 50, fig. 6, showing inward extension of a sensory fossette
and the upward passage of the vertically laminated outer cuticular layer of Dalingwater into the cornea
at the base of the eye-socle. X 40. Gr I 5503 (cf text-fig. 4c).

PLATE 50

7

CLARKSON, Asaphus eyes

9

438

PALAEONTOLOGY, VOLUME 16

for each eye, with rnarked overlap above, are not uncommon. This plot represents
the minimum possible range of vision, assuming that the peripheral prisms receive
only light coming parallel with the c-axis. The eye may have actually been capable of
receiving light from outside this zone and transmitting it to the photoreceptors;
this would depend upon whether or not screening pigment was present, isolating
the ommatidia as in modern compound eyes.

Optics of the calcite prisms. Calcite was used as a primary structural component in
trilobite cuticle. It was also present in the eye, where it was of particular value, being
rigid, easily secreted in the same way as the rest of the cuticle, and above all refrac-
tive and transparent. But calcite is an anisotropic mineral with the property of double
refraction, and its use in an optical system raises problems. Some modern arthropods
have an irregular mosaic of calcite crystals within the eye as mentioned earlier.

In the following discussion on calcite optics each prism is considered as an indivi-
dual calcite crystal, though it is recognized that each was probably penetrated through-
out by organic matter. How far this would have altered the refractive index, if at all,
cannot be assessed but the birefringent properties of calcite would not have been
eliminated by such interpenetration. The calcite cornea may also have had an organic
association.

Each prism is a single hexagonal crystal with its optic axis (c-axis) normal to the
surface of the eye. Any light rays entering the crystal normal to the visual surface
(i.e. parallel with the optic axis) would be transmitted, unpolarized, straight through
the crystal without any change of direction. A light ray entering obliquely, however,
will be resolved into two linearly polarized rays vibrating perpendicular to each
other. The ordinary ray has constant velocity whatever the direction of incidence,
but the extraordinary ray increases in velocity as the angle of incidence increases
from the normal. Oblique incident rays not only polarize, but produce double images
at different depths. Herein lies the disadvantage of calcite; some interference with
the visual process would be expected, unless there were some system within the eye
for ensuring that only normal or near normal rays were actually let through to the
photoreceptive organs below.

Let us consider the angular light receptivity of each crystal. Median sections of
two crystals, of the dimensions actually found in different parts of the eye, are illus-
trated in text-fig. \e,f. Text-fig. 1/is the typical form, occurring in all but the peri-
pheral regions of the eye, whereas text-fig. Ic is a crystal from the generative zone
hear the lower margin. If each prism is considered as an isolated unit, the most
oblique incident ray which it could transmit would be refracted along the line XY
which is the path and wave-normal of the ordinary ray.

Light entering at a higher angle of incidence would be refracted against the wall
of the prism. The angle of incidence for such a refracted ray travelling along the
line XY can be calculated using Snell’s Law and the following refractive indices
(n water L33, n o-ray = L66). The highest angles of incidence are then 44° and 22°
respectively for the two prisms. If n e-ray ^ 1'48, then using the optical indicatrix
for calcite, the path of the extraordinary ray incident at O may be constructed as
in the diagram; as it travels faster it is less highly refracted. Thus each prism, con-
sidered by itself, has quite a high range of angular receptivity.

CLARKSON; ASAPHUS EYES

439

The prisms in the eye do not, however, exist in isolation, but are contiguous one
with another. It is not known, of course, whether in life they were optically separated
by organic layers along their walls. There is no evidence of such layers, and their
former presence is rendered unlikely both by the close contiguity of the prisms and
the fact that there are small domains of optically associated prisms where the cleav-
ages pass through from one crystal to another. The lenses of modern insect and
crustacean eyes, where contiguous like those of Asaphus, are sometimes optically
isolated one from another by such highly refractive layers but not always, and indeed
the superposition system of vision, as mentioned later, depends upon their close
optical contiguity.

If, therefore, as we assume, the prisms of Asaphus were originally in optical con-
tiguity, then the dioptric system as a whole would actually be receptive to a much
higher range of light incidence, for the light could pass through neighbouring crystals.
In such a case the divergence between the path of the extraordinary and the ordinary
ray would increase, resulting in more extreme double refraction and hence double
images at different depths.

Such double images, however, could have been virtually eliminated and not trans-
mitted to the photoreceptors if the Asaphus eye were provided with a layer of ab-
sorbing pigment, just below the prismatic region, exactly as with the ‘distal retinal
pigment’ of many known crustaceans (text-fig. Ad, e), which occurs in both apposition
and superposition eyes. In the latter, there is a cylindrical sleeve of black or brown
pigment surrounding the upper parts of each ommatidium below the lens, embracing
the crystalline cone and the area below it, which absorbs unwanted light rays. A simi-
lar pigment sleeve in Asaphus located just below the prisms, would fulfil an identical
function, and if the central bore were narrow, would effectively restrict most of the
oblique rays, and screen them from the other ommatidia. If this were the case a
mosaic-type image could then be formed by the eye, using only rays parallel with the
axis or only slightly oblique and thus cutting out most of the adverse effects of double
refraction. Such a system would allow a degree of mosaic vision not greatly inferior
to that of modern arthropods. The extreme sphericity of the visual surface would
give a more or less undistorted image of the mosaic kind, and it might also relate to
a kind of superposition vision, as discussed in the next section.

Arrangement of deeper-lying structures. The visual surface of Asaphus raniceps has
an almost constant horizontal curvature, the radius of which is slightly less than
that in the vertical plane, though the latter is less regular, and decreases towards the
top of the eye. Thus the visual surface approximates a segment of a prolate spheroid
in which the vertical axis is slightly greater than the horizontal ones.

The eye has a slightly higher profile curvature anteriorly than posteriorly, and
the visual field is expanded latitudinally towards the front (text-fig. Ig). Apart from
this slight difference the curvature of the eye is otherwise regular, and follows the
almost perfectly radial arrangement of the primary prisms. In horizontal and verti-
cal sections of well-preserved specimens the edges of the prisms are clearly visible,
and may be used as a basis for some reconstruction of the internal parts. If lines are
drawn along the edges of adjacent prisms and extended inwards, they are found to
meet at a common centre.

440

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 4. a. Horizontal section through A. raniceps eye, with radii of curvature passing through every
tenth prism marked. From a thin section, x 12-5. Gr I 5512. b. Vertical section, x 12-5. Gr I 5503. c. Thin
section through lower part of visual surface and eye-socle. The outer cuticular layer of Dalingwater passes
upwards into the cornea at the base of the eye-socle, x 30. Gr I 5503. d. A single ommatidium of the
apposition eye of a shore crab. e. Ommatidium of the superposition eye of a lobster.

In d and c, the left-hand side is shown as dark-adapted, the right as light-adapted. Both after Kampa
(1965).

/, g. Diagrams illustrating surface convexity and centres of ommatidial convergence in the eyes of the
insect Apis mellifica (apposition eye), and Sarnia cecropia (superposition eye). Both redrawn from Portillo
(1936).

CLARKSON: ASAPHUS EYES

441

The radii of curvature in the two directions are illustrated in text-fig. Aa, b. It is
very probable that those parts of the eye, now destroyed, lying below the primary
prisms, were elongated ommatidia, following the radii of curvature more or less
exactly, just as in the eyes of many insects and crustaceans, though how deep they
were cannot be assessed. In modern arthropods the first optic ganglion of synaptic
region occupies the central part of the underlying space and the same was probably
true of the trilobites. Almost certainly the ommatidia must have terminated some
distance short of the centre of curvature.

The regularity of surface curvature, ommatidial separation, and radial arrange-
ment of the ommatidia, evident in Asaphus raniceps, is also characteristic of the
structure of many superposition eyes in modern arthropods. In this respect the
eye of A. raniceps is unlike both modern apposition eyes, and those of phacopid
trilobites.

Anatomical differences between apposition and superposition eyes are very well
known, and have been synthesized in various reviews (Waterman 1961 ; Goldsmith
1964; Wigglesworth 1965). Apposition eyes (text-fig. Ad) have the rhabdom in
contact with the crystalline cone, whereas in superposition eyes (text-fig. 4c), the
cone and rhabdom are separated by a long cone-stalk, believed to be a light con-
ductor. The rhabdom here is very much shorter than in apposition eyes.

These anatomical differences have been greatly discussed since the time of Exner
(1891), but their physiological significance is still controversial (Goldsmith 1964;
Miller c/ n/. 1968).

Superposition eyes, which are typical of nocturnal crepuscular or deep-water
arthropods, have an elaborate system of adaptation to dark conditions. In light-
adapted superposition eyes, when screening pigment surrounds each ommatidium,
the visual system probably forms a mosaic image rather like an apposition eye.
(Actually the ‘mosaic theory’ of image formation, first propounded by Muller in
1 826, is now held to be too simple, and Burtt and Catton ( 1 962, 1 966a, b) have shown
the important role of diffraction processes in arthropod vision, so as to modify
greatly and extend the mosaic theory.)

In dark-adapted superposition eyes, the pigment migrates away from the cone-
stalk region so that the isolation of the ommatidia is lost. In this way light coming
through many lenses can pass freely to any rhabdom, and is not confined to any
specific ommatidium. Sensitivity seems to be increased thereby, though resolution
is lost.

It is interesting that the only modern arthropod eyes of spherical, geometrically
perfect form with the radii meeting at the centre are of superposition type (Portillo
1936), though not all superposition eyes do, in fact, have such perfect form. Portillo
believed that such a structure was highly desirable for superposition vision, both
optically and physiologically.

The parallels between the regular structure of the eye of Asaphus raniceps, and
typical modem superposition eyes are illustrated in text-fig. Aa, b,f,g-, it is possible
that the number of points in common may imply some degree of functional similarity.

Some recent work by Stuermer (1970) appears at first sight to militate against the
above suggestions. Stuermer subjected the schizochroal eyes of some Devonian
phacopid trilobite to X-ray examination and found within the eyes very long, closely

B

442

PALAEONTOLOGY, VOLUME 16

packed bundles of fibres extending from the lentiferous region right into the axial
part of the body of the trilobite. He interpreted these as light guides (presumably
equivalent to very long cone-stalks in modern arthropods), which conducted light
from the lenses to the photoreceptors.

I have assumed that in Asaphus raniceps, as in virtually all recent arthropods, the
photoreceptive organs were of ommatidial kind, arranged radially, following the
radii of curvature more or less exactly and that optic ganglion was present cen-
trally. Though Stuermer’s material is tectonically distorted, it is clear that the
structures he described do not follow this pattern; they are numerous and quite
thick and because of their extreme length a radial pattern is not evident. As far
as can be seen they terminate outside the optic region altogether so that if they were
really light guides the optic ganglion would be displaced towards the central part
of the body.

Thus the eye as interpreted by Stuermer bears very little resemblance to any other
kind of arthropod eye. Even in certain mysids and euphausiids where the cone-stalks
are of exceptional length (thus accommodating multiple diffraction images at dif-
ferent depths), the optic ganglion remains centrally located, and the external part
of the eye becomes expanded outwards during ontogeny to accommodate the extra
length. It is difficult to see what purpose could be served in the large-lensed phaco-
pids by such an improbable distance between lens and photoreceptor, and it is
likely that the fibre-bundles are not part of the optical system at all. They could be
part of a circulatory system like the alimentary prosopon of many Cambrian trilo-
bites (Opik 1961) only located below the surface, or, as Dr. J. Bergstrom (pers. comm.)
has suggested, filaments of the ‘gill’ or exite branch of the appendages. Various
factors seem to support this suggestion, and in particular the random and oblique
orientation of the fibre-bundles, and their indistinct preservation directly below the
palpebral furrow, which would have acted as a ridge crushing against the filaments.
Such filaments may have been preserved only where trapped in the void below the
eye and the glabella, and that is why they extend only to the visual surface and not
beyond, thus giving the impression of being part of the optical system.

Until further evidence is forthcoming, therefore, Stuermer’s structures cannot be
unequivocally accepted as being light guides. And though the internal organization
of the schizochroal system may well have differed from that of holochroal eyes the
external morphology of the latter approximates that of insects and crustaceans in
so many ways that it seems more appropriate to infer some internal resemblance, at
least in so far as having radially arranged photoreceptors and a centrally placed
ganglion.

Acknowledgements. My thanks are due to Dr. John Dalingwater (University of Manchester) for the supply
of material which he had collected, for much correspondence on asaphid cuticles, and for critical reading
of the manuscript; to Dr. T. P. Scoffin for aid with diagenetic studies; to Drs. K. G. Cox and J. E. Dixon
for assistance with crystallographical interpretations; to Dr. R. B. Bathurst (University of Liverpool)
for valuable advice on diagenesis in an early stage of this work; to Mr. C. Eccles for discussion on many
aspects of this study; and to Mr. Jim Goodall (Department of Engineering, Edinburgh University) who
carried out all the Stereoscan preparation and photography.

CLARKSON: ASAPHUS EYES

443

REFERENCES

ANGELIN, N. p. 1878. Palaeontologica Scandinavica ; Part 1. Crustacea formationis transitionis, 1-96,
pis. 1-42 (issued 1878, ed. G. Lindstrom). Holmiae, Samson & Wallin.

BALASHOVA, E. A. 1948. On the tactile organs of trilobites. Dokl. Akad. Nauk. SSSR. 61, 509-51 1 . Moscow,
Leningrad. (In Russian.)

BECKMANN, H. 1951. Zur Ontogenie der Sehflache grossaiigiger Phacopiden. Paldont. Z. 24, 126-141,
pi. 10.

BOHLiN, B. 1949. The Asaphus Limestone in Northernmost Oland. Bull. geol. Instn. Univ. Upsala, 33,
529-570, pi. 1.

BURTT, E. T. and CATTON, w. T. 1962. A diffraction theory of insect vision. Part 1. An experimental study
of visual acuity in certain insects. Proc. Roy. Soc. B. 157, 53-82.

1966a. Image Formation and Sensory Transmission in the Compound Eye. In: Advances in

Insect Physiology, 6, 1-52.

\966b. The role of diffraction in compound eye vision. In Bernard, c. g. (ed.). Functional

Organisation of the Compound Eye. Wenner-Gren International Symposium Series, 7, 63-76. Pergamon
Press.

CLARKSON, E. N. K. 1966a. Schizochroal eyes and vision of some Silurian acastid trilobites. Palaeontology,
9, 1-29, pis. 1-3.

19666. Schizochroal eyes and vision in some phacopid trilobites. Ibid. 9, 464-487, pis. 73-75.

1967. Fine structure of the eye in two species of Phacops (Trilobita). Ibid. 10, 603-616, pi. 99.

1969. On the schizochroal eyes of three species of Reedops (Trilobita; Phacopidae) from the Lower

Devonian of Bohemia. Trans. R. Soc. Edinb. 68, 183-205, pis. 1-3.

1971. On the early schizochroal eyes of Ormathops (Trilobita, Zeliszkellinae). Mem. Bur. Rech.

Geol. Min. 73, 51-63, pi. 1, figs. 1, 2.

DALiNGWATER, J. (in press). Trilobite cuticle microstructure and composition. Palaeontology, 17.

DICKSON, j. A. D. 1966. Carbonate identification and genesis as revealed by staining. J. sedim. Petrol. 36,
491-505.

DUDiCH, E. 1931. Systematische und biologische Untersuchungen uber die Kalkeinlagerung des Crusta-
ceenpanzers in polarisiertem Lichte. Zoologica, 80, 1-154.

EGUCHi, E. and waterman, t. h. 1966. Fine structure patterns in crustacean rhabdoms . In Bernard, c. g.
(ed.). Functional Organisation of the Compound Eye. Wenner-Gren International Symposium Series,
7, 105-124. Pergamon Press.

EXNER, s. 1891. Die Physiologie der facettirten Augen von Krebsen und Insekten. Leipzig and Vienna.

FRIEDMAN, G. M. 1964. Early diagenesis and lithification in carbonate sediments. J. sedim. Petrol. 34, 377.

GOLDSMITH, T. H. 1964. The Visual System of Insects. In rockstein, m.. Physiology of Insecta, I. Ch. 10,
397-462. New York and London, Academic Press.

HARRINGTON, H. J. 1959. General Description of Trilobites. In moore, r. c. (ed.). Treatise on Invertebrate
Palaeontology, Part O, Arthropoda 1. O 38-0 117. Kansas, Lawrence.

HUPE, p. 1953. Classe des Trilobites. In piveteau, j. (ed.). Trade de Paleontologie, 44-246. Paris, Masson
et Cie.

lindstrom, g. 1901. Researches on the visual organs of the trilobites. K. Svensk. Vetensk. Akad. Handl.
34, 1-89, pis. 1-6.

MILLER, w. H., BERNARD, G. D. and ALLEN, J. L. 1968. The Optics of Insect Eyes. Science, 162, 760-767.

MULLER, J. 1826. Zur vergleichenden Physiologie des Gesichtsinnes, 1-462. Leipzig, Cnobloch.

OPiK, A. A. 1961. Alimentary caeca of agnostids and other trilobites. Palaeontology, 3, 410-438, pis. 68-70.

PORTILLO, J. DEL 1936. Die Beziehungen zwischen die Offnungswinkel der Ommatidien, Krummung, und
Gestalt der Insekten-Augen und ihr funktionellen Ausgabe. Z. vergl. Physiol. 23, 100-145.

RICHARDS, A. G. 1951. The Integument of Arthropods, 1-411. Univ. Minnesota Press.

ROSE, J. N. 1967. The Eyes of Isotelus and Nileus. Proc. Iowa Acad. Sci. 74, 178-185, pi. 1.

SCHMIDT, F. 1904. Revision der Ostbaltische Silurischen Trilobiten, Part 5. Asaphiden, sec. 3. Mem. Acad,
imp. Sci. St. Petersbourg, ser. 7, 30, 1-68, pis. 1-8.

SHAW, F. c. and ormiston, a. r. 1964. The eye socle of Trilobites. J. Paleont. 38, 1001-1002.

STUERMER, w. 1970. Soft Parts of Cephalopods and Trilobites; Some Surprising Results of X-Ray Exam-
ination of Devonian Slates. Science, 170, 1300-1302.

444

PALAEONTOLOGY, VOLUME 16

TJERNViK, T. 1972. Bolkhov stage — Limbata Limestone. Geol. Foren. Stock. Fork. 94, 301-310.
WATERMAN, T. H. 1961. Light Sensitivity and vision. In waterman, t. h. et al.. The Physiology of Crustacea,
vol. 2, 1-54. Academic Press.

WHITTINGTON, H. B. 1963. Middle Ordovician Trilobites from Lower Head, Western Newfoundland. Bull.
Mus. Comp. Zool. 129, 1-118, pis. 1-36.

1965. Trilobites of the Ordovician Table Head Formation, Western Newfoundland. Bull. Mus.

Comp. Zool. 132, 275-442, pis. 1-68.

wiGGLESWORTH, v. B. 1965. The Principles of Insect Physiology. 6th edition. London, Methuen.

E. N. K. CLARKSON
Grant Institute of Geology
University of Edinburgh

Revised typescript received 11 October 1972 Edinburgh EH9 3JW

REMOPLEURIDES AND OTHER
UPPER ORDOVICIAN TRILOBITES FROM
NEW SOUTH WALES

by B. D. WEBBY

Abstract. Thirteen trilobite species are described and illustrated from Upper Ordovician successions of central
New South Wales. They include the new species, Remopleurides saenuros, R. exallos, R. acer, Pseudobasilicusl
fortis, and Illaenus (Parillaenus)! incertus, and records of Shumardia, Geragnostusl, and a harpid. A discussion of
possible dimorphism in Remopleurides is presented.

In previous contributions on the Upper Ordovician trilobite fauna of central New
South Wales, several new genera and species were proposed. Webby, Moors, and
McLean (1970) introduced the new genera Malongullia and Encrinuraspis , and
Campbell and Durham (1970), Parkesolithus. In the following year Webby (1971)
added two new species of Pliomerina. The present work completes descriptions of
the Upper Ordovician agnostids, shumardiids, remopleuridids, asaphids, illaenids,
and harpids from central New South Wales based on collections housed in the
Department of Geology and Geophysics, University of Sydney.

The faunas have been collected through a considerable stratigraphic thickness of
beds, and from widely scattered localities on the Molong Rise and Parkes Platform
of the Lachlan Geosyncline (Packham 1969). Remopleurides acer sp. nov. and Pseudo-
basilicusl fortis sp. nov. occur with Pliomerina prima Webby and a scutelluid (not
Eobronteus) in the lower part of the Cliefden Caves Limestone. R. saenuros sp. nov.,
R. sp. A, Pseudobasilicusl sp. A, and a harpid are represented, together with Plio-
merina austrina Webby, Amphilichas, Sphaerocoryphe, and an encrinurid, in the
Ordovician limestone at Billabong Creek, considered to be approximately equivalent
in age to the upper part of the Cliefden Caves Limestone. Another, stratigraphically
younger fauna, with Illaenus (Parillaenus)! incertus sp. nov. as the dominant form,
is found in calcarenites at the top of the Ballingoole Formation (upper part of the
Bowan Park Group of Semeniuk 1970). Other elements of the fauna are too frag-
mentary to determine to species level, but include Remopleurides, an asaphid, a
proetid, a trinucleid, and encrinurid. The three stratigraphically distinct trilobite
faunas from the limestones correlate quite well with the three stromatoporoid/coral
faunas I, II, and III outlined previously (Webby 1969).

In the shales of the Malongulli Formation, directly overlying the Cliefden Caves
Limestone, at Trilobite Flill and Copper Mine Creek, R. exallos sp. nov., Malongullia
oepiki Webby, Moors, and McLean, Encrinuraspis optimus Webby, Moors, and
McLean, Parkesolithus, and a scutelluid (not Eobronteus) are found. These beds,
from their associated graptolites (Moors 1970), are taken to have an Upper Eastonian
age {Dicranograptus hians Zone). A similar fauna has been collected from near
Mirrabooka homestead, north of Cheeseman’s Creek, with additions of R. sp. B,
Illaenus (Parillaenus)! sp. A, and Toernquistia. Another distinct species of Remo-
pleurides, R. sp. C, occurs in a near-by locality, just north of the Cheeseman’s Creek

[Palaeontology, Vol. 16, Part 3, 1973, pp. 445-475, pis. 51-55.]

446

PALAEONTOLOGY, VOLUME 16

Post Office, and possibly comes from a higher stratigraphic horizon. The scutelluids
from the lower part of the Cliefden Caves Limestone and the Malongulli Formation,
previously assigned to Eobronteus (Webby 1971, p. 612; Whittington and Hughes
1972, p. 273), more correctly belong to a new genus having closest affinities to the
Silurian genera Kosovopeltis Snadjr and Plants cute Hum R. and E. Richter.

The beds of the Oakdale Formation in the Mumbil area, with their Geragnostusl
and Shumardia, have previously been assigned an Upper Eastonian age on the basis
of occurrences of associated species of Climacograptus, Dicellograptus, and Ortho-
graptus (Strusz 1960, 1961). They could be comparable in age to the beds of the
Malongulli Formation, but are probably slightly older, perhaps Gisbornian or
Lower Eastonian (above the base of the Nemagraptus gracilis Zone). It seems that
the Oakdale Formation includes two separate successions, an older belt to the west
containing the ‘upper Eastonian’ graptolites, and Geragnostusl and Shumardia, and
a much younger sequence in the core of the Oakdale anticline to the east, with a
rich coral fauna at the top, suggesting either an Upper Bolindian or, more probably,
a late Llandoverian age (see further discussion in Webby 1972).

Pittman (1900, p. 10) has recorded a probable agnostid from Ordovician grapto-
litic shales at Junction Reefs, near Mandurama. Alluding to this same trilobite.
Chapman (1914, p. 227) related it to Shumardia. Stevens (1957, p. 48) considered
the beds containing the trilobite to be associated with occurrences of Glyptograptus
teretiusculus, and Smith (1966) grouped these beds in the Malongulli Formation.
The region is east of the type area of the Malongulli Formation (Upper Eastonian
age) near Cliefden Caves, and the beds contain definite Darriwilian (Llanvirn-
Llandeilo) graptolite assemblages (Packham 1969, p. 80). According to Packham,
the upper limit of age of these beds is probably near the base of the Eastonian.
Smith’s ‘Malongulli Formation’ is therefore a distinct, older, Darriwilian-Gisbornian
unit. Pittman’s original trilobite is confirmed as a species of Shumardia closely similar
to the Oakdale form. It is possible that it is conspecific with the Oakdale species.
Perhaps both come from similar, Gisbornian, horizons.

In the silicified residues from the Ordovician limestone at Billabong Creek (Pack-
ham 1967), there is abundant material belonging to the small species of Remo-
pleurides, R. saenuros, allowing virtually complete description, and fragmentary
specimens of the thorax and pygidium of a much larger and broader form, Remo-
pleurides sp. A (PI. 52). The cephalon and anterior thoracic segments of this latter
form remain unknown, but it can be estimated to have been about four times larger
than R. saenuros. It is relatively much broader, especially across the axis, and appears
to lack a median axial spine on the third thoracic segment in front of the pygidium.
It has a large, rather undifferentiated, sub-trapezoidal, posteriorly rounded pygidium,
totally different from the pygidium of R. saenuros, which has a weakly differentiated
triangular axis, two pairs of short pleural spines, and a deeply incised median longi-
tudinal furrow. Another very large, incomplete and rather similar specimen, referred
to Remopleurides sp. B, occurs in the shales of the Malongulli Formation associated
with R. exallos (PI. 51, figs. 3-13). These associations suggest the possibility of the
New South Wales Remopleurides exhibiting a most distinctive type of dimorphism,
though the cephalon of the large form must be found to demonstrate the relationship
convincingly. Whittington (1963) has mentioned the possibility of dimorphism be-

WEBBY: ORDOVICIAN TRILOBITES

447

tween his two species of Remopleurides, R. eximius and R. simulus, from the Lower
Edinburg Formation of eastern North America, but these are separated by com-
paratively minor morphological characteristics compared with the features dis-
tinguishing the large and small New South Wales forms. R. eximius has a tiny spine
at the genal angle, long, curving pleural spines on the seventh thoracic segment, and
deep furrows on the pygidial axis. R. simulus, in contrast, has a slender spine originat-
ing in front of the genal angle, the pointed ends of the pleurae are equally extended,
the axis of the pygidium is less inflated, the furrows shallower, and it is only about
two-thirds the size of R. eximius.

A micrometer ocular and, for larger dimensions, a pair of dividers have been used for taking measure-
ments. Unfortunately, however, there are considerable errors in taking linear measurements across curved
surfaces of exoskeletons, especially in more highly convex forms. Therefore, only measurements taken
across less convex surfaces are cited.

SYSTEMATIC DESCRIPTIONS

Family agnostidae McCoy 1849

Genus geragnostus Howell 1935

Type species. Agnostus sidenbladhi Linnarsson 1869

Geragnostusl sp.

Plate 51, fig. 1

Material. One internal mould of a fairly complete specimen (SUP 26900) from
locality G (Strusz 1961, p. 334, text-fig. 1) of the Oakdale Formation, miles NNE.
of Newrea, south of Wellington.

Description. Although margins of cephalon not exposed, general appearance suggests
subquadrate outline ; slightly wider than long. Glabella expanded anteriorly, bounded
by deep, broad axial and preglabellar furrows, and divided into anterior and posterior
lobes by pronounced transglabellar furrow situated about glabellar mid-length,
slightly convex forwards. From nature of freshly broken surface immediately behind
transglabellar furrow, it seems likely that small median glabellar node may have
been present originally. Anterior lobe gently convex, widest just behind mid-length
(sag.), wider than long, with convexo-concave outline. Posterior lobe narrower (tr.),
moderately strongly convex, parallel-sided with posterior part sloping steeply back-
ward and having sharply V-shaped posterior outline. V-shaped occipital furrow
separates posterior lobe from pair of triangular, lateral occipital lobes (Whittington
1963, p. 28).

Cheeks and preglabellar area convex, with narrow, steeply sloping inward edge
to axial and preglabellar furrows, and wider more gently sloping outward margin.
Border unknown.

Thorax of two segments, subequal in size. Axis of anterior segment broad, divided
into median trapezoidal lobe and pair of ovate lateral lobes elongated inward and
forward at about 45° ; each pleura divided almost equally by pleural furrow. Posterior
segment similar, differing in having more obviously quadrate median lobe, suboval
lateral lobes elongated transversely, and each pleura curving forward with pro-
minent pleural furrow situated on anterior part.

448

PALAEONTOLOGY, VOLUME 16

Pygidium slightly wider than long; axis convex, rounded posteriorly, and widest
about mid-length, limited by deep, broad axial furrows. Transverse furrow well
defined, gently concave anteriorly and situated about one-third of total length of
axis from anterior margin. Anterolateral corners of axis exhibit pair of elongated,
triangular-shaped lobes. Weakly developed rounded median node forms highest
point on pygidium, just in front of transverse furrow. Posterior lobe of axis has
concavo-convex outline. Inner edge of pleural lobes inclined steeply in towards axial
furrow; outwardly more or less of equal width and convexity, bounded by border
furrow. Border poorly preserved; no spines seen. Ornamentation not observed.

Remarks. The New South Wales species seems to be nearest to Geragnostus mccoyii
(Salter) from the Llandeilo and basal Caradoc of Wales and the Welsh Borderland
(Whittard 1955 ; Hughes 1969), a form with a deeply impressed transglabellar furrow.
However, it has a much more expanded anterior glabellar lobe, more sharply V-
shaped posterior margin to the posterior lobe of the glabella, and more or less equal
thoracic segments. Since the borders are incompletely preserved and the anterior
glabellar lobe is rather enlarged, the species is only tentatively assigned to the genus.

Species of Geragnostus, as Whittington (1963, 1966) has pointed out, range from
Upper Cambrian to Middle Ordovician, with a doubtful record in the Ashgill of
Poland (Kielan 1960), and have a wide geographical distribution. There seems to
be no known occurrence of a species of Geragnostus with strongly developed trans-
glabellar furrow in beds younger than basal Caradoc.

EXPLANATION OF PLATE 51

Fig. 1 . Geragnostusl sp. Dorsal view of internal mould of incomplete exoskeleton from Oakdale Forma-
tion near Newrea ; SUP 26900, x 6.

Fig. 2. Shumardia sp. Latex impression of cephalon from Oakdale Formation, near Newrea; SUP 26901,

x8.

Figs. 3-12. Remopleurides exallos sp. nov., from the Malongulli Formation. 3-4, 6, 9-10, and 12 from
Copper Mine Creek, 1 1 from Trilobite Hill, and 5, 7-8 from near Mirrabooka homestead, Cheeseman’s
Creek area. 3, Dorsal view of latex impression of external mould of cranidium; holotype SUP 28913,
x4. 4, Dorsal view of internal mould of cranidium; paratype SUP 19937, x5. 5, Oblique view of
latex cast of external mould of cranidium; paratype SUP 27905b, x6. 6, Partly disarticulated speci-
men showing internal mould of dorsal side of part of cranidium and two thoracic segments, and ventral
side of last three thoracic segments and pygidium; paratype SUP 28911a, x6. 7-8, Oblique views
of internal and external moulds of left free cheek; paratype SUP 28905, x6. 9, Ventral view of com-
pressed specimen showing internal mould of anterior part of doublure and large pit, and external mould
of free cheek with distinctive anastomosing lines; paratype SUP 28914, x6. 10, Dorsal view of latex
impression of external mould of two thoracic segments; paratype SUP 26942b, x4. 11, Ventral side
of thoracic pleurae; paratype SUP 19917, x8. 12, Dorsal view of latex cast of external mould of

articulated posterior part of exoskeleton with four thoracic segments and pygidium. Note median axial
spine on third thoracic segment in front of pygidium; paratype SUP 28910, x6.

Fig. 13. Remopleurides sp. B from the Malongulli Formation near Mirrabooka homestead, Cheeseman’s
Creek area; SUP 28904, x2. Internal mould of six thoracic segments and damaged pygidium.

Figs. 14-15. Remopleurides sp. C. Ventral and dorsal views of latex impressions of left free cheek from
shales just north of Cheeseman’s Creek P.O.; SUP 28908, x4.

PLATE 51

WEBBY, Geragnostusl, Shumardia, Remopleurides

450

PALAEONTOLOGY, VOLUME 16

Family shumardiidae Lake 1907
Genus shumardia Billings 1862

Type species. S. granulosa Billings 1 862

Shumardia sp.

Plate 51, fig. 2

Material. An incomplete external and internal mould of a cephalon (SUP 26901)
and a fragment of the thorax (SUP 26902) from locality G (Strusz 1961, p. 334,
text-fig. 1) of the Oakdale Formation, 1^ miles NNE. ofNewrea, south of Wellington.

Description. Cephalon convex, approximately subsemicircular; glabella occupying
almost one-half cephalic width. Anterior part of glabella expanded with pair of
elongate, ovoid antero-lateral lobes, isolated by lateral glabellar furrow 2p running
forward and slightly inward from deeply indented axial furrow; posterior part of
glabella semicylindrical ; lateral glabellar furrow Ip not seen, and apparently no
basal lobes. Antero-lateral lobes narrower than intervening frontomedian lobe of
glabella. Axial furrow deeply impressed in posterior part of glabella but shallows
as it curves forward around antero-lateral lobe, being confluent with preglabellar
furrow, and coming together with a V-shaped outline, viewed from the front, with
an angle of about 90° between either side. Occipital furrow separates convex occipital
ring from rest of glabella; slightly wider than posterior part of glabella.

Cheek convex, sloping steeply distally, especially postero-laterally, narrowing into
preglabellar area anteriorly; also sloping steeply into posterior part of axial furrow.
Facial suture not observed. Posterior border furrow not shown. Surface of cephalon
appears to be unornamented.

Remarks. A specimen (MMF. 3414) in collections of the Geological and Mining
Museum, Geological Survey of N.S.W., from Ordovician graptolitic shales near
Junction Reefs, Mandurama (Pittman 1900), proves to exhibit a species of Shumardia
which may be conspecific with the Oakdale form. The slab shows two separate cephala
and a pygidium. The form and size of the cephala are similar to that of the Oakdale
species. The relatively slightly shorter, less slender glabella is probably not a signifi-
cantly diagnostic feature for differentiation of a separate species. These specimens
exhibit a little more of the posterior part of the cheek than in the Oakdale cephalon.
The posterior border is very deep, and undercuts the cheek on its inner course, but
shallows and curves slightly forward distally, dying out inside the lateral margin,
and not apparently continuous into a lateral border furrow. The posterior border
has a prominent wedge-shaped, distally widening outline. The pygidium is sub-
triangular, with raised, almost smooth axis, tapering backward to bluntly rounded
point. Anteriorly, the articulating furrow and articulating half-ring are well de-
veloped. Seven very weakly differentiated axial rings are seen. The pleural regions
exhibit a prominent first pair of pleural furrows, curving backward and outward to
the posterior edge of the articulating facet, and the next three, progressively shorter
and much weaker. Posteriorly the pleural regions are smooth. Broad (sag.), shallowly
curved posterior border furrow separates an elongated, flattened, tongue-like
posterior border from the steeply declined, convex pleural region behind the axis.

WEBBY: ORDOVICIAN TRILOBITES

451

An internal mould of the pygidium shows the inner edge of the doublure extending
from immediately behind axis in smooth, gentle curve antero-laterally toward pos-
terior edge of articulating facet, with the result that the doublure has a much ex-
panded, crescent-shaped outline. Pygidia of Koroleva’s (1964) species of Shumardia,
S. lacrima and S. analoga, from the Middle Ordovician of northern Kazakhstan,
seem to show a similar tongue-like projection of the posterior border, but both
exhibit more conspicuous differentiation of axial rings and pleural ribs.

The Oakdale species appears to bear close resemblances to S. minutula Harring-
ton from the Upper Tremadoc and Arenig of Argentina (Harrington and Leanza
1957). However, the glabella differs in being better differentiated with longer and
more slender antero-lateral lobes. S. sagittula Whittington (1965) from the Middle
Table Head Formation of Newfoundland may also be compared, but has a longer
(sag. and exsag.) occipital ring, a shorter subcylindrical part of glabella in front of
occipital furrow and more bulbous antero-lateral lobes. The present species bears
little resemblance to known Upper Ordovician species, S. scotica Reed (1903) from
the Whitehouse Group (late Caradoc or early Ashgill) of Girvan, Scotland, and
S. polonica Kielan (1960) from the Middle Ashgill of Poland. The posterior part of
the glabella of both these species is much broader, and they exhibit weakly differ-
entiated basal lobes and a relatively wider occipital ring. Of Koroleva’s (1964)
species of Shumardia from the Middle Ordovician of northern Kazakhstan, S. analoga
has most similar cephalic proportions, but differs in exhibiting slightly shorter
(exsag.) antero-lateral lobes, and sub-quadrate outline of both anterior parts of the
cephalon and the glabella. The outline of both the cephalon and glabella of the Oak-
dale species is more conspicuously V-shaped anteriorly.

Family remopleurididae Hawle and Corda 1847
Genus remopleurides Portlock 1843

Type species. R. colbii Portlock 1843; subsequently designated by Miller 1889.

Remopleurides saenuros sp. nov.

Plate 52, figs. 1-28

Material. Holotype (SUP 27932) and eleven paratypes (SUP 27933-27936, 27938-
27941, 27948-27949, 28900) from Ordovician limestone at Billabong Creek. All
fragmentary silicified specimens.

Description. Glabella convex, especially anteriorly along sagittal line, less convex
transversely; glabellar tongue broad, elevated, vertical to very slightly overhanging
anterior cephalic margin. Maximum width of glabella about three-fifths total dis-
tance behind anterior margin (sag.). Length of glabella (including occipital ring)
slightly greater than maximum width. Width at base of glabellar tongue slightly
more than width at base of median glabellar area, adjacent to occipital furrow;
glabellar tongue exhibits slight widening towards antero-lateral extremities. No
lateral glabellar furrows visible. Occipital ring gently convex, set below raised median
glabellar area, elongate (sag. and exsag.) but tapering rapidly towards distal ends;
occipital furrow transverse, with steep slope in front on to median glabellar area.

452

PALAEONTOLOGY, VOLUME 16

Glabellar tongue bounded laterally by axial furrow, with small anterior pit placed
in furrow at antero-lateral extremity; preglabellar furrow continuous with axial
furrow, becoming shallower and closer to anterior margin until it dies out a short
distance in from extremities. As seen in frontal view, anterior margin flattened to
very weakly concave ventrally (PI. 52, fig. 3). Axial furrow continuous with palpe-
bral furrow to rear. Palpebral rim flat, tilted slightly forward and outward (eye lobe
has similar tilt), widens posteriorly to maximum width adjacent to postero-lateral
part of median glabellar area; narrows rapidly as it descends to occipital furrow.
Posterior area of fixed cheek narrow (exsag.), relatively short (tr.), though not for
genus, near horizontal, tapering towards backwardly deflected tip; articulation with
first thoracic segment facilitated by small axial process on axial furrow, gently trans-
versely arched posterior flange and large fulcral sockets at distal end of posterior
area (PI. 52, figs. 1, 6-7). Doublure extends forwards slightly more than one-half
sagittal length of occipital ring. Surface ornamentation on glabella and occipital
ring consists of faint Bertillon pattern of lines; also row of small granules developed
along posterior margin of occipital ring (PI. 52, fig. 7).

Eye lobe long (exsag.) with visual surface curved through about 150°; anterior
part of lobe lies close to antero-lateral cephalic extremity, and posterior part virtu-
ally overhangs distal edge of occipital ring. Anterior branch of facial suture des-
cends subparallel with axial furrow across anterior end of eye lobe, then turns
sharply inwards and forwards to meet at mid-line ; posterior branch of suture, after
curving gently backward and inward between visual surface and palpebral rim, turns
sharply downwards and outwards to pass along anterior margin of posterior area
of fixed cheek, and then deflected sharply backwards to cut posterior margin between
rounded projection on free cheek and fulcral socket on posterior area of fixed cheek.
Rounded projection continuous into narrow, curved ridge of posterior border on
free cheek; posterior border furrow similarly curved and intersecting posterior
margin between rounded projection and short, slender genal spine (PI. 52, figs. 9,
11); genal spine projects backward and slightly outward from about opposite occi-
pital ring. Cheek outside eye lobe narrow, outward sloping, and subtriangular in

EXPLANATION OF PLATE 52

Figs. 1-28. Remopleurides saenuros sp. nov., from Ordovician limestone at Billabong Creek. 1-4, Dorsal,
lateral, anterior, and ventral views of cranidium; holotype SUP 27932; 1-3, x 5; 4, x4. 5-8, Lateral,
dorsal, posterior, and ventral views of cranidium; paratype SUP 27933; 5-7, x 5; 8, x4. 9-12, Dorsal,
ventral, dorso-lateral, and oblique interior (enlarged) views of left free cheek; paratype SUP 27940;
9-11, x5; 12, x8. 13-14, Lateral and ventral views of right free cheek; paratype SUP 27935, x5.
15-16, Dorsal and lateral views of right free cheek; paratype SUP 28900, X 5. 17-20, Dorsal, anterior,
ventral, and posterior views of thoracic segment; paratype SUP 27949, X 5. 21-22, Lateral and dorsal
views of two thoracic segments; paratype SUP 27941, x5. Note median axial spine. 23-25, Ventral,
lateral, and dorsal views of ten thoracic segments and pygidium; paratype SUP 27936, x5. 26-27,
Magnified views of dorsal and ventral surfaces of pygidium in articulated specimen of figs. 23-25;
paratype SUP 27936, x8. 28, Oblique view of five thoracic segments and pygidium; paratype SUP
27948, X 5. Median axial spine damaged.

Figs. 29-32. Remopleurides sp. A from Ordovician limestone at Billabong Creek, x4. 29-31, Dorsal,
lateral, and ventral views of pygidium and last three thoracic segments; SUP 27946. 32, Dorsal view
of pygidium and two thoracic segments of SUP 27945.

PLATE 52

WEBBY, Remopleurides

454

PALAEONTOLOGY, VOLUME 16

outline, with terrace lines running obliquely out to margins, separated from visual sur-
face by convex external rim. Visual surface not well enough preserved to show facets.

Doublure of free cheek widest anteriorly towards mid-line; cut by median suture.
Beneath eye lobe innermost part of doublure exhibits tendency to curl upward.
Doublure narrows to just behind widest part of glabella, then widens posteriorly
on to genal spine and into short, rounded projection of posterior margin. Crescent-
shaped, convex forward, facet in doublure lies in front of, and between base of genal
spine and rounded projection, allowing forward movement of first thoracic pleura
during enrolment (PI. 52, figs. 10, 14). Large pit on external surface of doublure in
vertical line with anterior pit on antero-lateral corner of glabellar tongue; no cone-
like structure seen to be developed. On inner part of doublure adjacent to median
suture there is sharp downward deflection of inner edge of doublure forming one
side of funnel-shaped opening on median suture (PI. 52, figs. 9-10, 12). A similar
feature is shown in R. caphyroides and R. ligulus (see Whittington 1959, pi. 7, figs.
4-7; 1963, pi. 4, fig. 3). Terrace lines prominent around doublure, running sub-
parallel to lateral margins; also run out along genal spine.

Broad, convex axis of thorax tapers back to about one-half its width in length of
ten thoracic segments; short, paddle-shaped pleurae inclined backward and out-
ward beyond prominent fulcrum. Axial rings raised, with gentle slope forward (sag.),
steepening on to anterior margin, and even steeper slopes around lateral and pos-
terior margins. Relatively deep articulating furrow and moderately long (sag.)
articulating half-ring. Median axial spine placed on third segment in front of pygi-
dium; almost straight, tapering nearly to point in backward and upward direction,
inclined at about 30° to crestal profile of adjacent axial rings. Spine 4 0 mm long,
supported by axial ring 2-6 mm wide (PI. 52, figs. 21-22). Articulating structures
along short, transverse and horizontal anterior and posterior margins of inner part
of pleurae. Anterior flange extends along anterior margin between enlarged rounded
fulcral process antero-laterally, and small axial socket lying on axial furrow; on
posterior side, posterior flange seems to be raised slightly between enlarged fulcral
socket and small rounded axial process, protruding as ridge on to dorsal surface
of inner part of pleura (PI. 52, fig. 20). Vague suggestion of ring processes and sockets
inside axial furrow, but not clearly differentiated. Outer parts of pleurae backwardly
turned, with outer ends rather bluntly rounded, and more sharply pointed, pos-
teriorly directed tips. Weak, diagonal pleural furrow runs across inner part of pleura,
becoming ill defined near posterior margin on outer part of pleura.

Inner edge of doublure with gentle sigmoidal course from fulcral process back-
ward to point just outside fulcral socket, giving over-all zigzag appearance to dou-
blure along inner margin of thorax (PI. 52, fig. 23). Posterior part of doublure on outer
part of pleura flattened to form articulating facet for under-riding during enrolment
of succeeding pleura; sharp curved (convex forward) break-in-slope between facet
and gently convex anterior part of doublure. Tongue-like extension of doublure
beneath axial ring, reaching forward about two-thirds total sagittal length. Orna-
mentation of thorax virtually restricted to faint lines in Bertillon pattern on dorsal
surface of axis and pleurae; also a few granules scattered on pleurae, and along
posterior margin of axial rings. Terrace lines run subparallel to lateral margins on
doublure.

WEBBY: ORDOVICIAN TRILOBITES

455

Pygidium subrectangular, wider than long, with lateral margins diverging slightly
backwards; axis relatively long, subtriangular, tapering backward seemingly almost
to deep U-shaped median notch on posterior border (PI. 52, fig. 26). Articulating
furrow sharply impressed, becoming deeper distally; articulating half-ring relatively
long (sag.). First axial ring expands distally, and second axial ring divided by deep
median longitudinal furrow. Owing to fusion of furrows it is rather difficult to inter-
pret the precise nature of segmentation in posterior part of axis and relationships
with pleural regions. Suggestion of pair of moderately large, oval muscle attachment
areas lying over ring furrow between first and second axial rings, just inside weakly
developed axial furrow. Two pairs of pleural ribs, with short, weak interpleural
furrow between them; ribs extend into short, pointed pleural spines; the first pair
shorter and diverging slightly outwards, the second pair tending to converge back-
wards. Prominent fulcral process on antero-lateral corner of pygidium. Doublure
extends beneath pleural regions, and apparently beneath posterior part of axis just
in front of deep median notch (PI. 52, fig. 27). Extension of pygidial doublure in
front of deep median notch perhaps implies axis is rather shorter than appears to
be the case on the dorsal surface. Terrace lines continue subparallel to margin on
pygidial doublure. In two specimens (PI. 52, figs. 26, 28) pygidium has sagittal length
of 1-6 and 1-5 mm, and maximum width of 2T and 2-2 mm, respectively.

Remarks. R. saenuros has a median axial spine placed on the third segment in front
of the pygidium, whereas in most North American and European species of Remo-
pleurides it occurs on the fourth segment in front of the pygidium, i.e. on the eighth
thoracic segment where eleven thoracic segments are typical for the genus (Whitting-
ton 1950fl, 1959; Shaw 1968; Ingham 1970). But in the most complete specimen of
R. saenuros only ten thoracic segments occur (PI. 52, figs. 23-27), and though it
remains doubtful, there is nevertheless the possibility that this represents the full
complement of thoracic segments for the species.

Even more distinctive is the pygidium of R. saenuros. Instead of the characteristic
short and rapidly tapering two-segmented axis, and much broader and longer,
flattened pleural regions, the axis of R. saenuros is relatively larger, triangular, and
appears to extend back almost to the median notch, being virtually bisected by a
deep and most prominent median longitudinal furrow, while the pleural regions
are relatively narrow, having two pairs of short, pointed pleural spines, the outer
pair divergent and the inner pair converging backward to either side of the median
notch. Other small New South Wales species (R. exallos and R. acer) have a similar
type of pygidium. None of the European and North American species seems to have
a closely comparable pygidial construction. The pleural regions of the pygidium of
R. validus Thorslund (1940) from the Lower Chasmops Limestone of Jemtland,
Sweden, are similar, but this is the only resemblance. The pygidium as a whole is
much broader, and the posterior part of the axis much shorter, with median longi-
tudinal furrow shallower and considerably less extended.

456

PALAEONTOLOGY, VOLUME 16

Remopleurides exallos sp. nov.

Plate 51, figs. 3-12

Material. Holotype (SUP 28913) and seven paratypes (SUP 19937, 20902, 26942b,
28910, 28911a, 28912, 28914) from Copper Mine Creek, one paratype (SUP 19917)
from Trilobite Hill, and three paratypes (SUP 27905b, 28905, 28906) from near
Mirrabooka homestead, north of Cheeseman’s Creek P.O. All specimens come from
shales in the Malongulli Formation.

Comparative description. This species differs from R. saenuros chiefly in being some-
what larger, having a relatively longer (sag.) glabella and larger, more conspicuous
spines. It has a much more marked, fine Bertillon pattern of terrace lines on the
dorsal surface of the exoskeleton, excepting the furrows, and abundant granules on
the median glabellar area adjacent to the occipital furrow and next to the posterior
part of the palpebral furrows. Three pairs of lateral glabellar furrows are visible in
some external and internal moulds (PI. 51, figs. 3-4), equally spaced exsagittally,
with the first curving and tapering outwards, the second, long, slender, and gently
curved, and the third, short and transversely elongate.

Anastomosing terrace lines developed on the free cheeks and extend outward in
a transverse direction from the region of the eye lobe, becoming deflected backwards
near the lateral border. The posterior border has terrace lines arranged transversely
along it. Granulation is also developed in the angle between the posterior part of
the external rim of the eye lobe and the inner end of the posterior border furrow
(PI. 51, fig. 9). The external rim of the eye lobe has parallel lines along it, and scat-
tered granules on its posterior part (PI. 51, fig. 8). Visual surface is covered with
many tiny, low facets arranged in diagonal rows. Well-developed genal spine with
its base opposite occipital ring, and having prominent terrace lines running along
its length. The posterior margin between the genal spine and the rounded posterior
projection is straight and transverse, rather than curved as in R. saenuros.

Fine granulation is also developed on the posterior edge of axial rings of the
thorax, particularly the anterior ones, and they occur to either side of the conspicuous
diagonal furrow on the inner part of the thoracic pleurae (PI. 51, fig. 11). Terrace
lines run diagonally outward, parallel with the diagonal furrow on the outer part
of the pleurae, towards the pointed spine-like pleural tips (more pointed than in
R. saenuros but less drawn out than in R. acer). Terrace lines have forwardly convex
course on axial rings, except for third axial ring in front of pygidium, where they are
deflected backwards along straight, tapering median axial spine (PI. 51, figs. 10, 12).

External moulds of pygidium show differentiation of raised, subtriangular axial
region divided by long, deep, median longitudinal furrow almost intersecting first
axial segment and continuous to just in front of median notch, and relatively narrow
pleural region with two pairs of moderately elongated pleural spines (PI. 51, fig. 12).
First axial segment very narrow medially, widening rapidly distally. The ring furrow
curves in arc outwards and backwards. Terrace lines represented on outer parts of
segment, aligned diagonally in harmony with those on lateral parts of thoracic
axial rings (PI. 51, figs. 6, 12). Behind ring furrow are extensive raised triangular
areas of the second axial segment to either side of longitudinal median furrow. Small

WEBBY: ORDOVICIAN TRILOBITES

457

granules cover surface especially adjacent to the longitudinal median furrow.
Laterally, close to angle between ring furrow and axial furrow, there is a smooth
oval ? muscle area. Terrace lines run along length of pleural spines. A few granules
also appear on the inner pleural area adjacent to the anterior margin. A narrow,
arched, slightly elevated pleural area lies between posterior median notch and
posterior tip of longitudinal median furrow, and exhibits terrace lines subparallel
to the curve of the notch.

Remopleurides acer sp. nov.

Plate 53, figs. 1-10

Material. Holotype (SUP 28922) and fourteen paratypes (SUP 28915-28921, 28923-
28926, 28928-28930) from the lower part of the Cliefden Caves Limestone, Fossil
Hill.

Comparative description. R. acer closely resembles R. saenuros but may be dis-
tinguished mainly by having a more prominent, coarser Bertillon pattern of lines
on the glabella, typically interrupted and deflected by the three pairs of lateral
glabellar furrows (PI. 53, figs. 1-3), by exhibiting more elongated, spine-like thoracic
pleurae, even anteriorly (PI. 53, fig. 8), and by having a pygidium with longer pleural
spines and more conspicuous muscle scars on the axial region (PI. 53, figs. 9-10).
Again it is difficult to interpret precise nature of the segmentation in the pygidial
axis because of fusion of the axial segments and imprint of deep, median longi-
tudinal furrow. The first axial ring narrows medially, and ring furrow has a concave
backward outline. The second axial ring is rather ill defined but bisected by deep,
median longitudinal furrow, and appears to have large pair of oval- to crescent-
shaped muscle scars antero-laterally, abutting ring furrow. Possibly another, slightly
smaller, pair of scars is situated directly behind the first pair (PI. 53, fig. 10). Terrace
lines and granulation tend to be more prominent on the dorsal exoskeleton of
R. acer than in equivalent parts of R. saenuros.

R. acer differs from R. exallos in being somewhat smaller, having a relatively less
elongate (sag.) glabella, ornamented by a coarser Bertillon pattern of lines, and more
elongated pleural spines on the thorax and pygidium. There is a less anastomosing,
more obliquely, backwardly directed pattern of lines on the free cheek between the
external rim of the eye lobe and the lateral border, and granulation is apparently
lacking. The visual surface of the eye lobe also exhibits tiny eye facets in diagonal
rows.

Remopleurides sp. A
Plate 52, figs. 29-32

Material. Three fragmentary silicified specimens (SUP 27945-27947) from Ordovi-
cian limestone at Billabong Creek.

Description. Only three thoracic segments in front of pygidium seen; no evidence
of median axial spine on third thoracic segment in front of pygidium (PI. 52, fig. 29).
Very broad, smooth, gently convex axis, with deep, narrow articulating furrow and
long (sag.) articulating half-ring. Pleurae divided by pronounced constriction at
c

458

PALAEONTOLOGY, VOLUME 16

fulcrum into narrow (tr.), transverse, gently convex inner part, strictly in continuity
with axis because axial furrow not clearly imprinted on dorsal surface, and an outer
part showing backward and outwardly inclined blunt, short, paddle-shaped form.
Enlarged fulcral processes and sockets developed on pleurae, but no apparent
diagonal furrow. Judging from rapid increase in width of thorax in successive seg-
ments forwards from pygidium, maximum width of thorax (and cephalon) is likely
to be more than twice width of pygidium, perhaps considerably more. Doublure
very wide, extending inwards to fulcrum; also extends forward beneath axial ring
(PI. 52, fig. 31).

Pygidium large, subtrapezoidal in outline, approximately two to two and one-half
times wider than long. Deep transverse articulating furrow and long articulating half-
ring at anterior margin. At antero-lateral extremities prominent fulcral processes,
and laterally, facets to receive last pair of pleurae in enrolment. Axial and pleural
areas not differentiated, but very broad, gently undulose surface of pygidium, weakly
convex and sloping gently backwards; moderately large, rounded, median depres-
sion in posterior half of pygidium, and very shallow, smooth, elongate, transverse
to gently curved furrow in front of median depression (PI. 52, figs. 29, 32); appearing
to represent pair of muscle attachment areas connected by narrower, gently for-
wardly curved band across mid-line ; gentle depression extends in curve from outer
ends of this structure to posterior margin between mid-line and postero-lateral ex-
tremities; dorsal surface exhibits raised lines running subparallel to gently curved

EXPLANATION OF PLATE 53

Figs. 1-10. Remopleurides acer sp. nov., from the lower part of the Cliefden Caves Limestone, Fossil Hill.

I, Dorsal view of latex impression of external mould of cranidium; holotype SUP 28922, x6. 2, Dorsal

view of latex impression of external mould of cranidium; paratype SUP 28924, x8. 3, Dorsal view of
latex cast of external mould of cranidium; paratype SUP 28915, x6. 4-5, Lateral and anterior views
of cranidium; paratype SUP 28923, x 5. 6, Lateral view of part of left free cheek; paratype SUP 28921,
X 10. 7, Lateral view of portion of right free cheek; paratype SUP 28930, x 10. 8, Dorsal view of
four thoracic segments; paratype SUP 28925, x 8. 9, Oblique dorsal view of part of pygidium and last
two thoracic segments; paratype SUP 28920, x 10. 10, Dorsal view of pygidium and last thoracic

segment; paratype SUP 28919, x 12.

Figs. 11-24. Pseudobasilicusl fortis sp. nov., from the lower part of the Cliefden Caves Limestone; 11-12,
14-17, 22-24 from ‘mixed fauna’ unit east of Fossil Hill; 13, 18 21 from ‘lower coral’ unit. Fossil Hill.

II, Dorsal view of latex impression of exfoliated cranidium; holotype SUP 37900, x3. Note right,
posteriorly placed palpebral lobe. 12, Dorsal view of cranidium showing internal mould of lateral
glabellar furrows, median tubercle, and median ridge; paratype SUP 29939, x4. 13, Dorsal view of
small cranidium; paratype SUP 18947, x5. Also note prominent posteriorly situated palpebral lobe.
14, Dorsal view of right free cheek and genal spine; paratype SUP 29941, x 2-5. Note prominent lateral
and posterior borders. 15, Dorsal view of right free cheek showing doublure anteriorly; paratype
SUP 17941, x2. 16, Dorsal view of left free cheek; paratype SUP 29940, x2-5. 17, Oblique lateral
view of part of left free cheek showing elevated, rounded eye; paratype SUP 18901, x2-5. 18, Ventral
view of part of doublure beneath left free cheek and base of genal spine; paratype SUP 18908, x2.
19, Ventral view of relatively small hypostome; paratype SUP 29948, x6. 20, Dorsal view of part of
thoracic segment; paratype SUP 17937, x3. 21, Oblique lateral view of incomplete pygidium; para-
type SUP 17943, x4. Note conspicuous anastomosing lines. 22, Dorsal view of internal mould of
part of pygidium; paratype SUP 37902, x3. 23, Dorsal view of doublure on left side of pygidium;
paratype SUP 29947, x 1-5. 24, Dorsal view of large pygidium showing prominent postero-lateral
border; paratype SUP 17945, x 1-5.

PLATE 53

•Mm'

WEBBY, Remopleurides, Pseudohasiliciisl

460

PALAEONTOLOGY, VOLUME 16

posterior margin, except for part of median depression and elongate furrow. Another,
smaller, rounded median depression occurs at posterior margin or just beneath on
posterior edge of doublure (PI. 52, fig. 31). Doublure broad, widest antero-laterally,
presumably mainly underlying undifferentiated pleural regions; narrowing slightly
backwards and inwards, but medially having slight forward and upwardly directed,
tongue-like extension (PI. 52, fig. 31). Terrace lines developed along pygidial and
thoracic doublure, subparallel to margin. In two specimens (PI. 52, figs. 29, 32)
pygidium has sagittal length of 2-6 and 3-5 mm, and maximum width of 5-8 and
8 0 mm, respectively.

Remarks. The association of the small R. saenuros and the large R. sp. A in the same
silicified limestone fauna at Billabong Creek suggests the possibility of dimorphism
(see earlier discussion, p. 446). However, until a more complete specimen is obtained,
it may still be argued that this large form represents an entirely new remopleuridid.

Remopleurides sp. B

Plate 51, fig. 13

Material. A single, large internal mould (SUP 28904) from the Malongulli Formation
near Mirrabooka homestead, north of Cheeseman’s Creek P.O.

Comparative description. The specimen shows six thoracic segments in contact with
a damaged pygidium. The thoracic segments exhibit a very wide axis with gentle
forward convexity, and shorter backward and outwardly curved pleurae. Terrace
lines on the thoracic doublure are more or less subparallel to the margin. On the
pygidium, there is an imprint of a large rounded median elevation (depression in
doublure) between anterior and posterior margins, and transversely aligned terrace
lines. The specimen reaches a maximum width of 34 mm measured across the fifth
thoracic segment in front of the pygidium. It closely resembles R. sp. A, but because
of its association in the Malongulli Formation with R. exallos, is thought to repre-
sent a dimorph of this species, rather than of R. saenuros.

Possibly the Australian forms belong to a distinct species group characterized by
(1) smaller ‘male’ dimorphs with a median spine on the third thoracic segment in
front of the pygidium, and a small pygidium with relatively large, triangular axis
bisected by deep, median longitudinal furrow, and narrow, marginal pleural regions,
and (2) large ‘female’ dimorphs, lacking a median thoracic spine and with a large,
wide, poorly differentiated, subtrapezoidal pygidium.

Remopleurides sp. C
Plate 51, figs. 14 15

Material. Part of a damaged cranidium (SUP 28907), free cheek (SUP 28908), and
posterior part of thorax and pygidium (SUP 28909) from Ordovician shales (not
in situ), ^ mile north of Cheeseman’s Creek P.O.

Comparative description. The character of the ornamentation on the free cheek of
this species distinguishes it from other smaller New South Wales forms. Though
perhaps most closely resembling R. exallos, the triangular area inside the lateral

WEBBY: ORDOVICIAN TRILOBITES

461

border, limited by the posterior border furrow and the eye lobe, exhibits a prominent
granulation (PI. 51, fig. 15), and terrace lines are confined mainly to the genal spine,
the lateral border, and the doublure. The posterior border is well developed, raised,
smooth, and transversely aligned, unlike the narrow, curved ridge of the posterior
border in R. saenuros.

Family asaphidae Burmeister 1843
Genus pseudobasilicus Reed 1931

Type species. Ptychopyge lawrowi F. Schmidt 1898.

Pseudobasilicusl fort is sp. nov.

Plate 53, figs. 11-24; Plate 54, figs. 1-4

Material. Holotype (SUP 37900) and eighteen paratypes (SUP 17940-17941, 17945-
17946, 18901, 29936-29944, 29946-29947, 37902-37903) from the ‘mixed fauna’ unit
east of Fossil Hill, lower part of the Cliefden Caves Limestone. Eleven paratypes
(SUP 17937, 17943, 18902, 18908, 18947, 19907, 29909b, 29948, 37901, 37904-37905)
from the ‘lower coral’ unit on Fossil Hill, lower part of the Cliefden Caves Limestone.

Description. Cephalon subsemicircular, wider than long (sag.), with prominent
borders and strong genal spines. Glabella moderately convex, occupies between
one-quarter and one-third total width of cephalon ; defined by fairly well-differentiated
axial furrows, continuous into preglabellar furrow anteriorly. Axial furrows not
well defined posteriorly, except in small specimens and on internal moulds of larger
ones, and extending more or less directly backwards on to posterior margin. One
pair of prominent lateral glabellar furrows (Ip) seen to run obliquely backwards
and inwards from intersection with axial furrows near mid-length of glabella, and
apparently meet at mid-line just in front of small, median glabellar tubercle, the
latter situated on gentle rise of posterior part of glabella immediately in front of
occipital furrow (PI. 53, fig. 12). Internal moulds may show faintly impressed short,
curved, convex forward, lateral glabellar furrows (2p) with weakly raised, crescent-
shaped ridge just in front of each furrow; distally, 2p furrows meet Ip furrows, just
behind their point of intersection with axial furrows. 2p furrows subdivide convex,
raised, pear-shaped anterior part of glabella into large frontal lobe and small, lateral
glabellar lobes {2p). No observable differentiation of this pear-shaped anterior part
of glabella is seen in external moulds (PI. 53, fig. 11). Arched, subtriangular to L-
shaped lateral glabellar lobes (Ip) slope antero-medially toward Ip furrow from
raised course of axial furrow adjoining elevated palpebral lobe, and postero-laterally
toward lateral development of occipital furrow. Occipital furrow becomes weakly
defined and appears to be deflected slightly forwards medially. Internal moulds of
glabella may show faint median ridge running forward from median tubercle across
2p and frontal lobes (PI. 53, fig. 12). Occipital ring gently convex, narrowing laterally
owing to forward deflection of occipital furrow and slight backward curve of pos-
terior margin medially. Frontal part of fixed cheeks expanded into flat wing-like
areas, continuous into preglabellar areas. Anterior branches of facial suture dorsal-
intra-marginal, but meet at margin frontally; branches of suture cut obliquely across

462

PALAEONTOLOGY, VOLUME 16

weak, transverse-curving antero-median ridge (representing part of antero-lateral
cephalic border) on preglabellar area; median suture not developed on dorsal sur-
face. In small specimen (PI. 53, fig. 13) more deeply impressed preglabellar furrow
separating frontal lobe from narrow, raised ridge on preglabellar area, and crossed
by low, median, connecting ridge at mid-line. Palpebral lobe most elevated part of
cranidium, set posteriorly and adjacent to lateral glabellar lobe Ip (PI. 53, figs. 1 1
and 13). Maximum width across posterior margin of cranidium similar to total
width across frontal lobe of glabella and wings of fixed cheek.

Ornamentation consists of fine anastomosing lines running concentrically around
frontal part of glabella, constricted posteriorly toward median tubercle (PI. 54,
fig. 1). Anastomosing lines on occipital ring also exhibit concentric, convex forward
pattern. Lines directed outwards and forwards across axial furrow on to wing of
fixed cheek just in front of palpebral lobe, and run more or less parallel to antero-
lateral margin of cephalon on to preglabellar area. Across palpebral lobes, lines
orientated longitudinally (exsag.).

Free cheek exhibits large elevated, rounded eye, situated posteriorly, only about

EXPLANATION OF PLATE 54

Figs. 1-4. Pseudobasilicusl fortis sp. nov., from the lower part of the Cliefden Caves Limestone; 1, 3, from
the ‘mixed fauna’ unit east of Fossil Hill, and 2, 4, from the ‘lower coral’ unit, Fossil Hill. 1, Enlarged
dorsal view of external mould of part of cranidium showing pattern of concentric anastomosing lines;
paratype SUP 29938, x6. 2, Oblique lateral view of exfoliated pygidium, showing part of doublure;
paratype SUP 37901, X 2. 3, Dorsal view of incomplete pygidium; paratype SUP 29946, x 3. 4, Ven-
tral view of incomplete hypostome ; paratype SUP 29909b, X 4.

Figs. 5-18. Illaems {Parillaenus)! incertus sp. nov., from the ‘calcarenite’ unit at the top of the Ballingoole
Formation, Malachi’s Hill. 5, Dorsal view of internal mould of cranidium, showing posteriorly situ-
ated palpebral lobe, lateral muscle impressions, and median tubercle; holotype SUP 29932a, x6.
6-7, Dorsal and oblique lateral views of exfoliated cranidium; paratype SUP 18934, x4-5. 8, Dorsal
view of small cranidium, showing one pair of glabellar muscle scars on internal mould; paratype SUP
18931, X 8. 9, Anterior view of latex impression of cranidium, showing pattern of anastomosing lines;
paratype SUP 19941, x5. 10, Ventral view of doublure on pygidium, showing smooth, even curve
of inner edge; paratype SUP 29928, x4. 11, Dorsal view of internal mould of pygidium; paratype
SUP 29931c, x3'5. 12, Antero-dorsal and oblique lateral views of paratypes SUP 18933c (upper)

and SUP 18933b (lower), x 5. Note smooth curve of inner edge of doublure and of terrace lines on ex-
foliated part of lower specimen, and fine pitting and radiating lines on dorsal surface of upper speci-
men. 13, Dorsal view of internal mould of pygidium; paratype SUP 29932b, x4. 14, Oblique lateral
view of internal mould of cranidium; paratype SUP 18933a, x4. 15, Dorsal view of partly exfoliated
pygidium showing lateral muscle impression on one side and six pairs of small axial muscle scars on
internal mould; paratype SUP 29928b, x5. 16, Dorsal view of exfoliated pygidium; paratype SUP
29931a, x4. 17, Dorsal view of internal mould of small cranidium showing diverging axial furrows;
paratype SUP 17700, x 6. 18, Ventral view of small pygidium showing trace of axial furrows anteriorly,
one pair of lateral muscle impressions, fine, weak median ridge (groove ventrally) posteriorly, and smooth
curve of inner edge of doublure; paratype SUP 29929, x 8.

Figs. 19-20. lUaenus {Parillaenus)1 incertus sp. nov.?, also from ‘calcarenite’ unit at top of Ballingoole
Formation, Malachi’s Hill. 19, Dorso-lateral view of left free cheek, showing moderate-sized, crescent-
shaped faceted eye; SUP 29928a, x8. 20, Ventral view of internal mould of incomplete hypostome;
SUP 18900, x3.

Fig. 21. Illaenus (Parillaenus)! sp. A. Dorsal view of internal mould of compressed pygidium from the
Malongulli Formation near Mirrabooka homestead, Cheeseman’s Creek area; SUP 29935, x4.

PLATE 54

WEBBY, Pseudobasilicusl, Illaenus (Parillaenus)!

464

PALAEONTOLOGY, VOLUME 16

one-half its exsagittal length in front of posterior margin. Almost upper three-
quarters of eye surface composed of visual area, which extends in large arc of nearly
270°; visual area merges into eye-socle without break-in-slope, but there is change
in colour of exoskeleton and subhorizontal terrace lines appear on socle. Eye-socle
curves smoothly down into broad, gently convex platform of free cheek inside pro-
minent lateral border. Change-in-slope at inner edge of lateral border represents
lateral border furrow, which weakens posteriorly along genal spine. Anterior part
of free cheek has long, narrow extension of border, which tapers and descends
approaching mid-line (PI. 53, fig. 16). Posterior border relatively broad, gently con-
vex, transverse on inner part but curving gently backward into genal spine laterally,
flanked by shallow, depressed posterior border furrow, which appears to die out
approaching lateral border. Posterior border raised slightly above convex platform
area of free cheek. Genal spine has subtriangular cross section, with dorsal keel on
inner edge being continuous into raised posterior border, weakly concave outward
slope as continuation of lateral border, and ventro-lateral keel as extension of lateral
margin of cephalon. Inner and ventral surfaces of spine flattened and at right angles
to each other (PI. 53, fig. 18). Spine estimated to extend back to near three-quarters
of total length of thorax; sharply pointed and evenly curved backward in continuity
with lateral border.

Doublure underlies entire free cheek except near eye lobe and apparently beneath
posterior border furrow; also extends under wing of fixed cheek in front of palpebral
lobe and preglabellar area. Anteriorly, crossed by median suture, and indented by
prominent, broad, deep anterior notch; small, tongue-like deflection in even forward
and inward curve of anterior notch close to mid-line (PI. 53, fig. 15). Nature of inner
edge of doublure postero-laterally to eye lobe not known, and Panderian structures
not clearly observed. Terrace lines run subparallel to margins of cephalon, along
lateral and posterior borders, genal spine, and doublure; anastomosing lines on plat-
form area inside border curve around eye lobe, and directed backward to posterior
border furrow.

Hypostome of forked asaphid type, subovate, somewhat longer than wide, and
with middle body gently convex ; divided by deep, transverse, slot-like, middle furrows
into large anterior, and very small posterior lobe. Macula conspicuous, transversely
ovate, steeply inclined forward, situated on posterior side of middle furrow. Lateral
and posterior borders broad and long (exsag.) with deep posterior median notch;
sides only slightly divergent; posterior notch occupies about one-quarter of total
length (sag.) of hypostome. Inner edges of posterior notch, raised; laterally descend-
ing on to flattened lateral borders; just in front of hypostome mid-length, lateral
borders narrow into shoulder, which dies out anteriorly. Lateral notch lies between
shoulder and curving antero-lateral margin of middle body; extends upward and
outward into anterior wing. No anterior border. Doublure unknown. Anastomosing
terrace lines have concentric, convex forward pattern across middle body, and more
or less transverse arrangement of lines on posterior and lateral borders, with forward
deflection at lateral margins.

One poorly preserved specimen shows five thoracic segments and pygidium articu-
lated together. Axial rings convex (sag. and tr.), each separated by relatively shallow
articulating furrow from convex, tongue-like articulating half-ring, about three-

WEBBY: ORDOVICIAN TRILOBITES

465

quarters of length (sag.) of axial ring. Inner part of pleura flat, transverse; outer
part deflected slightly backwards and downwards beyond fulcrum. Character of
pleural ends and doublure unknown. Deep, sharp pleural furrow commences close
to front edge of inner corner of pleura and follows slightly oblique course, becoming
deepest near middle of pleura about fulcrum, and dying out near rear edge of pleura
some distance inside pleural termination. Large triangular facet on anterior side
of pleura beyond fulcrum, allowing articulation beneath posterior edge of preceding
pleura; bounded posteriorly by sharp ridge running obliquely outwards and back-
wards from fulcrum toward postero-lateral corner of pleural termination. Anasto-
mosing lines on axial ring concentrically arranged convex forward, and extend in
continuity across articulating furrow on to articulating half-ring. On inner part of
pleura, lines directed forward and very slightly outward; articulating facet apparently
smooth.

Pygidium subsemicircular, moderately convex; axis occupies about one-quarter
of total width at anterior margin; articulating half-ring and articulating furrow as
in thoracic segments. Axis raised, gently tapering posteriorly to rounded termination
set inside broad, flattened posterior border. As seen in internal moulds, number of
axial rings difficult to determine but at least twelve (seven to nine in anterior two-
thirds of axis). Ring furrows more deeply impressed laterally than medially. In
external moulds axial rings faintly or not at all defined. Pleural region exhibits raised
first pleural rib, continuous to lateral margin, with large triangular facet on its antero-
lateral surface. Broad, flattened border terminates against first pleural rib. Inside
border up to seven additional pleural ribs may be differentiated in internal moulds,
but only anterior four or five are clearly seen on external surfaces, separated by
sharply indented pleural furrows. Faint, discontinuous interpleural furrows may be
developed on a few anterior pleural ribs (PI. 53, fig. 22). Doublure broad, extending
beneath border areas and outer part of pleural field. Inner edge runs in gentle curve
from fulcrum towards posterior part of axis and is deflected in smooth sharp curve
around termination, giving rounded notched appearance. Broad spaced, anasto-
mosing terrace lines on dorsal surface run in rather sharp forward curve across axis,
extend outward along pleural ribs, and outward and forward across border, usually
with backward deflection at margin ; they also cross triangular facet. Well-developed
terrace lines run along doublure subparallel to posterior and lateral margins.

Judging from some large fragments of free cheeks, it is estimated that this species
may have reached a maximum length of about 105 mm, and a maximum width
across cephalon of about 60 mm.

Remarks. P.l fortis bears moderately close resemblances to the type species of
Pseudobasilicus, P. lawrowi (Schmidt 1898), from the Middle Ordovician (C^) of
the Leningrad region and Estonia, but differs in exhibiting a slightly more inflated,
pear-shaped glabella, a slightly narrower preglabellar area, more prominent lateral
cephalic borders, larger genal spines, and less pointed, backwardly curving thoracic
pleurae. Also the proportions across the posterior part of the cephalon are rather
different, with the glabella occupying about one-quarter of the total cephalic width
in P. lawrowi, and nearer one-third of the total width in E.? fortis. In consequence
the present species is only tentatively assigned to the genus. Schmidt’s (1898, 1904)

466

PALAEONTOLOGY, VOLUME 16

Other two Middle Ordovician species of Pseudobasilicus, P. kuckersianus and P.
kegelensis, from horizons Cjj and D^j, respectively, of Estonia, with their much
elongated, flattened preglabellar area, seem to be less closely related. Balashova
(1971), in recent revision of Schmidt’s species, has erected a new genus, Pseudo-
basiliella, for P. kuckersianus and P. kegelensis. Pseudobasilicusl brachyrachis (Tdrn-
quist) from the Middle Ordovician of the Siljan district, Sweden (Jaanusson 1953),
is also similar to the present species, but it too has several distinguishing points
— a more obtuse angle of intersection of anterior branches of the facial suture
at the mid-line, lack of prominent lateral cephalic borders, shorter, more rapidly
tapering genal spine, and wider pygidial doublure with more acute V-shaped inner
edges.

Of South-East Asian forms, Basiliella satunensis Kobayashi and Hamada (1964)
from Ordovician shales at Satun, southern Thailand, seems to be most closely related.
Though fragmentary, the cranidia of the Thai species are very closely similar to those
of P. ? fortis. Kobayashi and Hamada described the species as having anterior branch
of facial suture ‘a little intramarginal’, seemingly on dorsal side, ‘then abruptly bent
forward to meet its fellow on the axis’, presumably marginally. From their state-
ment and illustrations (pi. IX, fig. la-b), it is clear they correctly interpreted the suture
as extending dorsal-intra-marginally along the border, swinging across the ridge-
like edge of the border approaching the mid-line, and descending to meet the other
branch on the margin, almost exactly analogous to that seen in P. ? fortis. But they
assigned the species to Basiliella Kobayashi, a sub-genus of Basilicus Salter, charac-
terized by having a marginal suture in front of the glabella (Jaanusson 1959, p. O 336).
The eyes of B. satunensis are stated as being placed about mid-glabellar length (Koba-
yashi and Hamada 1964, p. 208). Yet a poorly preserved imprint of the free cheek
(pi. IX, fig. 6) actually shows the posterior edge of the eye lobe running into the
posterior border furrow. The large eyes seem to be situated posteriorly with raised
palpebral lobes as in P. ? fortis. Therefore, the Thai species should perhaps like P. ?
fortis be tentatively assigned to Pseudobasilicus. The basic differences between the
two species lie in the shorter and less slender genal spine, and hypostome with wide
diverging inner margins of posterior notch of /*.? satunensis.

The presence of a raised, median ridge across the preglabellar area of the small
specimen of P.l fortis (PI. 53, fig. 13) may invite the suggestion of a distant link with
species like Basiliella carinata Harrington from the Upper Tremadocian of Argentina
(Harrington and Leanza 1957, p. 144).

Pseudobasilicus! sp. A
Plate 55, figs. 1-19

Material. Fragmentary silicified cranidium, large and small free cheeks, an almost
complete hypostome, and two fragmentary pygidia from Ordovician limestone at
Billabong Creek (SUP 37906-37913); also incomplete hypostome from Quondong
Formation, Bowan Park (SUP 37914).

Comparative description. This species is described with particular emphasis on
characters which distinguish it from P.! fortis, and on additional features not seen
in the latter. Small, weakly convex, fragmentary cranidium has less inflated frontal

WEBBY: ORDOVICIAN TRILOBITES

467

lobe, narrower, posteriorly tapering wings of fixed cheek, and relatively larger,
though less elevated, palpebral lobe (PI. 55, fig. 1). Large specimen of free cheek
(PI. 55, figs. 2-4) retains part of visual surface of eye, showing on inside surface
numerous tiny hexagonal facets; relatively longer (exsag.) and less elevated eye
than in P.l fortis. Inner edge of doublure extends in gentle curve outward from just
inside posterior margin (PI. 55, fig. 3); detailed nature of course uncertain because
of incomplete silicification, and no evidence of Panderian structures. In small speci-
mens, lateral border narrows posteriorly toward relatively smaller genal spine (PI. 55,
figs. 5-7). Faint terrace lines on doublure, borders, and genal spine.

Hypostome with gently convex, subrounded middle body, subdivided into large
anterior, and very small posterior lobes by marked, transversely elongate middle
furrows. Small, ovate, forwardly tilted macula situated near lateral border just
behind middle furrow. Broad, flattened, lateral, and posterior borders with bluntly
pointed postero-lateral prongs to either side of notch, and diagnostic sharp nick
half-way along margin of lateral border. Inner, ventral edges of notch sharply
crested ; sides slope upward and outward on to convex, dorsal surface of posterior
part of doublure. Inner edge of doublure turns sharply upward into short, posterior
wing directly above and behind macula; inclined inward and backward at about 45°
to exsagittal line. Broad, shallow groove running forward along doublure outside
posterior wing, continuous into downward and forwardly curving lateral notch
between shoulder and anterior wing; possibly formerly occupied by antenna or
similar structure. Inner edge of doublure between posterior wing and shoulder
slightly upturned. Anterior margin curved gently forward, but with small, shallow
notch across mid-line. Backward and upward continuation of anterior margin runs
into very broad anterior wing. Anterior wing broad, subtriangular, upwardly and
backwardly directed; dorsal edge slightly damaged, but well enough preserved to
show, in side view, asymmetrical profile with gentle backward curve anteriorly and
sharply deflected postero-dorsal tip. Anastomosing terrace lines directed more or
less transversely across middle body and in front of posterior median notch; de-
flected forwards on either side across lateral borders, shoulders, and lateral notches
to intersect anterior margin obliquely.

Pygidium gently convex, with much less well-differentiated axis and pleural regions
than in P. ? fortis. Only very faint trace of about seven pleural ribs on internal surface
of large specimen, 39-5 mm long (sag.) and estimated to have been about 46 mm
wide (PI. 55, fig. 13); evenly curved margin but for slight flattening of posterior
margin and very weak upward deflection medially (PI. 55, fig. 16). Less prominent
lateral and posterior borders than in P.l fortis. First pleural furrow very weakly
developed. Axis occupies about three-sevenths of total width of pygidium anteriorly.
Broad tongue-like appearance of articulating half-ring. Triangular articulating facet
with moderately sharply rounded antero-lateral corner. Doublure occupies about
one-half width of pleural field anteriorly ; inner edge gently curved between fulcrum
and posterior end of axis, broadly V-shaped, sharply rounded at posterior end of
axis; terrace lines well exhibited subparallel to margin (PI. 55, figs. 13 and 18). In
large specimen, ornamentation of scallop-shaped markings especially on lateral
border; irregular to transverse, widely spaced lines inside lateral border (PI. 55,
figs. 15-16). Smaller specimen exhibits anastomosing lines intersecting margin

468

PALAEONTOLOGY, VOLUME 16

obliquely, directed forward and outward, and fine pitting (PI. 55, figs. 17 and 19);
terrace lines also run outward on facet.

To sum up, P. ? sp. A is distinguished from P. ? fortis in the following main charac-
ters: (1) a less inflated frontal lobe, narrower wing-like lateral areas of fixed cheeks
of cranidium, and relatively longer (exsag.) and less elevated eyes, (2) a prominent
nick on lateral margin of border of hypostome, and (3) a poorly differentiated axis
and pleural regions of pygidium.

Family illaenidae Hawle and Corda 1847
Genus illaenus Dalman 1827
Subgenus parillaenus Jaanusson 1954

Type species. Illaenus fallax Holm 1882.

Illaenus {Parillaenus)! incertus sp. nov.

Plate 54, figs. 5-18

Material. Holotype (SUP 29932a) and nineteen paratypes (SUP 17700, 18928,
18931-18932, 18933a-c, 18934, 19914, 19941, 20905, 20910, 29928b, 29929, 29931a-c,
29932b, 29933) from the ‘calcarenite’ unit of the Ballingoole Formation, Bowan Park
Group (Semeniuk 1970), at Malachi’s Hill.

Description. Cranidium moderately convex, subsemicircular in outline, wider than
long; widest across posteriorly placed palpebral lobes; glabella in posterior part of
cranidium bounded by prominent, fairly deep axial furrows as seen in internal
moulds, but only weakly developed in external moulds. From posterior margin,
axial furrows directed forward and slightly inward to inner edges of pair of large,
exsagittally elongate, oval lateral muscle impressions (lunettes); appear to be de-
flected forward and outward in front of impressions; internal moulds of small
specimen (PI. 54, fig. 17) exhibits weakly impressed axial furrows diverging outwards
in front of lateral muscle impressions. Lateral muscle impressions set only slightly
in front of palpebral lobes, at about one-third of total cranidial length (sag.) from
posterior margin. Just in front of anterior edges of lateral muscle impressions, and
immediately across axial furrows on glabella, are pair of elongate, obliquely placed

EXPLANATION OF PLATE 55

Figs. 1-19. Pseudobasilicusl sp. A from Ordovician limestone at Billabong Creek. 1, Dorsal view of
part of cranidium; SUP 37906, x 3. 2-4, Dorsal, ventral, and lateral views of part of left free cheek;
SUP 37907, X 1-5. 5, Dorsal view of left free cheek of small specimen; SUP 37908, x4. 6, Dorsal view
of anterior part of right free cheek of small specimen; SUP 37909, x4. 7, Ventral view of posterior
part of left free cheek of small specimen; SUP 37910, x4. 8-12, Ventral, dorsal, anterior, lateral, and
posterior views of nearly complete hypostome; SUP 37911, X 3. 13-16, Ventral, dorsal, enlarged dorsal,
and posterior views of large fragmentary pygidium; SUP 37912; 13-14, 16, x 1; 15, x3. Note scallop-
shaped ornament on dorsal surface of lateral border. 17-19, Lateral, ventral, and dorsal views of
incomplete pygidium; SUP 37913, x2-5.

Figs. 20-21. Harpid gen. et sp. ind. Dorsal and ventral views of incomplete specimen of fringe from
Ordovician limestone at Billabong Creek; SUP 29949, x4.

PLATE 55

WEBBY, Pseudohasilicusl, harpid

470

PALAEONTOLOGY, VOLUME 16

glabellar muscle scars (basal scars of Jaanusson 1954); they are orientated more or
less at right angles to adjacent axial furrows and subparallel to antero-lateral margins
of cephalon (PI. 54, figs. 5 and 8). Glabellar width at level of palpebral lobes about
three-sevenths of total cranidial width. Toward posterior margin, tiny median
tubercle well exhibited on internal moulds of glabella. Internal mould of posterior
margin of glabella shows narrow rolled edge, or incipient doublure. Posterior branch
of facial suture runs only short distance backward on to posterior margin; anterior
branch of suture has somewhat sigmoidal course, firstly forward and slightly out-
ward, then curving gently inward to meet margin antero-laterally; sharp curve at
margin into connective suture which continues in gentle arc flanked by doublure of
free cheek and presumably rostral plate. Anteriorly, cranidium strongly curved
downward, with greatest curvature towards ventral edge, giving slightly backwardly
deflected anterior margin. In anterior aspect, outline made by deflected anterior
margin seems to be flattened and horizontal. Anterior part of cranidium exhibits
anastomosing terrace lines subparallel to margin, appearing like more or less hori-
zontally arranged contours with slight backward and upward deflection along
anterior branch of suture (PI. 54, fig. 9).

From outline of facial suture and position and size of palpebral lobe, free cheek
seems to have been rather small, with eye situated posteriorly. One large fragmentary
specimen (PI. 54, fig. 19) of a free cheek from the same locality and horizon, possibly
belonging to the species, shows a moderately large, crescent-shaped eye with num-
erous tiny facets, and evenly rounded genal angle.

Only broken fragments of rostral plate seen; surface with very coarse transverse
terrace lines.

Hypostome (PI. 54, fig. 20), also from same locality and horizon, provisionally
placed in species; relatively large, and exhibits almost straight, unnotched anterior
margin, with expanded anterior wing. Anterior lobe of middle body subcircular,
moderately convex, separated by deep furrow from lateral border. Lateral margins
incompletely preserved. Middle furrow runs backward and inward from deep lateral
border furrow, and dies out adjacent to small, oval macula. Posterior lobe of middle
body poorly differentiated, especially medially; separated from raised, gently arched
posterior border by broad (sag. and exsag.) and deep postero-lateral border furrow.
Faint, small pits and subconcentrically, well-spaced lines developed on surface of
middle body, and lines run parallel to margin on posterior border.

Thoracic segments unknown, but judging from relative widths of posterior margin
of glabella and anterior edge of pygidial axis, narrows only slightly backward along
thorax.

Pygidium moderately convex (tr. and sag.), subsemicircular in outline; widest
across anterior part, just behind antero-lateral corner; slightly wider than long.
Axis occupies about one-third of pygidial width anteriorly; usually only differ-
entiated from pleural areas at anterior margin; axial furrows deeply impressed at
margin with small downward and backward deflection of anterior edge. Relatively
broad, shallow, diagonal furrow arises at anterior extremity of axial furrow and
continues back and out toward lateral margin; isolates from rest of pygidium rather
small, narrow, raised first pleural lobe with steep forward and outward facet-like
area on antero-lateral corner. Doublure narrow, about one-fifth length (sag.) of

WEBBY: ORDOVICIAN TRILOBITES

471

pygidium, evenly rounded laterally and posteriorly, with very slight narrowing in
antero-lateral direction; not indented or notched sagittally; terrace lines, usually
about 10-12, run subparallel to margin. On undersurface of one well-preserved
small specimen (PI. 54, fig. 18), trace of axial furrows seen extending backward from
anterior margin toward large pair of oval, exsagittally elongate ‘lateral’ muscle
impressions; rear edge of these impressions near mid-length of pygidium; axis forms
subtriangular area lying inside axial furrows, with lateral muscle impressions appar-
ently situated on outer side of axial furrows, strictly on inner part of pleural field.
Axis tapers to point immediately behind rear edge of these muscle impressions, with
faint, fine median ridge, seen as furrow on undersurface, continuing to inner edge of
doublure. Pattern of muscle scars on axis preserved on internal mould of one well-
preserved specimen (PI. 54, fig. 15). Faint trace of large, crescent-shaped lateral
muscle impression (one of a pair) and, on axis, inside and to front of it, six pairs of
small muscle scars. First three pairs faint, elongate scars, closely spaced and almost
transversely aligned; fourth pair, larger, pear-shaped and placed diagonally just
in front of lateral muscle impression; last two pairs, subequal, rounded, set in line
(but inside line of fourth pair), opposite anterior part of lateral muscle impression.
Dorsal surface usually smooth, but one specimen (upper, PI. 54, fig. 12) shows fine,
widely spaced lines with intervening tiny pits running postero-laterally away from axis.

Remarks. This species may eventually prove to belong to a new genus, but addi-
tional, preferably articulated, material must be found and prepared to elucidate
details of the rostral plate. It exhibits similarities to both Illaenus (Parillaenus) and
Bumastus Murchison, though strictly is not identical with either. The relatively
narrow posterior part of the glabella and the less outwardly diverging axial furrows
to the front are features more typical of Illaenus {Parillaenus) than Bumastus. Neither
genus, nor for that matter any other illaenid, appears to have the same pattern of
an elongate, ‘basal’ glabellar muscle scar situated beside the axial furrow as in
I. (P.)? incertus.

The even curve of the inner margin of the pygidial doublure, which characterizes
the Parillaenus group (Jaanusson 1954), is displayed by the New South Wales species,
though not with a shallow median furrow on the surface of the doublure as some-
times occurs in the type species, I. fallax Holm, from the Kullsberg and Chasmops
Limestones of Sweden (Warburg 1925). Unfortunately, no pygidial muscle impres-
sions have been recorded from species of Illaenus {Parillaenus) for comparison. In
Bumastus bouchardi (Barrande) from the Silurian of Bohemia, Snajdr (1957) has re-
corded two pairs of pygidial muscle impressions, a large postero-lateral, and a small,
antero-median pair. The large, postero-lateral pair resembles the large, lateral pair
of impressions in I. {P.)l incertus. But the one small antero-median pair bears little
resemblance to the pattern of six small ‘axial’ pairs of scars of I. (P.)? incertus.

The present species also has similarities with forms like Bumastus {Bumast aides)
aplatus (Raymond) from the Chazyan of New York (Shaw 1968). However, the
relatively more widely spaced axial furrows on the cranidium behind the lateral
muscle impression, and the relatively wider and more flattened pygidium, with the
pygidial doublure exhibiting on its ventral surface a shallow median furrow, readily
distinguish the Chazyan species.

472

PALAEONTOLOGY, VOLUME 16

Features of the cranidium of /. (F.)? incertus, particularly the moderately small,
posteriorly situated eyes and the palpebral lobes placed well out from the axial
furrows, suggest a close relationship with Stenopareia Holm 1886. But the hypo-
stome and pygidium are markedly different. In the type species of Stenopareia,
S. linnarssoni (Holm 1882) from the Boda Limestone (Middle Ashgill) of the Siljan
district, Sweden, the hypostome is short (sag.) and subpentagonal, and the pygidium
has strongly truncated antero-lateral angles and a pygidial doublure with a sharply
pointed, forwardly directed projection on the mid-line (Holm 1886, Warburg 1925).

Illaenus {Parillaenus)! sp. A
Plate 54, fig. 21

Material. One specimen (SUP 29935) from shales of the Malongulli Formation near
Mirrabooka homestead, north of Cheeseman’s Creek.

Comparative description. This specimen is somewhat flattened by compression and
poorly preserved, but does show a relatively narrower axis at anterior margin of
pygidium, and apparently a somewhat broader doublure.

Family harpidae Hawle and Corda 1847
Harpid gen. et sp. ind.

Plate 55, figs. 20-21

Material. Numerous silicified fragments of fringe from upper part of Cliefden Caves
Limestone and Ordovician limestone at Billabong Creek. One diagnostic specimen
(SUP 29949) from Billabong Creek shows prolongation of fringe and girder.

Description. On lower lamella of fringe, raised girder extends obliquely in to gently
curving internal rim of prolongation behind postero-lateral corner of cheek. On
upper lamella, narrow, slightly irregular band corresponds to girder beneath, lying
between pitted areas of genal roll prolongation and brim prolongation respectively.
Genal roll prolongation steeply sloping; brim prolongation concave, becoming
flatter posteriorly, with gentle outward slope. Prominent gently curving, convex
external rim with adjacent row of coarse pits; pronounced ridge on upper lamella
and corresponding weaker ‘pseudogirder’ on lower lamella inside row of coarse pits;
toward tip of prolongation, ridge dies out. External rim has trace of marginal suture
dividing it into two; marginal band relatively broad, slightly concave above and
below fine median ridge with trace of suture. Numerous, irregular pits covering
surface of brim prolongation, with slightly larger pits adjacent to outer ridge, girder
and preserved fragment of genal roll prolongation. Approximately eleven rows of
pits between girder and external rim. Some fringe fragments show typical harpid
feature of fine radiating, somewhat anastomosing ridges across row of pits.

Remarks. Judging from the number of silicified fragments, this harpid species is
fairly abundant in the limestones. It probably represents a new Australian genus,
bearing resemblances to Selenoharpes Whittington (19506) in having a girder which
intersects the internal rim, but differing fundamentally in showing a ridge and

WEBBY: ORDOVICIAN TRILOBITES

473

corresponding ‘pseudogirder’ immediately inside the first row of large pits at the
external margin, and narrower prolongations with relatively coarser and fewer rows
of pits. A Lower Ordovician harpid has been recorded from Waratah Bay, Victoria
(Singleton, in Lindner 1953), and, more recently, Whittington and Hughes (1972)
have listed a harpid from the Gordon Limestone of Tasmania.

Acknowledgements. I wish to express my gratitude to Professor H. B. Whittington for providing facilities
and counsel during work at the Sedgwick Museum, University of Cambridge, while on study leave in
1971-1972. Professor Whittington and Dr. C. P. Hughes kindly read and criticized the manuscript. The
work has been aided by the award of a Royal Society and Nulheld Foundation Commonwealth Bursary,
and by funds from the Australian Research Grants Committee and a Sydney University Research Grant.
In addition to collections made by myself and Dr. G. H. Packham, material has been kindly provided
by Professor G. M. Philip, Messrs. D. Morris, V. Semeniuk, L. Sherwin, and M. Tuckson.

REFERENCES

BALASHOVA, E. A. 1971. K Ustanovlenijo novogo podsemejstva Pseudobasilicinae. Voprosy Paleontologii,
6, 52-60.

BILLINGS, E. 1862. Palaeozoic Fossils, I. Containing descriptions and figures of new or little known species
of organic remains from the Silurian rocks. Geol. Surv. Canada, 25-168.

BURMEISTER, H. 1843. Die Organisation der Trilobiten. 147 pp. Berlin.

CAMPBELL, K. s. w. and DURHAM, G. J. 1970. A New Trinucleid from the Upper Ordovician of New South
Wales. Palaeontology, 13, 573-580.

CHAPMAN, F. 1914. Australasian Fossils. 341 pp. Melbourne.

DALMAN, J. w. 1827. Om palaeoderma eller de sa kallade Trilobiterna. K. svenska Vetensk-Akad. Handl.
1, 113-152, 226-294.

HARRINGTON, H. J. and LEANZA, A. F. 1957. Ordovician Trilobites of Argentina. Univ. Kansas Dept. Geol.,
Spec. Publ. 1, 1-276.

HAWLE, I. and CORDA, A. J. c. 1847. Prodrom einer Monographic der bohmischen Trilobiten. 176 pp. Prague.

HOLM, G. 1882. De svenska arterna af trilobitslagter Illaenus (Dalman). Bihang till. Kungl. svenska Vetensk-
Akad. Handl. 7, (3), 1-148.

1886. In SCHMIDT, F., Revision der ostbaltischen silurischen Trilobiten, Abt. III. Illaeniden. Mem.

Acad. imp. Sci. St. Petersbourg, (7), 33, (8), 1-173.

HOWELL, B. F. 1935. Cambrian and Ordovician trilobites from Herault, southern France. J. Paleont. 9,
222-238.

HUGHES, c. p. 1969. The Ordovician trilobite faunas of the Builth-Llandrindod inlier. Central Wales.
Part I. Bull. Brit. Mus. nat. Hist. {Geol.), 18, 39-103.

INGHAM, J. K. 1970. The Upper Ordovician trilobites from the Cautley and Dent districts of Westmor-
land and Yorkshire. Part 1. Palaeontogr. Soc. Monogr. 1-58.

JAANUSSON, V. 1953. Untersuchungen fiber baltoskandische Asaphiden. I. Revision der mittelordovizischen
Asaphiden des Siljan-Gebietes in Dalarna. Arkiv f. Mineral. Geol. 1 (14), 377-464.

1954. Zur Morphologie und Taxonomie der Illaeniden. Ibid. 1 (20), 545-583.

1959. Family Asaphidae. In moore, r. c. (ed.). Treatise on Invertebrate Paleontology, Part O, Arthro-

poda 1, 334-355. Univ. Kansas Press.

KiELAN, z. 1960. Upper Ordovician trilobites from Poland and some related forms from Bohemia and
Scandinavia. Palaeont. Polonica, 11, 1-198.

KOBAYASHi, T. and HAMADA, T. 1964. On the Middle Ordovician Fossils from Satun, the Malaysian Fron-
tier of Thailand. Jap. J. Geol. Geogr. 35, 205-211.

KOROLEVA, M. N. 1964. Novye sredneordovikskie trilobity Shumardia sevemogo Kazakhstana. Paleont.
zhurn. 1964 (1), 71-75.

LAKE, p. 1907. A Monograph of the British Cambrian trilobites. Part 2. Palaeontogr. Soc. Monogr., 29-48.

LINDNER, A. w. 1953. The Geology of the Coastline of Waratah Bay between Walkerville and Cape Lip-
trap. Proc. R. Soc. Viet. 64, 77-92.

D

474

PALAEONTOLOGY, VOLUME 16

LiNNARSSON, j. G. o. 1869. Om Vestergotlands Cambriska och Siluriska aflagringar. K. svenska Vetensk-
Akad. Hand!. 8, (2), 1-89.

MCCOY, F. 1849. On the classification of some British fossil Crustacea, with notices of some new forms in
the University collection at Cambridge. Ann. Mag. nat. Hist. (2), 4, 161-179, 330-335, 392-414.
MILLER, s. A. 1889. North American geology and paleontology. 664 pp. Cincinnati.

MOORS, H. T. 1970. Ordovician graptolites from the Cliefden Caves Area, Mandurama, N.S.W., with a
re-appraisal of their stratigraphic significance. Proc. R. Soc. Viet. 83, 253-287.

PACKHAM, G. H. 1967. The Occurrence of Shelly Ordovician Strata near Forbes, New South Wales. Aust.
J. Sci. 30, 106-107.

1969. Ordovician System. The Geology of New South Wales. J. geol. Soc. Aust. 16, 76-103.

PITTMAN, E. F. 1900. The auriferous ore-beds of the Lynhurst Gold Field. Rec. geol. Surv. N.S.W. 7,
9-15.

PORTLOCK, J. E. 1843. Report on the geology of the County of Londonderry^ and of parts of Tyrone and Fer-
managh. xxxH-784 pp. Dublin and London.

REED, F. R. c. 1903. The Lower Palaeozoic Trilobites of the Girvan District, Ayrshire. Part I. Palaeontogr.
Soc. Monogr. 1-48.

1931. A review of the British species of Asaphidae, Part 2. Ann. Mag. nat. Hist. (10), 7, 441-472.

SCHMIDT, F. 1898. Revision der ostbaltischen Trilobiten, Abt. V, Lief. I. Mem. Acad. imp. Sci. St. Peters-
bourg, (8), 6, (11), 1-45.

1904. Revision der ostbaltischen Trilobiten, Abt. V, Lief. III. Mem. Acad. imp. Sci. St. Petersbourg,

(8), 14,(10), 1-68.

SEMENIUK, v. 1970. The Lower- Middle Palaeozoic Stratigraphy of the Bowan Park Area, Central-Western
New South Wales. J. Proc. R. Soc. N.S.W. 103, 15-30.

SHAW, F. c. 1968. Early Middle Ordovician Chazy Trilobites of New York. N.Y. State Mus. Mem. 17,
1-163.

SMITH, R. E. 1966. The Geology of Mandurama-Panuara. J. Proc. R. Soc. N.S. W. 98, 239-262.

Snajdr, m. 1957. Classification of the family Illaenidae (Hawle et Corda) in the lower Palaeozoic of Bohe-
mia. Sbornik Ust. ust. Geol. 23, 125-284.

STEVENS, N. c. 1952. Ordovician stratigraphy at Cliefden Caves, near Mandurama, N.S.W. Proc. Linn.
Soc. N.S.W. 77, 114-120.

1957. Further notes on Ordovician formations of central New South Wales. J. Proc. R. Soc. N.S.W.

90, 44-50.

STRUSZ, D. L. 1960. The Geology of the Parish of Mumbil, near Wellington, N.S.W. J. Proc. R. Soc. N.S. W.
93, 127-136.

1961. Lower Palaeozoic corals from New South Wales. Palaeontology, 4, 334-361.

THORSLUND, p. 1940. On the Chasmops Series of Jemtland and Sodermanland (Tvaren). Sver. geol. Unders.
(C), 436, 1-191.

WARBURG, E. 1925. The trilobites of the Leptaena Limestone in Dalarne. Bull. Geol. Inst. Univ. Uppsala,
17, 1-446.

WEBBY, B. D. 1969. Ordovician stromatoporoids from New South Wales. Palaeontology, 12, 637-662.

1971. The trilobite Pliomerina Chugaeva from the Ordovician of New South Wales. Palaeontology,

14, 612-622.

1972. The Rugose Coral Palaeophyllum Billings from the Ordovician of central New South Wales.

Proc. Linn. Soc. N.S.W. 97, 150-157.

MOORS, H. T. and mclean, r. a. 1970. Malongullia and Encrinuraspis, New Ordovician Trilobites from

New South Wales, Australia. J. Paleont. 44, 881-887.

WHITTARD, w. F. 1955. The Ordovician Trilobites of the Shelve Inlier, West Shropshire, Part I. Palaeontogr.
Soc. Monogr. 1-40.

WHITTINGTON, H. B. 1950cr. Sixteen Ordovician genotype trilobites. J. Paleont. 24, 531-565.

19506. Monograph of the British trilobites of the family Harpidae. Palaeontogr. Soc. Monogr.

1-55.

1959. Silicified Middle Ordovician trilobites: Remopleurididae, Trinucleidae, Raphiophoridae,

Endymioniidae. Bull. Mus. Comp. Zool. Harv. 121, 371-496.

1963. Middle Ordovician trilobites from Lower Head, western Newfoundland. Bull. Mus. Comp.

Zool. Harv. 129, 1-118.

WEBBY: ORDOVICIAN TRILOBITES

475

WHITTINGTON, H. B. 1965. Trilobitcs of the Ordovician Table Head Formation, western Newfoundland.
Bull. Mus. Comp. Zool. Harv. 132, 275-442.

1966. Phylogeny and distribution of Ordovician trilobites. J. Paleont. 40, 696-737.

and HUGHES, c. p. 1972. Ordovician geography and faunal provinces deduced from trilobite distri-
bution. Phil. Trans. R. Soc. {B), 263, 235-278.

B. D. WEBBY

Department of Geology
University of Sydney
Sydney, N.S.W., 2006

Revised typescript received 17 October 1972 Australia

jA- 9

■* -vv

tf <r

’i. ,... • ,'\([ff'j^5lj; ' ~ % -I

•.' V, * . .. ‘^at. f •

• V-Hf > M ..... ■ i ;.*:*■ ’ #: ' I

I"

■?

o

“ 1

^

J>

. V

■'' \

r.'i

U-

K.?!.

LOWER CARBONIFEROUS CONODONT
FAUNAS FROM

THE EASTERN MENDIPS, ENGLAND

by MALCOLM BUTLER

Abstract. In the Mendip Hills, the lower part of the Carboniferous Limestone succession (the Lower Limestone
Shale and Black Rock Groups) extends to greater thickness than is found in the Avon Gorge at Bristol. Conodont
faunas from this range of the Mendip succession are described here. A distinctive lag-type deposit occurs low in
the Lower Limestone Shale Group. Faunas below this lag (with Patrognathus, Pseudopolygnathus dentilineatus.
Polygnathus symmetricus) cannot yet be dated precisely. Above, faunas with Siphonodella, Elictognathus, and
Gnathodus correlate with those known from the late Kinderhookian of the U.S.A. and probably represent the range
of the upper Siphonodella crenulata-Zone of Germany. Higher, in the middle part of the Black Rock Group, faunas
typical of the German anchoralis-Zone appear. Particular interest attaches to the occurrence of Scaliognathus anchor-
alis, Dollymae bouckaerti, Pelekysgnathus bultyncki, and related forms. These compare closely with faunas recently
described from Tn 3c in Belgium. None of the anchoralis-Zone species are known from the Avon Gorge succession,
which is considered to be incomplete.

The Carboniferous Limestone of the Bristol area provided the evidence on which
Vaughan (1905) based his scheme of coral-brachiopod zones. More recently Kella-
way and Welch (1955) have erected a series of lithological units for the Carboniferous
Limestone of this area, intended to supersede Vaughan’s zones. Neither of these
schemes enables correlation to be made on anything more than a local scale. In an
attempt to correlate on an international scale, Rhodes, Austin, and Druce (1969)
studied the conodont faunas of the Avon Gorge section. However, these authors
failed to recover certain important species of conodont, in particular the indices
of the anchoralis-Zone (Bischoff 1957; Voges 1959, 1960).

Workers using both Vaughan’s (1905) zones (e.g. Welch 1932) and the litho-
logical units of Kellaway and Welch (1955) have noted that the strata near the
top of the Black Rock Group (Z to CJ thicken southwards from Bristol into the
Mendips (see Kellaway and Welch 1955, pi. 1). It was decided to sample the Lower
Limestone Shale and Black Rock Groups in the eastern Mendips to test the possi-
bility that since the succession is thicker there, it might also be more complete,
and might therefore produce conodont faunas which are not available in the Avon
Gorge.

The sedimentary petrology of the sections sampled for conodonts has been
studied in detail, and in addition a brief study has been made of the nature of equi-
valent levels in the Avon Gorge. The results of these investigations are considered
elsewhere (Butler 1972). The sections were divided into sedimentological units,
a brief description of each being given in text-figs. 3 to 10, along with details of
sampling localities and sample numbers.

In the eastern Mendips the best section in the Lower Limestone Shale Group is
that seen in the disused railway cutting at Maesbury (ST 606 475). Here the upper
two-thirds of the Group are exposed, together with the transition to the overlying
Black Rock Group (Green and Welch 1965). The lower and middle parts of the

[Palaeontology, Vol. 16, Part 3, 1973, pp. 477-517, pis. 56-59.]

478

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 1. Locality map, showing outcrop of the Carboniferous Limestone in the Bristol- Mendip area.

BUTLER: CARBONIFEROUS CONODONTS

479

Black Rock Group are seen in disused quarries at Windsor Hill and Ham Woods
(ST 615 452) and the middle and upper parts of this group are well exposed in a
working quarry at Halecombe (ST 702 475). These sections were sampled syste-
matically for conodonts, 2-kg samples being taken at 3-metre intervals. In addition
a number of samples were taken from the middle and upper parts of the Black Rock
Group in Vallis Vale (ST 755 490). This section is likely to provide a more per-
manent record than Halecombe Quarry, which is being actively worked. Isolated
samples were also taken from Lower Limestone Shale Group exposures at Portis-
head (ST 465 775), Clevedon (ST 402 718), and Asham (ST 717 463) and from the
Palate Bed in the Avon Gorge (ST 555 746). Faunal lists for these samples are given
in an appendix, and the forms recovered are not treated systematically.

Conodont occurrence charts are given in text-figs. 3 to 10. 167 samples were pro-
cessed and yielded 5324 identifiable specimens. 2491 of these are ‘bar’ forms. Yields
per sample averaged 36 specimens in the Lower Limestone Shale Group and the
lower and middle parts of the Black Rock Group. Certain samples gave relatively
high yields, over 500 in one case (HW 18). Yields were especially poor in the upper
part of the Black Rock Group (unit br 4), only three samples producing conodonts,
and in this part of the succession samples were processed from 6-metre intervals
only. Samples whose number bears a suffix ‘a’ come, in each case, from a level Ij
metres above the preceding sample (e.g. HW 18a came from 1^ metres above HW18).
Duplicate samples were processed in a number of cases. None showed any marked
differences from what was found in the first sample. Acetate peels were made of
all rocks sampled for conodonts. The peels, bearing sample numbers, are stored in
the Geology Museum, University of Bristol. There is no clear evidence of any link
between conodont abundances and particular lithologies.

Samples were digested in 10% acetic acid and residues collected by sieving to
125-mesh sieve. Heavy liquid separation was carried out on the dried residues, using
the methods described by Collinson (1963).

CONODONT OCCURRENCES IN THE EASTERN MENDIPS

Details of the occurrences of conodonts discussed here are available in text-
figs. 3-10.

The Lower Limestone Shale Group. Conodont faunas recovered from the lowest
part of the Lower Limestone Shale Group at Maesbury include Polygnathus sym-
metricus Branson and Pseudopoly gnathus dentilineatus Branson. In addition, two
specimens of Patrognathus variabilis Rhodes, Austin, and Druce were recovered
from sample Ma 4.

At the base of unit m 2 coarse limestones with phosphatic nodules occur, forming
a lag-type deposit, and an abrupt change in fauna takes place. The genera Siphono-
della, Elictognathus, and Gnathodus first appear here (in sample Ma 7). The interval
Ma 7 to Ma 19a is characterized by the presence of Siphonodella obsoleta Hass,
S', isosticha (Cooper), S. cf. S. isosticha (Cooper), S. cooperi Hass, and S. cf. crenu-
lata (Cooper). Two specimens of Gnathodus punctatus (Cooper) were recovered from
sample Ma 7 and two specimens of G. delicatus Branson and Mehl were found in
sample Ma 9. Polygnathus inornatus Branson ranges from Ma 7 to Ma 19a, and

480

PALAEONTOLOGY, VOLUME 16

Lower Limestone
Shale Group

Block Rock Group

to D o

I-’ O =i,
' o 5

3 ;

► *■
»■ ►

|3 ‘n

Patrognothus vonobilis

Siphonodella spp.

■ Elictognothus lacerotus

Gnathodus punctatus

Grothodus delicatus
■■ Pseudopolygnothus multistnatus
Polygnathus communis carina
Pelekysgnothus bultyncki
Dollymoe bouckoerti

Polygnothus cf P symmetricus
Gnothodus bulbosus

■ Pseudopolygnothus tnongulus pinnotus

■ Scoliognothus anchorolis

G texonus pseudosemiglober

G. texonus texonus

■ M beckmonni

TEXT-FIG. 2. Chart to show the ranges of selected conodont species in the Carboniferous Limestone
of the eastern Mendips. See text-figs. 3 to 10 for full information. Firm lines represent proven ranges,

broken lines are extrapolations.

Elictognathus laceratus (Branson and Mehl) ranges from Ma 7 to Ma 19 but appears
in only three samples.

Sample Ma 20 marks another abrupt change in the conodont faunas, siphono-
dellids and associated forms disappearing. The fauna in this uppermost part of the
Lower Limestone Shale Group includes Polygnathus communis communis Branson
and Mehl and Spathognathodus stobilis (Branson and Mehl), and also includes
poorly preserved specimens of Spathognathodus aculeatus (Branson and Mehl).

The Black Rock Group. The faunal assemblage seen in the uppermost part of the
Lower Limestone Shale Group continues into the Black Rock Group. However,
Pseudopolygnathus multistriatus Mehl and Thomas appears in sample WH 5, im-
mediately above a minor chert development at the base of unit br 2, and Sp. aculeatus

BUTLER: CARBONIFEROUS CONODONTS

481

is no longer seen. Gnathodus delicatm Branson and Mehl occurs in sample WH 12,
but is not seen again until sample HW 2, where it becomes abundant. Pseudo-
polygnathus primus Branson and Mehl occurs in unit br 2. Pelekysgnathus bultyncki
(Groessens) first appears in sample HW 3 and occurs along with Polygnathus com-
munis Carina Hass in samples HW 8 and HQ 1.

In the lower part of the Main Chert (unit br 3) at Ham Woods Quarry (samples
HW 18 and HW 18a) there is an occurrence of abundant representatives of Dollymae
bouckaerti Groessens. This genus was not recovered from any other section. Associ-
ated with D. bouckaerti in these samples are Pe. bultyncki, Ps. multistriatus, and
Gnathodus delicatus. In addition, a single specimen referable to Doliognathus sp.
was recovered from HW 18.

Near the upper limit of the range of P. communis carina, Gnathodus bulbosus
Thompson first appears (samples HW 20, HQ 12, and Va 2). G. bulbosus rapidly
takes over from G. delicatus as the dominant gnathodid. At this level in both
Ham Woods and Halecombe Quarries specimens referred to Bactrognathus cf. B.

2kg. Samples taken at 3 metre intervals,
except those from unit m2, which were
taken where limestone bands were
available.

25 Metres

Dork and light
grey pockstones
ond groinstones.

Lenticular pockstones
and groinstones, with
mud flasers

Channel and mego-
ripple shaped limestone
bodies within fissile
sil tstones

Fissile siltstones
with rare stringers
of skeletal debris
Limestones with phosphatic
nodules neor base
Lo m mat ed f me- g ra ined
limestones. with coorse
lenses, in fissile siltstones

TEXT-FIG. 3. Location of sampled section in Maesbury railway cutting (ST 606 475) and notes on lithologies.

482 PALAEONTOLOGY, VOLUME 16

MAESBURY Samples Ma

1

la

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

19;

20

21

22

23

24

25

26

27

28

F^eudopolygnathus dentilineatus
Polygnathus symmetricus
Fbtrognathus voriobilis
Spathognathodus crossidentatus
P communis communis

1

6

3

15

4

25

21

2

1

4

2

2

11

2

19

1

1

Elictognothus lacerotus
Siphonodello isosticha
Scf.S.isosticha
S obsolete

Polygnathus inornatus

2

3

5

7

5

1

1

4

3

1

1

3

1

1

1

11

68

3

40

3

10

3

11

11

2

Spathognathodus sto bills
Gnathodus punctatus
G. delicatus

Siphonodello cf. crenuloto
Spathognothodus oculeotus

2

5

4

1

2

3

2

1

3

Siphonodello cooper!
Polygnathus sp,

Gnathodus sp indet,
Siphonodello sp.indet.

2

1

2

8

2

3

3

6

1

3

21

2

Spathognathodus sp.indet.
BARS

1

2

3

3

2

2

1

6

2

1

1

1

8

6

4

9

6

6

39

8

28

7

14

5

5

5

85

7

13

2

30

1

2

1

TOTAL

-

4

27

8

42

28

8

134

13

98

18

36

6

-

-

13

23

-

1

173

18

4

18

3

40

-

1

2

1

-

TEXT-FIG. 4. Chart to show conodont occurrences in samples from Maesbury railway cutting. (Sample

prefix Ma.)

distortus Branson and Mehl occur (samples HW 21, 22, and HQ 14a). At Halecombe
these are followed by the first occurrence of Scaliognathus anchoralis Branson and
Mehl (in sample HQ 15). Pseudopoly gnat hus triangulus phmatus Voges first appears
in samples HW 23, HQ 16, and Va 1, and is an important constituent of the fauna
in the middle part of unit br 3. Pseudopoly gnathus triangulus triangulus Voges is
present in samples throughout the range of Ps. triangulus pinnatus, although speci-
mens possibly attributable to this subspecies also occur in samples HW 14 and
HQ 14a.

Ps. multistriatus dies out near the beginning of the range of Ps. triangulus pinnatus
(sample HQ 19). Near the top of the range of Ps. multistriatus. Polygnathus nodo-
marginatus Branson makes a brief appearance. Polygnathus cf. P. symmetricus
Branson first occurs here and ranges up into the top of unit br 3, where conodonts
become rare.

In the middle part of unit br 3 both at Halecombe and in Vallis Vale (samples
HQ 22 and Va 9) G. bulbosus disappears and Gnathodus texanus pseudosemiglaber
Thompson and Fellows first appears. Towards the top of unit br 3 (samples HQ 32
and Va 13) Gnathodus texanus texanus Roundy first appears and Ps. triangulus pin-
natus disappears. Scaliognathus anchoralis occurs in sample HQ 33, but is not seen
again until HQ 46, where it is present together with Hindeodella segaformis Bischoff.
A single specimen of Pelekysgnathus sp. A Voges was recovered from sample Va 13,
and additional specimens of H. segaformis were also found in this sample.

Associated with G. texanus texanus and P. cf. P. symmetricus in the upper part
of unit br 3 are common apatognathids and Spathognathodus scitulus (Hinde).

BUTLER: CARBONIFEROUS CONODONTS 483

Dark grey, muddy crinoidal packstones and wackestones

Dark and light grey packstones and grainstones.
Cross - bedding occasionally visible

Lenticular limestone bodies , with mud flasers.

r

Metres

-L 0

2 kg. Samples taken at intervals of approximately 3 Metres

TEXT-FIG. 5. Location of sampled sections in Windsor Hill and Ham Woods Quarries (ST 615 452) and

notes on lithologies.

484 PALAEONTOLOGY, VOLUME 16

WINDSOR HILL Qv. Samples WH

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

Polygnathus communis communis

1

5

2

1

6

7

1

2

3

4

2

Pseudopolygnothus multistriotus

1

2

2

1

2

4

1

5

13

1

1

Spothognothodus cross identotus

1

1

1

1

Sp. stobilis

1

1

1

Ps. primus

1

1

4

Gnothodus delicotus

1

Spothognothodus sp.indet.

1

1

1

2

2

2

3

3

BARS

3

5

3

25

1

5

4

4

10

1

24

3

1

11

17

20

7

3

2

TOTAL

3

6

-

4

32

2

5

10

9

19

3

40

4

-

1

12

26

44

16

6

3

TEXT-FIG. 6. Chart to show conodont occurrences in samples from Windsor Hill Quarry. (Sample prefix WH.)

HAM WOODS Qv. Samples HW

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

18a

19

20

21

22

23

Spothognothodus crossidentotus
Sp. stobilis

Pseudopolygnothus multistriotus
Ps. postinodosus
Gnothodus delicotus

1

2

1

3

6

1

1

2

2

2

1

1

3

2

18

5

1

10

4

3

6

2

2

1

5

8

1

2

3

17

8

3

7

9

6

6

116

31

10

1

5

Pelekysgnothus bultyncki
Polygnothus communis communis
P.communis corino
Gnothodus cf. delicotus
Ps.trionquius cf. trionqulus

1

2

12

10

6

1

1

20

2

8

1

3

4

4

20

45

9

1

4

1

2

5

6

8

11

12

7

68

1

2

1

1

1

1

Polygnathus cf. P symmetricus
Dollymoe bouckoerti
Doliognothus sp.

P nodomorginotus

Gnothodus n.sp. B, THOMPSON 1967

2

16

3

6

57

6

1

1

1

2

2

G. bulbosus

Boctrognothus cf. B.distortus
Ps. triongulus triongulus
Ps. triongulus pinnofus
Dollymoe sp.

19

11

28

33

1

1

1

23

1

Gnothodus sp.indet,
Pseudopolygnothus sp indet.

1

11

6

2

1

1

3

Spothognothodus sp.indet.
BARS

3

4

6

9

2

4

3

32

12

5

19

1

4

60

27

3

34

20

1

19

15

2

1

13

20

11

154

31

17

28

24

24

72

TOTAL

5

84

43

1

4

71

-

38

4

57

1

40

16

27

38

43

51

519

97

33

104

51

55

145

TEXT-FIG. 7. Chart to show conodont occurrences in samples from Ham Woods Quarry. (Sample prefix HW.)

BUTLER: CARBONIFEROUS CONODONTS

485

Light grey pelletal
grainstone , with
micritic intraclasts.

Light grey cnnoid -
foraminifer grainstone.

2 kg. Samples taken at intervals of approximately 3 metres.

TEXT-FIG. 8. Location of sampled sections in Halecombe Quarry (ST 702 475) and notes on lithologies.

HALECOMBE Qy. Samples H Q

1

2

3

5

6

7

8

9

10

11

12

13

14

14a

15

16

17

18

19

20

21

22

23

24

25

Pelekysgnathus bultyncki

2

Pseudopolygnathus multistriatus

2

1

1

1

Polygnathus communis communis

6

6

4

3

2

9

11

4

1

34

1

86

34

3

6

Pcommunis corina

6

13

4

1

5

Gnathodus dellcatus

1

1

1

6

6

2

7

1

G. cl delicatus

1

1

Spathognathodus stabllis

1

3

1

2

1

7

1

1

11

1

2

Sp. crassidentatus

1

1

4

5

1

Polygnathus cf.P. symmetric us

1

2

3

1

1

8

3

11

1

6

1

Gnathodus bulbosus

1

18

124

11

26

55

25

Gnathodus n.sp.B, THOMPSON 1967

1

1

2

1

5

Bactrognathus cf. B distortus

1

Ps.triangulus cf triangulus

1

Scaliognathus anchoralis

3

1

1

Ps, triangulus pinnatus

1

4

5

11

5

Polygnathus nodomarginatus

1

Ps.triangulus triangulus

1

G.texonus pseudosemiglaber

5

1

Gnathodus sp. indet.

1

1

1

5

10

17

3

1

Pseudopolygnathus sp. indet.

1

1

1

1

1

Spathognathodus sp. indet.

1

1

1

5

1

19

1

22

2

2

BARS

29

6

5

2

5

7

4

2

33

9

31

6

32

44

5

21

116

25

1

5

90

31

4

20

TOTAL

49

7

12

3

13

30

19

10

65

13

78

1

6

67

191

18

60

186

103

2

10

239

73

10

30

HALECOMBE Qy. Samples HQ

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

G.texonus pseudosemiglaber
Ps.triangulus pinnatus
Ps.triangulus triangulus
Spathognathodus stabilis
Sp. crassidentatus

1

4

2

1

1

3

1

1

3

1

5

2

1

1

2

1

1

1

1

1

G.texonus texanus
Polygnathus cl Psymmetricus
Scaliognathus anchoralis
Hindeodella segaformis
Spathognathodus scitulus

1

5

3

1

1

1

1

2

3

2

3

8

9

2

2

1

22

5

2

1

2

1

1

1

1

2

4

opatognathids

Gnathodus sp. indet.
Spathognathodus sp. indet.
BARS

5

1

1

2

5

1

2

1

1

3

1

4

2

7

36

24

3

17

4

3

30

6

8

14

2

2

77

8

1

3

17

6

3

TOTAL

4

7

50

31

4

19

5

8

3

49

20

12

2

19

3

3

109

15

3

6

25

15

7

-

-

HALECOMBE Qy. Samples HQ

52

54

56

58

60

62

64

66

68

70

72

74

75

76

78

80

82

Mestognathus beckmanni

Spathognathodus sp. indet.
BARS

2

1

1

3

TOTAL

1

1

5

-

-

-

TEXT-FIG. 9. Chart to show conodont occurrences in samples from Halecombe Quarry. (Sample prefix HQ.)

occurrences. (Sample prefix Va.)

r

o

o

D

O

■-t.

c/3

3

o

3

<

<

C/3

H

4:^

so

O

P

3

a

o

3*

P

O

3

O

a

o

3

NJ

ID

if)

fb

3

x

T>

in

3

1 TOTAL 1

1 BARS

Spathognathodus sp. indet.

Pseudopolygnathus sp indet.

Gnothodus sp indet.

Spathognathodus scitulus

apatognathids

Hindeodella segaformis

1 Pelekysgnathus sp.A,VOGES 1959

1 G.texanus texanus

G.texanus pseudosemiglaber

Polygnathus cf, P symmetricus

Bactrognathus sp

1 Gnothodus nsp.B, THOMPSON 1967

j Pseudopolygnathus multistriatus |

Gnothodus bulbosus

Polygnathus nodomarginatus

Ps. triangulus pinnatus

1 Sp.crassidentatus

1 Spathognathodus stabilis |

G. cf. delicatus

Gnothodus delicatus

P comm unis carina

1 Polygnathus communis communis I

1 VALLIS VALE Samples Va |

OJ

NJ

NO

CO

NO

-

ro

'sj

-

•vj

CD

NJ

CO

CO

CD

NJ

''O

-

NJ

o

03

NO

U1

4>

-

GO

'

JN

OD

CD

cn

GO

NO

-

-

CO

CO

OJ

CJ

cn

Fo

CD

-

NJ

-

CD

4>-

NJ

-

CD

U)

O

cn

cn

CO

00

-

-

G)

CD

CO

cn

OD

CO

CO

-

O

o

GO

(D

NO

NJ

-

NJ

'vj

CD

M

CD

CD

03

-

Fo

nJ

CD

NJ

--

NO

NO

cn

UD

cn

NJ

oo

00

NJ

cx>

NJ

NO

-

-

NO

-

C71

CO

-sj

NJ

io

-

cn

CO

CD

NJ

NO

NJ

00

NJ

OJ

cn

■'j

NJ

CO

cD

-

CD

UD

NJ

•o

CO

Co

OD

O

CO

-

NO

NO

cn

CD

00

1

NO

NJ

NO

O

488

PALAEONTOLOGY, VOLUME 16

Faunas from unit br 4 are particularly poor, only three samples producing any
conodonts at all, and samples were processed from 6-metre intervals after HQ 50
(see text-fig. 11). Sample HQ 76 (at the base of the Vallis Limestone) produced two
specimens of Mestognathus beckmanni Bischoff.

COMPARISONS AND CORRELATIONS WITH NORTH AMERICA

AND EUROPE

Conodonts recovered during the course of this study are all referable to the
Lower Carboniferous (Dinantian). The first definitive studies of conodonts of this
age were carried out in Germany by Bischoff (1957) and Voges (1959). These authors
erected a series of conodont zones for this part of the Carboniferous. In recent years
it has become clear that although the German sections offer good information on the
lowest Carboniferous and also on the anchoralis-Zonc and higher levels, there is a
paucity of information on the intermediate parts of the succession. The reason for
this is that the rocks which include the Siphonodella crenulata-Zone of Voges (1959),
that is the Liegende Alaunschiefer, have not produced good conodont faunas.

In Belgium, conodonts from that part of the succession characterized by the
genus Siphonodella are still little known, although here again both lower and higher
levels are better documented (for example, Conil et al. 1964; Conil et al. 1969;
Austin et al. 1970). Groessens (1971) has recently described faunas of anchoralis-
Zone age from Belgium. These latter faunas compare closely with some recovered
during the present study.

In North America Collinson et al. (1962) erected a series of conodont zones for
the Mississippian, based on evidence from the upper Mississippi Valley. These
zones have been recently revised by Collinson et al. 0971). Recent work by Thomp-
son (1967) and Thompson and Fellows (1970) has indicated that the upper Missis-
sippi Valley sections are incomplete (see Collinson et al. 1971, fig. 2). It is with their
more complete sequence of Mississippian rocks from south-western Missouri and
Arkansas that the Mendip faunas best compare.

Lowest part of the Lower Limestone Shale Group at Maesbury. The conodonts from
this part of the succession do not readily provide for detailed comparison with
information from either German or North American sections. Voges (1959) found
that Pseudopolygnathus dentilineatus has a range from the uppermost Devonian to
the base of the Siphonodella erenulata-Zone, and possibly into that zone. This
species has been reported from North America by several authors (for example
Branson 1934; Collinson et al. 1962; Klapper 1966; Canis 1968) and has a range
from basal Carboniferous, and possibly Upper Devonian, into the Siphonodella
quadruplicata-S. crenulata Zone of Collinson et al. (1962). Thompson and Fellows
(1970) reported Ps. dentilineatus ranging up into their Gnathodus delicatus-
Siphonodella cooperi cooperi Zone. The genus Patrognathus has been recorded from
the lowest Carboniferous of Belgium by Austin et al. (1970), but from an upper part
of the Kinderhookian of North America by Klapper (1971). Klapper (1971) re-
covered the genus from within a range of stratigraphy which also yielded Siphono-
della isosticha and S. eooperi, although siphonodellids and patrognathids were
found together in only one sample.

BUTLER: CARBONIFEROUS CONODONTS

489

It seems clear that the conodonts recovered from this part of the stratigraphy at
Maesbury do not, at present, offer any possibility of precise correlation with other
areas.

Faunas with Siphonodella. Siphonodellids first appear at Maesbury in beds which
include phosphatic nodules and which are interpreted as lag concentrates. The
species of Siphonodella present are advanced, in terms of the North American pro-
gression of forms (see Collinson et al. 1971), and their association with Gnathodus
punctatus enables comparison to be made with what Thompson and Fellows (1970)
regard as the uppermost Kinderhookian faunas. However, the ranges of Gnathodus
punctatus, G. delicatus, Siphonodella isosticha, S. obsoleta, and Elictognathus laceratus
are not shown to coincide on table 1 of Thompson and Fellows (1970). Their zonal
scheme is not consistent with their own recorded results. An examination of the
stratigraphic logs and conodont occurrence charts of these authors reveals that in
the Baird Mountain Quarry section (Thompson and Fellows 1970, p. 149) several
samples included this association of forms. Their samples 6 to 8 included Gnathodus
punctatus, G. delicatus, Siphonodella cooperi hassi ( = 5. isosticha, fide Klapper, see
systematic part), and Elictognathus laceratus, even though these forms do not have
corresponding ranges in their table 1. Since these Baird Mountain Quarry samples
are placed within the Siphonodella cooperi hassi-Gnathodus punctatus Zone, it would
seem reasonable to correlate samples Ma 7 to Ma 19 with a part of this zone.

This zone of Thompson and Fellows (1970) represents a part of the stratigraphy
which is not seen in the upper Mississippi Valley (Collinson et al. 1971, fig. 2). The
uppermost Kinderhookian of the upper Mississippi Valley was correlated with the
Belgian Tn 2b by Sando et al. (1969) on the basis of foraminiferids. Groessens (1971)
gives an upper limit to the range of Siphonodella at the top of Tn 2c in Belgium.
In Germany the range of the genus extends into the anchoralis-Zone (Voges 1959;
Meischner 1971), and a similar situation has been noted in south-west England
by Matthews (1969a, b). Matthews et al. (1972) report the occurrence of siphono-
dellids in a fauna including Gnathodus punctatus, Dollymae hassi, and Polygnathus
communis carina from Devon.

A tentative correlation is here suggested between samples Ma 7 to Ma 19 and
some part of the interval Tn 2b to Tn 2c in Belgium.

Uppermost Lower Limestone Shale Group and the lower part of the Black Rock Group.
The lowest part of this interval is characterized by the dominance of Pseudopoly-
gnathus multistriatus and Polygnathus communis communis. This association of
forms conforms to faunas lying immediately above the range of Siphonodella in
North America (for example Rexroad and Scott 1964) with the exception that there
they include abundant gnathodids. There appears to be a scarcity of gnathodids
throughout the Mendip sections at this time, these forms appearing in only two
samples at Maesbury. Two possibilities emerge here: first, the lack of gnathodids
at this level in the eastern Mendips could be due to some facies control ; or secondly,
this association of forms could be present in North America also, but has not been
recognized since the successions there are thinner at this level. Further study of this
problem is necessary. Whatever the cause of this absence, gnathodids reappear in
the middle part of the Black Rock Group.

E

490

PALAEONTOLOGY, VOLUME 16

The first occurrence of Polygnathus communis carina is taken by Thompson and
Fellows (1970) to define the lower limit of their Gnathodus semiglaber-P. communis
carina Zone. This form first appears in sample HW 8 and is present in sample HQ 1.
The first appearance of Pseudopolygnathus multistriatus is taken by Thompson and
Fellows (1970) to mark the base of the overlying Bactrognathus~Ps. multistriatus
Zone. This form appears earlier than P. communis carina in the Mendip successions
(in sample WH 5). There does not appear to be any distinction to be made in the
Mendip succession of conodonts between assemblages characteristic of the Gnathodus
semiglaber-P. communis carina Zone and those of the Bactrognathus-Ps. multi-
striatus Zone of Thompson and Fellows (1970). Nor is it clear where the equivalent
of the boundary between the G. semiglaber-P. communis carina Zone and the under-
lying G. punctatus-Siphonodella cooperi hassi Zone should be taken in the Mendips.

Pelekysgnathus bultyncki (Groessens) first appears in sample HW 3, but is present
with P. communis carina in samples HW 8 and HQ 1. Groessens (1971) shows two
maxima in the occurrence of P. communis carina, one of which coincides with the
first appearance of Pe. bultyncki. It would seem reasonable to correlate sample
HW 8 with a point near the beginning of the second maximum of P. communis
carina, that is with the base of Tn 3c of Groessens (1971). In Belgium Groessens has
shown that a level characterized by abundant Dollymae bouckaerti lies at the top
of the range of Pe. bultyncki. Above this level Doliognathus latus makes a brief
appearance and the range of Scaliognathus anchoralis begins. This progression of
forms compares with that recognized in the Mendip successions. It is possible there-
fore to correlate with a fair degree of certainty between the eastern Mendips and
Groessens’s Belgian successions at these levels.

Faunas comparable with the German anchoralis-Zonc. The anchoralis-Zone was
established in Germany by Bischoff (1957). Voges (1959, 1960) revised the zone and
selected Scaliognathus anchoralis, Hindeodella segaformis, and Doliognathus latus
as indices. Although Scaliognathus anchoralis was first described from North America
by Branson and Mehl (1941), the anchoralis-ZonQ has not been widely recognized
there. It is now clear that discontinuities within the Upper Mississippi Valley sections
are in part responsible for this.

The recognition of a line of development leading to Scaliognathus anchoralis
suggests that the first occurrence of this form at Halecombe Quarry must be at least
as early as any occurrence elsewhere. Correlation is possible with Groessens’s (1971)
sections and with the base of the anchoralis-Zone in Germany.

Again it is necessary to point out some differences in the record of conodont
occurrences between the eastern Mendip sections and those of Thompson and Fellows
(1970). According to these authors the ranges of Scaliognathus anchoralis and
Pseudopolygnathus triangulus pinnatus do not coincide with that of Gnathodus bul-
bosus. In the eastern Mendips, Ps. triangulus pinnatus ^ndj or Scaliognathus anchoralis
occur together with G. bulbosus in certain samples between HQ 15 and HQ 21 and
with G. texanus subspp. in certain samples within the range HQ 22 to HQ 46. For
the reasons given above, it seems unlikely that Scaliognathus would appear earlier
in Thompson and Fellows’s sections than it does in the eastern Mendips. The two
first occurrences are therefore tentatively correlated, suggesting an equivalence

BUTLER: CARBONIFEROUS CONODONTS

491

between sample HQ 15 and the base of the Bactrognathus distortus-Gnathodus
cuneifonnis Zone of Thompson and Fellows (1970).

The sequence of development from G. bulhosus to G. texanus pseudosemiglaber
and G. texanus texanus, recognized by Thompson and Fellows (1970, p. 89) can
also be seen in the Mendip faunas. The recognition of this sequence enables correla-
tion to be made between sample HQ 22 and the base of the Gnathodus texanus-
Taphrognathus Zone of Thompson and Fellows (1970), although it should be noted
that no Taphrognathus has been recovered from the eastern Mendips. Below this
zone in Thompson and Fellows’s scheme lies the Gnathodus bulbosus Zone, which
must fall within the interval HQ 15 to HQ 22 along with the B. distortus-G. cuneiformis
Zone. As already stated, there seems to be no real distinction in the Mendips be-
tween the G. semiglaber-P. communis carina Zone and the Bactrognathus- Ps. multi-
striatus Zone. In addition it does not seem to be possible to separate a B. distortus-
G. cuneiformis Zone from a G. bulbosus Zone here. It appears therefore that the four
zones suggested by Thompson and Fellows (1970) for this part of the stratigraphy
in Missouri and Arkansas resolve themselves into only two distinct divisions in
the Mendips.

Groessens (1971) reported Mestognathus beckmanni from the uppermost part of
Tn 3c and ranging up into the Visean. Two specimens of M. beckmanni were re-
covered during the present study from a single sample (HQ 76) near the base of the
Vallis Limestone. Voges (1959) reported M. beckmanni from the anchoralis-Zone,

W.GERMANY BELGIUM E. MENDIPS MISSOURI.USA

(Voges ,1959 ) (Groessens ,1971 ) (This paper) (Thompson and

V la

Vallis

Lst

HO 76

Fellows 1970 )

Unit

G. texanus -

br4

Taphrognathus

Zone

HQ_50

anchoralis -

Tn 3c

CL

D

O

O

Unit HQ

o

a:

br3 22

HQ 15 -

G.bulbosus 8. B.dist-
ortus -G.cun. Zones

JC S

M “S

Bactrognathus -
Ps.multistnatus &

Siphonodella

m z
o

HW 8

Gsemiglaber -
Pcomm carina Zones

crenulata -

Zone

Tn 3b

Unit

br2

Tn 3a

Mo27WH4

Unit brl

(b

O 0-

Mo20 ■
Unit m3

S-Cooperi hassi -

£ 2

—

G punctalus

Tn 2c

<_ .3;

Oi ro

Unit m2

Ma7 -

Zone

o 1/1

Unit ml

7

TEXT-FIG. 11. Chart showing suggested correlations between the eastern Mendip composite section and
sections in Europe and North America. Eastern Mendip column is drawn to scale, the others are not.

492

PALAEONTOLOGY, VOLUME 16

and it is therefore not possible to recognize an equivalent of zones higher than this
in the sections studied.

Text-fig. 11 sets out the suggestions made here on correlation with sections in
Europe and North America.

COMPARISONS AND CORRELATIONS WITH THE AVON GORGE

Rhodes et al. (1969) sampled the Lower Carboniferous succession in the Avon
Gorge and compared it with a composite section from the North Crop of the South
Wales Coalfield and with scattered sections in northern England and Scotland.
They erected a zonal scheme, and suggested that facies-control might explain the
absence of certain forms used as zonal indices in Germany and North America.
This work has been criticized by Ziegler (1971) on methodological and other grounds.
Among other points, Ziegler observes that it is difficult to extract from Rhodes et
al. (1969) information on the exact occurrences and ranges of species and on the
composition of faunas. This being the case, it will be understood that proposals
on correlation between the Mendips and the Avon Gorge must be tentative at the
present time. The proposals made here are set out in text-fig. 12, and discussed below.

TEXT-FIG. 1 2. Chart to show suggested correlations between the eastern
Mendips composite section and the section exposed in the Avon Gorge.
The two stratigraphic columns are drawn to the same scale.

AVON GORGE
CONODONT ZONES

(after RHODES ,

AUSTIN aORUCE, 1969 )

200 -n

100 -■

0— ^

# = S. cf. S.robustus - S. tridentatus Zone (one isolated sample).

* * = Placinatus - Ps.cf.Ps longiposticus Zone. B.R.D. = Black Rock Dolomite.

BUTLER: CARBONIFEROUS CONODONTS

493

Faunas from the lowest parts of the Lower Limestone Shale Group {Samples K 1 to
K 11 of Rhodes et al.; samples Ma 1 to Ma 6 of this paper). Faunas recovered from
this part of the succession by Rhodes et al. (1969) include Patrognathus, Pseudo-
polygnathus vogesi ( = Ps. dentilineatus, see systematic part), and also Clydagnathus
and clydagnathid-like spathognathodids. This association compares in some re-
spects with the Maesbury faunas, where Patrognathus was recovered from one
sample and Ps. dentilineatus occurs in several samples.

It is possible that the absence of the clydagnathid group of conodonts at Maesbury
may be explained by restriction of these conodonts to a narrow range of facies. They
occur both at Portishead and Clevedon (see Appendix) in facies different from that of
the Shirehampton Beds in the Avon Gorge, but it may be worth while to note that in
the North Crop, where stromatolitic and oolitic rocks predominate (George 1 954), this
group of conodonts extends into the Z zone (Black Rock Group) (Rhodes et al. 1969).

The Palate Bed in the Avon Gorge (which lies between samples K 11 and K 12
of Rhodes et al. 1969) contains a large conodont fauna, including abraded speci-
mens (see Appendix). Siphonodella duplieata appears in this bed, but has not been
reported from elsewhere in the Bristol district. A discontinuity is present below the
Palate Bed in the Avon Gorge. Caliche formation occurred within the underlying
Bryozoa Bed, and pebbles of Bryozoa Bed material occur within the Palate Bed.
The possibility cannot therefore be ruled out that the lower part of the Maesbury
section (unit m 1) might include faunas which have no representation in the dis-
continuous Avon Gorge stratigraphy.

Faunas with Siphonodella {Samples K 12 and K 17 of Rhodes et al.; Samples Ma 7
to Ma 19a of this paper). Rhodes et al. (1969) recovered two specimens of Siphono-
della from the Avon Gorge. Both were broken, but were identified as advanced
forms on the basis of the rostral ridge development (Rhodes et al. 1969, p. 55).
Patrognathids occur with these siphonodellids in the Avon Gorge. Both at Maes-
bury and in the Avon Gorge the incoming of siphonodellids is associated with a
phosphatic lag deposit. At this level a comparison is therefore possible both in
faunal and sedimentological terms.

The presence of patrognathids along with the siphonodellids in the Avon Gorge
section could be due entirely to reworking associated with the Palate Bed, which
(see Appendix) certainly includes reworked patrognathids. It is also possible that
facies differences play some part in this. This problem may become clearer when
the full distribution of Patrognathus is known.

It should be noted that in a later paper, Austin et al. (1971, text-fig. 2) extend the
range of Siphonodella below the Bryozoa Bed. There is no justification for this on
the basis of any published information. In the Avon Gorge a covered interval is
present above sample K 17 of Rhodes et al. (1969). In their paper these authors give
this interval as 25 ft (7-5 m) (Rhodes et al. 1969, fig. 59). This gap has been recog-
nized by other workers and is usually given a value of some 200 ft (60 m). Thus,
Kellaway (1971) gives the total thickness of the Lower Limestone Shale Group in
the Avon Gorge as 350 ft (107 m) compared with the 175 ft (53-5 m) given by Rhodes
et al. (1969). This covered interval may, in part, explain the rarity of siphonodellids
in the Avon Gorge reported by Rhodes et al. (1969).

494

PALAEONTOLOGY, VOLUME 16

Faunas from the uppermost part of the Lower Limestone Shale Group and those from
the lower part of the Black Rock Group {Samples K2I to Z 37 of Rhodes et al. ; Samples
Ma 20 to Ma 28; WH 1 to WH 21 ; HW 1 to HW22; HQ 1 to HQ 14a; Va 1 to Va 2
of this paper). Faunas recovered from this part of the succession by Rhodes et al. ( 1 969)
compare well with those from the Mendips. In the lower part of this interval the faunas
are dominated by Pseudopolygnathus multistriatus (see systematic description for
synonymy) and Polygnathus communis communis. Gnathodus delicatus appears some
way above the base of the Black Rock Limestone (Z zone). It should be noted here
that Rhodes et al. (1969) did not in fact recover Gnathodus delicatus from their
Spathognathodus costatus costatus-Gnathodus delicatus Zone in the Avon Gorge.
The first occurrence of G. delicatus there is given by Rhodes et al. (1969, p. 97) as
sample Z 28, and this may tentatively be correlated with the point where gnathodids
become abundant in the Mendips, that is sample HW 2 at Ham Woods Quarry.

The lower chert horizon noted in the Mendips is represented in the Avon Gorge
(sample Z 13 of Rhodes et al. 1969, fig. 60). However, the Main Chert is not present
there, nor are any of the conodonts found at this level in the Mendip succession.
Polygnathus communis carina was not reported from the Avon Gorge and no cono-
donts of the kind of Pelekysgnathus bultyncki or Dollymae bouckaerti were recorded.
Since P. communis carina first appears in sample HW 8 in the Mendips, it seems
likely that the interval Z 28 (the first appearance of G. delicatus) to Z 37 (first appear-
ance of G. bulbosus in Z 38, see below) in the Avon Gorge lies below the level of
HW 8 in the Mendips, but above that of HW 2.

An equivalence appears therefore to have been established between the upper-
most part of the Black Rock Limestone (Z zone) in the Avon Gorge and a level
less than half-way up this division in the Mendips. It seems that a considerable
discontinuity may exist at the top of the Black Rock Limestone in the Avon Gorge,
as has already been suggested by Mitchell (1971, 1972) on the basis of coral faunas.

Faunas comparable with the German anchoralis-Zonc {Samples Z 38 to C 14 of
Rhodes et al. ; Samples HW 23 ; HQ 15 to HQ 82; Va 3 to Va 20 of this paper). Rhodes
et al. (1969) did not record any trace of anchor alis-Zont faunas (with the exception
of one specimen of Pseudopolygnathus triangulus cf. pinnatus, which is removed from
this species altogether in the systematic part of this paper, see below). However, an
examination of their faunas from the Black Rock Dolomite in the Avon Gorge has
revealed that certain gnathodids characteristic of the anchoralis-Zone equivalents
in the Mendips do in fact occur here. Thus, G. bulbosus occurs in sample Z 38 and
G. texanus pseudosemiglaber occurs in sample C 4. S. C. Matthews (pers. comm.)
has recovered Ps. triangulus pinnatus from the Black Rock Dolomite in the Avon
Gorge.

It appears therefore that although the Mendip succession is much the thicker at
this level, and the Avon Gorge succession broken, there might nevertheless be some
isolated remnants of the fuller succession of faunas to be found in the Black Rock
Dolomite at Bristol. All of this question will be clearer when the conodont faunas
of the Black Rock Dolomite have been restudied.

Sparse faunas with Mestognathus beckmanni were recorded from the Gully
Oolite of the Clifton Down Group in the Avon Gorge by Rhodes et al. (1969), and

BUTLER; CARBONIFEROUS CONODONTS

495

they may perhaps be compared with those recovered from sample HQ 76 in the
Mendips. This would indicate that the base of the Gully Oolite in the Avon Gorge
is no lower than the top of the Black Rock Group in the eastern Mendips, which
is consistent with the proposals advanced by Kellaway and Welch (1955). Mitchell
(1972) has recently suggested that the Tournaisian-Visean boundary should lie at,
or below, the base of the Gully Oolite in the Avon Gorge. The conodont faunas
recovered from the eastern Mendips during this study do not contradict this sug-
gestion when compared with those recovered from Belgium by Groessens (1971).

SYSTEMATIC PALAEONTOLOGY

The entire conodont collection has been deposited in the Leeds Office of the
Institute of Geological Sciences. Four-figure numbers prefixed LZA identity 32-
cavity slides in the collection. Individual cavities are identified by a suffix to the
four-figure number. Each figured specimen occupies its own cavity. Bar-type cono-
donts are included in the collection, but since they have in almost every case no
stratigraphic significance, they are not treated in the systematics. The sole exception
is Hindeodella segaformis.

The synonymy lists carry annotations according to the system proposed by
Richter (1948).

Genus bactrognathus Branson and Mehl 1941
Bactrognathus cf. B. distortus Branson and Mehl 1941

Plate 58, figs. 11-13

Remarks. This form possesses a posterior lateral process, suggesting comparison
with the genus Bactrognathus. The prominent hornlike denticle is similar to that
described for B. distortus Branson and Mehl by Rexroad and Scott (1964). These
authors figure a lateral view of B. distortus which is similar to that of the present
specimens (cf. Rexroad and Scott 1964, pi. 3, fig. 9). The distal part of the posterior
lateral process of B. distortus is, however, normally redirected towards the posterior,
giving a Z-shaped appearance in oral view.

In lateral view the main bar resembles that of Pelekysgnathus bultyncki (Groes-
sens) and in oral view the species has much in common with immature forms of
Scaliognathus anchoralis Branson and Mehl, which, however, have two posterior
lateral processes (cf. PI. 58, figs. 12, 13, and figs. 21, 22).

Material. 3 specimens, from 3 samples.

Genus doliognathus Branson and Mehl 1941
Doliognathus sp.

Plate 58, figs. 17, 18

Remarks. A single specimen referable to this genus was recovered in association
with Pelekysgnathus bultyncki (Groessens) and D. bouckaerti Groessens. It re-
sembles early forms of Pe. bultyncki, displaying a short posterior bar behind the
prominent hornlike denticle, but has a small node on the inner side of this denticle.

496

PALAEONTOLOGY, VOLUME 16

The basal cavity has a corresponding lobe underneath this lateral node. Although
this specimen is an immature form, it is considered to resemble the genus Dolio-
gnathus Branson and Mehl. The lateral view of a specimen assigned to D. latus
Branson and Mehl by Groessens (1971, pi. 2, fig. 1) is similar in appearance.

Material. 1 specimen.

Genus dollymae Hass 1959
Dollymae bouckaerti Groessens 1971

Plate 58, figs. 19, 20, 25, 26

1959 Dollymae sp. B Voges, p. 276, pi. 33, figs. 15-17.

*1971 Dollymae bouckaerti Groessens, p. 14, pi. 1, figs. 6-8.

Remarks. Specimens found during the present study demonstrate similar morpho-
logical variations to those described by Groessens (1971). The diagnosis given by
Groessens (1971, p. 14) runs as follows: ‘A species of the genus Dollymae with a
straight blade and two lateral processes, each ornamented by a single node or row
of nodes. The basal cavity occupies the whole of the aboral surface of the platform
and includes a median groove.’.

In agreement with Groessens (1971, p. 15) this species is considered to have been
derived from his own Spathognatliodus bultyncki (referred to Pelekysgnathus below)
by an expansion of the basal cavity and the development of nodes on the resulting
lateral processes.

Material. 63 specimens, from 2 samples.

Dollymae sp.

Remarks. A single specimen referable to the genus Dollymae Hass was recovered
from sample HW 20. This specimen was broken, only part of the platform and
lateral process from one side being found. The specimen shows the development of
a broad basal cavity, with a platform ornamented by a laterally directed arc-shaped
row of denticles. The platform ornamentation does not fall within the range of
variation of that of D. bouckaerti Groessens.

Material. 1 specimen.

Genus elictognathus Cooper 1939
Elictognathus laceratus (Branson and Mehl 1934)

Plate 59, figs. 25, 28

*1934 Solenognathus lacerata Branson and Mehl, p. 271, pi. 22, figs. 5, 6.

1970 Elictognathus laceratus (Branson and Mehl); Thompson and Fellows, p. 81, pi. 5, figs.
20, 21 (with synonymy).

Remarks. Klapper (1966) and Thompson and Fellows (1970) have discussed the
range of variation in ornament exhibited by this species. Specimens recovered from
Maesbury are simple forms, similar to Branson and Mehl’s holotype.

Material. 10 specimens, from 3 samples.

BUTLER: CARBONIFEROUS CONODONTS

497

Genus gnathodus Pander 1856

Remarks. In recent years much confusion has arisen in the classification of gnatho-
dids, particularly in that of G. texanus Roundy. It is now becoming clear that part
of the reason for this is that the record of gnathodids from the upper Mississippi
Valley is not complete (compare Rexroad and Scott 1964 and Thompson and Fellows
1970). In addition, the stratigraphy immediately below the anchoralis-Zone in
Germany has not produced many conodonts. It now appears that the gnathodids
referred to as G. antetexanus Rexroad and Scott lie below the anchoralis-Zone and
that the German anchoralis-Zone gnathodids (including the G. texanus group of
Voges 1959) are not represented in the upper Mississippi Valley sections.

In the eastern Mendips, G. delicatus Branson and Mehl dominates faunas imme-
diately below the anchoralis-Zone and faunas in the lower part of this zone are
dominated by G. bulbosus Thompson.

Gnathodus bulbosus Thompson 1967

Plate 56, figs. 12, 15, 16, 19-24, 27

*1967 Gnathodus bulbosus Thompson, p. 37, pi. 3, figs. 7, 1 1, 14, 15, 18-21 ; pi. 6, figs. 2, 7.
vl969 Gnathodus punctatus (Cooper); Rhodes, Austin, and Druce, p. 105, pi. 18, figs. 1, 10, 11.
vl969 Gnathodus punctatus-Gnathodus semiglaber transition; Rhodes, Austin, and Druce, pi. 30,
figs. 2, 8.

1970 Gnathodus bulbosus Thompson; Thompson and Fellows, p. 84, pi. 1, figs. 3, 6, 8, 9, 12, 13.

Remarks. Thompson (1967, p. 37) restricted this species to those forms whose carina
is bulbous where it protrudes beyond the posterior end of the platform. Specimens
he illustrated are characterized by high peg-like nodes at the anterior end of the
platform, appearing almost to ‘pinch’ the carina. In the present study, many speci-
mens possess further nodes lying in rows parallel to the carina in the posterior part
of the platform. These are considered to lie within the range of variation of G.
bulbosus.

As suggested by Thompson and Fellows (1970, p. 89) G. texanus Roundy appears
to have developed from G. bulbosus by a reduction of the platform and the expan-
sion of the peg-like inner node into a short parapet.

Material. 417 specimens, from 15 samples.

Gnathodus delicatus Branson and Mehl 1938

Plate 56, figs. 3-5, 7-11, 13, 14

*1938 Gnathodus delicatus Branson and Mehl, pi. 34, p. 145, figs. 25-37.

1967 Gnathodus sp. cf. bilineatus (Roundy); Thompson, p. 37, pi. 3, figs. 8, 10, 12, 17.
vl969 Gnathodus semiglaber Bischoff; Rhodes, Austin, and Druce, p. 106, pi. 30, fig. 1.

1970 Gnathodus sp. cf. bilineatus (Roundy); Thompson and Fellows, p. 84, pi. 1, figs. 5, 10.
vl972 Gnathodus delicatus Branson and Mehl; Matthews, Sadler, and Selwood, pp. 559-560,
pi. 110, figs. 5, 7-9 (with synonymy).

Remarks. This species is considered to include all those gnathodids which possess
platforms ornamented by rows of nodes sub-parallel to the carina, but without the
low broad outer platform characteristic of G. bilineatus (Roundy). Forms towards

498

PALAEONTOLOGY, VOLUME 16

the top of the range of the species show the development of a distinct parapet,
situated anteriorly on the inner side, the rows of nodes posterior to it being less well
developed. These forms resemble those called G. antetexanus Rexroad and Scott
and G. typicus Cooper but are here considered to lie within the range of variation
of G. delicatus. A transition exists between G. delicatus and forms with a bulbous
platform on the inner side, referable to G. semiglaber Bischolf. Matthews et al.
0972) have drawn attention to forms of G. delicatus which resemble G. punctatus
(Cooper). The specimen illustrated here in Plate 56, figs. 10, 11 is of this type.

Material. 275 specimens, including 8 cf. determinations, from 27 samples.

Gnathodus punctatus (Cooper 1939)

Plate 56, figs. 1, 2

*1939 Dryphenotus punctatus Cooper, p. 386, pi. 41, figs. 42, 43; pi. 42, figs. 10, 11.
vl972 Gnathodus punctatus (Cooper); Matthews, Sadler, and Selwood, pp. 560-562, pi. 109, figs.
5, 13; pi. 110, figs. 1-4, 11-15 (with synonymy).

Remarks. Matthews et al. (1972) have described several variants of this species.
The two specimens recovered during the present study both lie within variant 5 of
these authors. A specimen illustrated by Rhodes et al. (1969) as G. delicatus is con-
sidered here to lie within the range of variation of G. punctatus (variant 1 of Matthews
et al. 1972, and cf. Plate 56, figs. 10, 11 of this paper). Specimens referred to G.
punctatus by Rhodes et al. (1969) are placed in G. bulbosus (see above). G. punctatus
has a curved inner parapet, convex to the carina, and is therefore distinguished from
G. bulbosus, which has a peg-like node on the inner side.

Material. 2 specimens, from 1 sample.

EXPLANATION OF PLATE 56
Specimens dusted with ammonium chloride. All x 30.

Figs. 1, 2. Gnathodus punctatus (Cooper). Aboral and oral views of LZA 6008/1 (sample Ma 7).

Figs. 3-5, 7-11, 13, 14. Gnathodus delicatus Branson and Mehl. 3, 4, Aboral and oral views of LZA 6060/1
(HW 10). 5, LZA 6010/1 (Ma 9). 7, 8, Oral and aboral views of LZA 6086/1 (HQ 10). 9, LZA 6080/1
(HQ 3). 10, 11, Aboral and oral views of LZA 6063/1 (HW 13). 13, 14, Aboral and oral views of
LZA 6068/10 (HW 18).

Figs. 6, 25, 26. Gnathodus n. sp. B Thompson, 1967. 6, LZA 6011 jA (HW 23). 25, 26, Lateral and oral
views of LZA 6093/4 (HQ 15).

Figs. 12, 15, 16, 19-24, 27. Gnathodus bulbosus Thompson. 12, LZA 6093/6 (HQ 15): specimen transi-
tional from G. delicatus. 15, 16, Oral and aboral views of LZA 6093/1 (HQ 15). 19, LZA 6093/9
(HQ 15). 20, 21, Aboral and oral views of LZA 6100/1 (HQ 19). 22, LZA 6093/2 (HQ 15). 23, LZA
6093/5 (HQ 15). 24, LZA 6093/3 (HQ 15). 27, LZA 6098/1 (HQ 18).

Figs. 17, 18. Gnathodus sp. (juv?). 17, LZA 6093/7 (HQ 15). 18, LZA 6093/8 (HQ 15).

Figs. 28, 29, 36. Gnathodus texanus pseudosemiglaber Thompson and Fellows. 28, 29, Oral and aboral
views of LZA 6112/3 (HQ 29). 36, LZA 6150/1 ( Va 1 3).

Figs. 30-35. Gnathodus texanus texanus Roundy. 30, LZA 6118/5 (HQ 35). 31-33, Lateral, oral, and
aboral views of LZA 61 18/1 (HQ 35). 34, 35, Oral and aboral views of LZA 6155/1 (Va 17).

PLATE 56

BUTLER, Carboniferous conodonts

500

PALAEONTOLOGY, VOLUME 16

Gnathodus texanus Roundy 1926

Remarks. Thompson and Fellows (1970) have recognized two subspecies of G.
texanus Roundy. Both have been recovered from the eastern Mendips.

Gnathodus texanus pseudosemiglaber Thompson and Fellows 1970

Plate 56, figs. 28, 29, 36

71967 Gnathodus texanus Roundy; Wirth, p. 213, pi. 23, fig. 18 (only).

V . 1969 Gnathodus antetexanus Rexroad and Scott; Rhodes, Austin, and Druce, pi. 18, fig. 13 (only).
*1970 Gnathodus texanus pseudosemiglaber Thompson and Fellows, p. 88, pi. 2, figs. 6, 8, 9, 11-13
(with synonymy).

71970 Gnathodus typicus Cooper; Marks and Wensink, p. 264, pi. 4, figs. 1, 4 (only).

Remarks. Thompson and Fellows (1970, p. 89) suggest that this form arises from
G. bulbosus and in turn gives rise to G. texanus texanus Roundy. This view is sup-
ported by evidence from the eastern Mendips. Specimens recovered are similar to
those illustrated by Thompson and Fellows (1970).

Material. 62 specimens, from 1 1 samples.

Gnathodus texanus texanus Roundy 1926

Plate 56, figs. 30-35

*1926 Gnathodus texanus Roundy, in Roundy, Girty, and Goldman, p. 12, pi. 2, figs. 7, 8.

1970 Gnathodus texanus texanus Roundy; Thompson and Fellows, p. 89, pi. 2, figs. 15, 16 (with
synonymy).

71970 Gnathodus typicus Cooper; Marks and Wensink, p. 264, pi. 4, figs. 2, 3 (only).

Remarks. Specimens recovered from the eastern Mendips conform to the descrip-
tion given by Thompson and Fellows (1970, p. 89). As these authors suggest, the
subspecies seems to develop from G. texanus pseudosemiglaber by a reduction of
the outer platform.

Material. 31 specimens, from 12 samples.

Gnathodus n. sp. B Thompson 1967

Plate 56, figs. 6, 25, 26

1967 Gnathodus n. sp. B Thompson, p. 43, pi. 4, figs. 14.

Remarks. This species is characterized by a platform ornamented by scattered low
nodes or rows of nodes. It may deserve to be included within the range of variation
of G. delicatus.

Material. 16 specimens, from 9 samples.

Genus hindeodella Ulrich and Bassler 1926
Hindeodella segaformis Bischoff 1957

Plate 58, fig. 29

*1957 Hindeodella segaformis Bischoff, p. 28, pi. 5, figs. 40, 41, 43.

1959 Hindeodella segaformis Bischoff; Voges, p. 285.

BUTLER: CARBONIFEROUS CONODONTS

50]

1964 Hindeodella segaformis Bischoff; Burton, range chart, facing p. 74.

1965 Hindeodella segaformis Bischoff; Budinger, p. 66, pi. 5, figs. 19-21 (with synonymy).

1967 Hindeodella segaformis Bischoff; Zikmundova, pi. 2, fig. 3; pi. 3, figs. 2a, 2b.

1969a Hindeodella segaformis Bischoff; Matthews, pi. 47, figs. 10, 11.

1970 Hindeodella segaformis Bischoff; Marks and Wensink, p. 265, pi. 1, fig. 2.

Material. 3 specimens, from 2 samples.

Genus mestognathus Bischoff 1957
Mestognathus beckmanni Bischoff 1957

Plate 58, figs. 1, 2

*1957 Mestognathus beckmanni Bischoff, p. 37, pi. 2, figs. 4, 5, 6, 8, 9.

1960 Mestognathus beckmanni Bischoff; Ziegler in Kronberg, Pilger, Scherp, and Ziegler, p. 14,
pi. 3, fig. 1.

vl969 Mestognathus beckmanni Bischoff; Rhodes, Austin, and Druce, p. 150, pi. 15, fig. 7.

1971 Mestognathus beckmanni Bischoff; Groessens, pi. 2, fig. 8.

Material. 2 specimens, from 1 sample.

Genus patrognathus Rhodes, Austin, and Druce 1969
Patrognathus variabilis Rhodes, Austin, and Druce 1969

Plate 59, figs. 1, 2

v*1969 Patrognathus variabilis Rhodes, Austin, and Druce, p. 179, pi. 2, figs. 8-11.

1970 Patrognathus variabilis Rhodes, Austin, and Druce; Austin, Conil, Rhodes, and Streel, pi. 1,
fig. 7.

Remarks. Klapper (1971) has distinguished two species of Patrognathus, based on
the size of the basal cavity. The two specimens recovered from the Maesbury sec-
tion have widely-flared basal cavities and can therefore be included in P. variabilis.

Material. 2 specimens, from 1 sample.

Genus pelekysgnathus Thomas 1949

Remarks. This genus was originally described from Upper Devonian rocks. Voges
(1959) reported its occurrence in rocks of anchoralis-Zone age, and further records
from Carboniferous rocks of this age have come from New Mexico (Burton 1964)
and Belgium (Groessens 1971). Klapper (1966) has discussed the difficulty of dis-
tinguishing Pelekysgnathus from Icriodus in the Upper Devonian. Forms recovered
from the eastern Mendips are here referred to as Pelekysgnathus, following the lead
of Voges (1959). The relationship of these to any Devonian forms, whether Pelekys-
gnathus or Icriodus, is not understood.

Pelekysgnathus bultyncki (Groessens 1971)

Plate 58, figs. 8-10, 14-16

*1971 Spathognathodus bultyncki Groessens, p. 115, pi. 1, figs. 2-5.

Remarks. Groessens’s diagnosis (1971, p. 15) runs as follows: ‘A species of Spatho-
gnathodus with a straight, or slightly arched, blade. The denticles are upright in the
anterior part, and progressively show more posteriorward inclination towards the

502

PALAEONTOLOGY, VOLUME 16

posterior end of the blade. A prominent posteriorly pointing denticle is present in
the posterior part of the blade. The basal cavity is situated at the posterior end of
the blade and is rounded posteriorly and pointed anteriorly.’

It is considered here that the features described by Groessens are more consistent
with the genus Pelekysgnathus than with Spathognathodus.

Early forms possess denticles behind the main cusp, but these are not present in
later specimens (cf. Groessens 1971, pi. 1, figs. 2-4). The origin of the species is
not clear.

Pe. bultyncki appears to have formed the stock from which both Dollymae bou-
ckaerti and Scaliognathus developed. Expansion of the basal cavity and the develop-
ment of nodes on the oral surface of the expansion seem to have led to D. bouckaerti.
A range of forms showing this transition is available in sample HW 18 (slide
LZA 6070). Scaliognathus anchoralis appears to have been derived by the develop-
ment of lateral processes, the basal cavity remaining relatively restricted. Bactro-
gnathus cf. B. distortus is considered to be a transitional form, with only one lateral
process. In addition, the specimen of Doliognathus sp. (PI. 58, figs. 17, 18) is close
to Pe. bultyncki but has a single rudimentary lateral bar.

Material. 17 specimens, from 4 samples.

Pelekysgnathus sp. A Voges 1959

Plate 58, figs. 3-5

1959 Pelekysgnathus sp. A Voges, p. 287, pi. 33, fig. 44.

1971 1 Pelekysgnathus sp. A Voges; Groessens, pi. 2, fig. 3.

Remarks. Klapper (1966) has stated that this form, because of its double row of
nodes, should be placed in the genus Icriodus. It has been retained in Pelekysgnathus
for the present, however, until its relationship to other forms is understood. It appears
to be related to Pe. bultyncki. In lateral view the two species are almost identical
(see PI. 58).

Material. 1 specimen.

Genus polygnathus Hinde 1879
Polygnathus communis Branson and Mehl 1934

Remarks. Hass (1959) subdivided P. communis on the basis of platform ornament.
In the eastern Mendips the zonally important form P. communis carina Hass has
been recovered, in addition to the long-ranging P. communis communis Branson
and Mehl.

Immature specimens of this species have relatively large basal cavities and larger
forms have relatively small cavities. The actual size of the basal cavity appears to
remain constant throughout growth.

Polygnathus communis carina Hass 1959

Plate 59, figs. 10-13, 26

*1959 Polygnathus communis var. carina Hass, p. 391, pi. 47, figs. 8, 9.

BUTLER: CARBONIFEROUS CONODONTS

503

1969 Polygnathus communis dentatus Druce, p. 95, pi. 18, figs. 13, 14.
vl972 Polygnathus communis carina Hass; Matthews, Sadler, and Selwood, pp. 563-564, pi. Ill,
figs. 6, 7, 13 (with synonymy).

Remarks. Variants of P. communis showing the development of nodes or ridges on
the anterior part of the platform or platform margin are placed within this subspecies.

Material. 162 specimens, from 16 samples.

Polygnathus communis communis Branson and Mehl 1934

Plate 59, figs. 8, 9, 15-17

*1934 Polygnathus communis Branson and Mehl, p. 293, pi. 24, figs. 1-4.

Remarks. Some members of this subspecies have exceptionally long blades and
abbreviated platforms. In these specimens the basal cavity often lies on the aboral
surface in a position anterior to the blade-platform junction. These forms resemble
P. varcus Stauffer.

Material. 455 specimens, from 55 samples.

Polygnathus inornatus Branson 1934

Plate 59, figs. 6, 7, 19, 20

*1934 Polygnathus inornata Branson, p. 309, pi. 25, figs. 8, 26.

1971 Polygnathus inornatus Branson; Klapper, p. 6 (with synonymy).

Remarks. Klapper (1971, p. 7) discussed the distinction between this species and
P. inornatus sensu Branson and Mehl 1934. All specimens recovered from the Men-
dips lie within the range of variation of P. inornatus Branson, including forms which
might previously have been called P. lobatus Branson and Mehl. The species is
distinguished from P. symnietricus Branson by its well-defined subcircular basal
cavity. In juvenile forms, the basal cavity resembles that of P. communis. A short
free blade is common in this species.

Material. 151 specimens, from 9 samples.

The Polygnathus symmetricus Branson Group

Remarks. A number of polygnathids figured by other workers resemble P. inornatus
in their platform development but differ in that they possess an elongate basal
cavity. Klapper (1966) placed specimens with this type of cavity in P. symmetricus
Branson, and his lead is followed here.

Within this group of polygnathids are recognized two subgroups. The first has
been recovered from the lowest part of the Lower Limestone Shale Group seen at
Maesbury and has been found in the Shirehampton Beds at Stoke Bishop (Bristol)
by S. C. Matthews (pers. comm.). This subgroup closely resembles the specimens
of P. symmetricus recovered from the Hannibal Shale by Branson and Mehl (1934)
and is included within that species.

The second subgroup includes a broader range of polygnathids found higher up
the stratigraphy, around the equivalent of the base of the anchor alis-Zont. These

504

PALAEONTOLOGY, VOLUME 16

were reported from Missouri by Thompson (1967) and Thompson and Fellows
(1970). Some representatives are also available in the Avon Gorge (see Rhodes
et al. 1969). These forms are included here as P. cf. P. symmetricus, since they have
the characteristic basal cavity but are generally asymmetrical. This subgroup appears
to develop from Pseudopoly gnathus multistriatus by reduction of the size of the
basal cavity and a change in the platform ornament. Polygnathus nodomarginatus
Branson is considered to be a transitional form. Members of this subgroup show
characteristic pseudopolygnathid asymmetry in juveniles, the platform on the right-
hand side tending to be better developed than that on the left.

Juvenile forms of both subgroups have basal cavities which are deep and widely
excavated along the underside of the platform, resembling those of spathognathodids.

Polygnathus nodomarginatus Branson 1934

Plate 57, figs. 2, 3

*1934 Polygnathus nodomarginata Branson, p. 310, pi. 25, fig. 10.

1956 Polygnathus nodomarginata Branson; Bischoff and Ziegler, p. 156, pi. 12, fig. 6.

1956 Polygnathus inornata Branson; Bischoff and Ziegler, p. 157, pi. 12, fig. 5.
vl969 Pseudopolygnathus nodomarginatus (Branson); Rhodes, Austin, and Druce, p. 212, pi. 9,
figs. 1-4; pi. 12, figs. 6-8, 10.

V. 1969 Pseudopolygnathus cf. longiposticus Branson and Mehl; Rhodes, Austin, and Druce, p. 210,
pi. 30, figs. 11-16 (only).

V. 1969 Pseudopolygnathus triangulus cf. pinnatus Voges; Rhodes, Austin, and Druce, p. 216, pi. 30,
fig. 19.

Remarks. This species is considered to be closely related to Pseudopolygnathus
multistriatus Mehl and Thomas. The form appears briefly near the top of the range
of Ps. multistriatus and some specimens have basal cavities similar to that of this
pseudopolygnathid. The platform ornament is coarse, but polygnathid-like. Members
of this species show a characteristic asymmetry, the right-hand side of the platform
often extending farther anteriorly than the left.

Material. 3 specimens, from 3 samples.

EXPLANATION OF PLATE 57
Specimens dusted with ammonium chloride. All x 30.

Fig. 1. Pseudopolygnathus postinodosus Rhodes, Austin, and Druce. LZA 6052/1 (sample HW 2).

Figs. 2, 3. Polygnathus nodomarginatus Branson. Oral and aboral views of LZA 6093/12 (HQ 15).

Figs. 4, 5, 7, 8, 11, 12, 19, 20. Polygnathus cf P. symmetricus Branson. 4, 5, Aboral and oral views of
LZA 6098/1 1 (HQ 18). 7, 8, Aboral and oral views of LZA 61 19/1 (HQ 36). 11,12, Aboral and oral
views of LZA 6118/6 (HQ 35). 19, 20, Aboral and oral views of LZA 6077/3 (HW 23).

Figs. 6, 9, 10, 15, 16, 23-25. Pseudopolygnathus multistriatus Mehl and Thomas. 6, LZA 6046/9 (WH 17).
9, 10, Aboral and oral views of LZA 6100/2 (HQ 19). 15, 16, Aboral and oral views of LZA 6078/1
(HQ 1). 23, 24, 25, Oral, aboral, and lateral views of LZA 6047/2 (WH 18).

Figs. 13, 14, 17, 18. Pseudopolygnathus primus Branson and Mehl. 13, 14, Aboral and oral views of LZA
6047/1 (WH 18). 17, 18, Aboral and oral views of LZA 6046/1 (WH 17).

Figs. 21, 22. Pseudopolygnathus dentilineatus Branson. Oral and aboral views of LZA 6003/1 (Ma 2).

PLATE 57

BUTLER, Carboniferous conodonts

506

PALAEONTOLOGY, VOLUME 16

Polygnathus symmetricus Branson 1934

Plate 59, figs. 22, 23

*1934 Polygnathus symmetrica Branson, p. 310, pi. 25, fig. 11.

1966 Polygnathus symmetrica Branson; Klapper, p. 21, pi. 4, figs. 7, 9; pi. 6, figs. 1, 5 (with
synonymy).

1968 Polygnathus symmetrica Branson; Straka, p. 35, pi. 1, figs. 6, 9 (only).

Remarks. This species is common at the base of the Lower Limestone Shale Group
in the Mendips. Immature forms have a basal cavity which is spathognathodid-like,
underlying the entire platform, and a symmetrical platform.

Material. 68 specimens, from 5 samples.

Polygnathus cf. P. symmetricus Branson 1934

Plate 57, figs. 4, 5, 7, 8, 11, 12, 19, 20

The following are regarded as falling within this subgroup :

1967 Polygnathus mehli Thompson, p. 47, pi. 2, figs. 1-6.

1969 Polygnathus bischoffi Rhodes, Austin, and Druce, p. 184, pi. 13, figs. 8-11.

1969 Polygnathus lacinatus asymmetricus Rhodes, Austin, and Druce, p. 188, pi. 11, figs. 1-4.
1969 Polygnathus lacinatus circaperipherus Rhodes, Austin, and Druce, p. 189, pi. 11, figs. 12-15.
1969 Polygnathus lacinatus lacinatus Huddle; Rhodes, Austin, and Druce, p. 189, pi. 11, figs.
8-10.

1969 Polygnathus lacinatus prelobatus Rhodes, Austin, and Druce, p. 190, pi. 11, figs. 5-7, 11.

1970 Polygnathus bischoffi Rhodes, Austin, and Druce; Marks and Wensink, p. 268, pi. 1, fig. 18.

Remarks. Included in this subgroup are all those ‘late’ polygnathids having a
narrow platform and elongate basal cavity. Immature members of this species have
spathognathodid-like basal cavities, similar to those seen in P. symmetricus. The
platform in immature forms is, however, often asymmetrical, with the right-hand
side consistently better developed than the left (see PI. 57, fig. 12).

The subgroup appears to develop from Pseudopoly gnat hus multistriatus, and forms
with basal cavities resembling those of this pseudopolygnathid are found.

Material. 217 specimens, from 39 samples.

Genus pseudopolygnathus Branson and Mehl 1934
Pseudopolygnathus dentilineatus Branson 1934

Plate 57, figs. 21, 22

*1934 Pseudopolygnathus dentilineata Branson, p. 317, pi. 26, fig. 22.

1966 Pseudopolygnathus dentilineata Branson; Klapper, p. 14, pi. 5, figs. 10, 11 (with synonymy).
V. 1969 Pseudopolygnathus expansus Rhodes, Austin, and Druce, p. 209, pi. 5, figs. 2, 4.

V. 1969 Pseudopolygnathus vogesi Rhodes, Austin, and Druce, p. 216, pi. 5, figs. 1, 3, 5-8.

1970 Pseudopolygnathus dentilineatus Branson; Thompson and Fellows, p. 99, pi. 5, figs. 1, 5.

Remarks. As suggested by Klapper (1966, p. 15), Ps. dentilineatus is characterized
by a basal cavity which, in mature specimens, is as wide as the platform. The basal
cavity tends to be subcircular and widely flared, and is therefore distinct from that
of Ps. multistriatus, which is narrow and attenuate posteriorly.

Material. 1 specimens, from 2 samples.

BUTLER: CARBONIFEROUS CONODONTS

507

Pseudopolygnathus multistriatus Mehl and Thomas 1947

Plate 57, figs. 6, 9, 10, 15, 16, 23-25

*1947 Pseudopolygnathus multisthata Mehl and Thomas, p. 16, pi. 1, fig. 36.

1967 Pseudopolygnathus multistriata Mehl and Thomas; Thompson, p. 49, pi. 4, figs. 15, 16, 19,
20 (with synonymy).

1968 Pseudopolygnathus multistriata Mehl and Thomas; Canis, p. 547, pi. 73, figs. 13, 16.
vl969 Pseudopolygnathus multistriatus Mehl and Thomas; Rhodes, Austin, and Druce, p. 211,

pi. 5, figs. 14-16; pi. 6, fig. 2.

V.1969 Pseudopolygnathus dentilineatus Branson; Rhodes, Austin, and Druce, p. 208, pi. 5, figs.
9-13; pi. 6, fig. 8.

v.1969 Pseudopolygnathus primus Branson and Mehl; Rhodes, Austin, and Druce, p. 214, pi. 6,
figs. 4, 5, 7, 11, 12 (only).

Remarks. Mature specimens of this form possess an elongate basal cavity which is
not as wide as the platform. Late forms may have a much-reduced basal cavity.
Hass (1959, pi. 47) figured an ontogenetic series of Ps. lanceolata (now placed in
synonymy with Ps. multistriatus). Specimens found during the present study con-
form to this series, certain immature forms having nodes on the right-hand side only,
and resembling Spathognathodus aculeatus (Branson and Mehl). Members of this
species are again more fully developed on the right-hand side. The basal cavity may
be relatively large in immature specimens. In certain forms the blade does not
continue into the carina on the platform, but tends towards the nodes on the right-
hand side of the platform (see PI. 57, fig. 23). These forms bear some resemblance
to the genus Clydagnathus Rhodes, Austin, and Druce.

Material. 73 specimens, from 25 samples.

Pseudopolygnathus postinodosus Rhodes, Austin, and Druce 1969

Plate 57, fig. 1

v*1969 Pseudopolygnathus postinodosus Rhodes, Austin, and Druce, p. 213, pi. 6, fig. 6.

Remarks. This form lacks a platform, but the lateral view is similar to that of Ps.
multistriatus (compare PI. 57, fig. 1 with PI. 57, fig. 25).

Material. 1 specimen.

Pseudopolygnathus primus Branson and Mehl 1934

Plate 57, figs. 13, 14, 17, 18

*1934 Pseudopolygnathus prima Branson and Mehl, p. 298, pi. 24, figs. 24, 25.
vl969 Pseudopolygnathus primus Branson and Mehl; Rhodes, Austin, and Druce, p. 214, pi. 6,
fig. 10 (only).

1970 Pseudopolygnathus primus Branson and Mehl; Thompson and Fellows, p. 101, pi. 5, figs.
15, 16, 18, 19 (with synonymy).

Remarks. The characteristics of this species are taken as follows: The platform
possesses a pronounced lateral lobe on either the inner or outer side, separated
from the carina by a depression. The basal cavity is not as wide as the platform in
mature forms. Immature forms again show a more fully developed right-hand side,
irrespective of whether this is the inner or outer side. As Klapper (1966, p. 14) has

508

PALAEONTOLOGY, VOLUME 16

pointed out, some specimens of Ps. primus resemble Ps. triangulus pinnatus Voges
(see PL 57, fig. 18). This latter species possesses a smaller basal cavity, and tends to
have a broad, flat platform.

Material. 6 specimens, from 3 samples.

Pseudopoly gnat hus triangulus Voges 1959
Pseudopoly gnat hus triangulus pinnatus Voges 1959

Plate 58, figs. 23, 24, 27, 28, 30, 31, 33, 34

*1959 Pseudopoly gnathus triangula pinnata Voges, p. 302, pi. 34, figs. 59-66.

1964 Pseudopolygnathus triangula pinnata Voges; Higgins, Wagner-Gentis, and Wagner, pi. 4,
fig. 16.

1967 Pseudopolygnathus triangula pinnata Voges ; Boogaert, p. 285, pi. 3, figs. 9, 10 (with synonymy).
1969 Pseudopolygnathus triangula pinnata Voges; Matthews (1969a), p. 271, pi. 48, figs. 3, 4, 8,
10, 11.

1969 Pseudopolygnathus triangula pinnata Voges; Matthews (19696), pi. 51, fig. 8.

non vl969 Pseudopolygnathus triangulus cf. pinnatus Voges; Rhodes, Austin, and Druce, p. 216, pi. 30,
fig. 19 (= Polygnathus nodomarginatus Branson).

1970 Pseudopolygnathus triangulus pinnatus Voges; Marks and Wensink, p. 269, pi. 1, fig. 17.
1970 Pseudopolygnathus triangulus pinnatus Voges; Thompson and Fellows, p. 102, pi. 6, figs. 6,

11, 12.

Remarks. Specimens found during the present study show a similar range of varia-
tion to that illustrated by Voges (1959). Members of this subspecies again show the
tendency for the right-hand side to be more fully developed than the left. A row of
platform denticles is often present extending along the blade towards the anterior
on the right-hand side, regardless of which side includes the pinnation (compare
PI. 58, fig. 30 with PI. 58, fig. 34).

Material. 65 specimens, from 14 samples.

EXPLANATION OF PLATE 58
Specimens dusted with ammonium chloride. All x 30.

Figs. 1, 2. Mestognathus beckmanni Bischoff. Aboral and oral views of LZA 6135/1 (sample HQ 76).

Figs. 3-5. Pelekysgnathus sp. A Voges 1959. Lateral, oral, and aboral views of LZA 6150/2 (Va 13).

Figs. 6, 7, 21, 22. Scaliognathus anchoralis Branson and Mehl. 6, 7, Oral and aboral views of LZA 6104/2
(HQ 22). 21, 22, Oral and aboral views of LZA 6093/1 1 (HQ 15).

Figs. 8-10, 14-16. Pelekysgnathus bultyncki (Groessens). 8-10, Lateral, aboral, and oral views of LZA
6078/2 (HQ 1). 14-16, Lateral, aboral, and oral views of LZA 6068/6 (HW 18).

Figs. 11-13. Bactrognathus cf. B. distortus Branson and Mehl. 1 1, LZA 6092/1 (HQ 14a): specimen later
accidentally destroyed. 12, 13, Aboral and oral views of LZA 6075/1 (HW 21).

Figs. 17, 18. Doliognathus sp. Aboral and oblique lateral/oral views of LZA 6068/3 (HW 18).

Figs. 19, 20, 25, 26. Dollymae bouckaerti Groessens. 19, 20, Aboral and oral views of LZA 6068/1 1 (HW
18). 25, 26, Oral and aboral views of LZA 6068/8 (HW 18).

Figs. 23, 24, 27, 28, 30, 31, 33, 34. Pseudopolygnathus triangulus pinnatus Voges. 23, 24, Aboral and oral
views of LZA 6077/2 (HW 23). 27, LZA 6104/1 (HQ 22). 28, LZA 6106/1 (HQ 23). 30, 31, Oral and
aboral views of LZA 6077/1 (HW 23). 33, 34, Aboral and oral views of LZA 6098/2 (HQ 18).

Fig. 29. Hindeodella segaformis Bischoff. LZA 6129/1 (HQ 46).

Fig. 32. Pseudopolygnathus triangulus triangulus Voges. LZA 6111/1 (HQ 28).

PLATE 58

BUTLER, Carboniferous conodonts

510

PALAEONTOLOGY, VOLUME 16

Pseudopolygnathus triangulus triangulus Voges 1959

Plate 58, fig. 32

*1959 Pseudopolygnathus triangula triangula Voges, p. 304, pi. 35, figs. 7-13.

1969 Pseudopolygnathus triangula triangula Voges; Matthews (1969a), p. 271, pi. 48, figs. 2, 7.

1970 Pseudopolygnathus triangulus triangulus Voges; Thompson and Fellows, p. 103, pi. 7, figs.
6, 7 (with synonymy).

Remarks. Specimens recovered are similar to those figured by Voges (1959, pi. 35).
Voges (1959, table 1) gave this subspecies a lower range than that of Ps. triangulus
pinnatus. In the eastern Mendips the two forms have almost identical ranges.

Material. 8 specimens (including 2 cf. determinations), from 6 samples.

Genus scaliognathus Branson and Mehl 1941
Scaliognathus anchoralis Branson and Mehl 1941

Plate 58, figs. 6, 7, 21, 22

*1941 Scaliognathus anchoralis Branson and Mehl, p. 102, pi. 19, figs. 29-32.

1964 Scaliognathus anchoralis Branson and Mehl; Burton, range-chart, facing p. 74.

1967 Scaliognathus anchoralis Branson and Mehl; Zikmundova, pi. 1, figs. 1, 2, 4.

1968 Scaliognathus anchoralis Branson and Mehl; Schulze, p. 220, pi. 20, fig. 32 (with synonymy).

1969 Scaliognathus anchoralis Branson and Mehl; Matthews (1969a), p. 272, pi. 49, figs. 1-10.

1969 Scaliognathus anchoralis Branson and Mehl; Matthews (19696), pi. 51, figs. 1, 2.

1970 Scaliognathus anchoralis Branson and Mehl; Marks and Wensink, p. 269, pi. 4, fig. 12.

1970 Scaliognathus anchoralis Branson and Mehl ; Thompson and Fellows, p. 103 (with additional
synonymy).

1971 Scaliognathus anchoralis Branson and Mehl; Groessens, pi. 1, figs. 9, 10.

Remarks. The species is one of the indices of the German anchor alis-Zone. It has, in
the past, been reported mainly from ‘basinal’ sequences, and there have been sugges-
tions (e.g. Rhodes et al. 1969, p. 65) that facies-control resulted in virtual exclusion
of this species from the ‘shelf’ limestone successions. It now seems clear that the
species is present in the limestone sequences and that its absence at many places is
due in part to discontinuities within the successions.

The species appears to have developed from Pelekysgnathus bultyncki (Groessens)
by the development of lateral processes at the posterior end (see discussion under
Pe. bultyncki, above).

Material. 7 specimens, from 5 samples.

Genus siphonodella Branson and Mehl 1944

Remarks. The suggestions put forward by Klapper (1971) on the systematic palae-
ontology of this genus are accepted here.

Siphonodella cooper i Hass 1959

Plate 59, figs. 39, 40

*1959 Siphonodella cooperi Hass, p. 392, pi. 49, figs. 35, 36.

1971 Siphonodella cooperi Hass; Klapper, p. 10, pi. 1, figs. 13-15; pi. 2, figs. 1-3 (with synonymy).

BUTLER: CARBONIFEROUS CONODONTS

511

Remarks. This species is distinguished from S. isosticha (Cooper) by the ornament
of transverse ridges on the outer side of the platform. Most specimens of S. isosticha
also show faint ridges on the outer side, but some are considered to be sufficiently
well developed in this respect to justify inclusion within S. cooperi.

Material. 3 specimens, from 2 samples.

Siphonodella cf. erenulata (Cooper 1939)

Plate 59, figs. 36, 37

Remarks. Matthews and Butler (in press) discuss occurrences of siphonodellids
with platform outlines resembling those of S. erenulata (Cooper), but without the
characteristic ornament of this species.

Material. 2 specimens, from 1 sample.

Siphonodella isostieha (Cooper 1939)

Plate 59, figs. 21, 30, 31

*1939 Siphonognathus isosticha Cooper, p. 409, pi. 41, figs. 9, 10.

1971 Siphonodella isosticha (Cooper); Klapper, p. 10, pi. 1, fig. 16 (with synonymy).

Remarks. Klapper (1971, p. 10) made clear the definition of Siphonodella isosticha
and reillustrated Cooper’s holotype. In this species the longest rostral ridge ter-
minates at the outer margin. The inner platform is weakly nodose and the outer
margin may be unornamented or have weak transverse ridges. This redefinition
makes it clear that Siphonodella cooperi hassi Thompson and Fellows is synonymous
with S. isosticha.

Material. 21 specimens, from 6 samples.

Siphonodella cf. S. isosticha (Cooper 1939)

Plate 59, fig. 38

1971 Siphonodella cf. S. isosticha (Cooper); Klapper, p. 12, pi. 1, figs. 17-20 (with synonymy).

Remarks. Klapper (1971, p. 12) applied this name to those siphonodellids with
weak ornament but with rostral ridges terminating on the platform without reaching
the margin.

Material. 5 specimens, from 3 samples.

Siphonodella obsoleta Hass 1959

Plate 59, figs. 33, 34

*1959 Siphonodella obsoleta Hass, p. 392, pi. 47, figs. 1, 2.

1971 Siphonodella obsoleta Hass; Klapper, p. 12, pi. 1, fig. 25 (with synonymy).
vl972 Siphonodella obsoleta Hass; Matthews, Sadler, and Selwood, p. 565, pi. Ill, figs. 4, 5 (with
additional synonymy).

Remarks. This species is characterized by the presence of a long outer rostral ridge,
extending to the posterior part of the platform before meeting the margin. The

512

PALAEONTOLOGY, VOLUME 16

platform is relatively long and narrow. As Klapper (1971, p. 12) has pointed out,
the species may have more than two rostral ridges. Certain specimens here have an
additional ridge (see PI. 59, fig. 33).

Material. 15 specimens, from 5 samples.

Genus spathognathodus Branson and Mehl 1941
Spathognathodus aculeatus (Branson and Mehl 1934)

*1934 Spathodus aculeatus Branson and Mehl, p. 186, pi. 17, figs. 11, 14.

Remarks. Matthews (in Matthews and Naylor 1973) gives full systematic treatment
to this species. Specimens recovered during the present study are not well enough
preserved or numerous enough to justify lengthy discussion.

Material. 4 specimens, from 2 samples.

Spathognathodus crassidentatus (Branson and Mehl 1934)

Plate 59, figs. 18, 24, 29

*1934 Spathodus crassidentatus Branson and Mehl, p. 276, pi. 22, fig. 17.

1966 Spathognathodus crassidentatus (Branson and Mehl); Klapper, p. 23, pi. 5, figs. 15-17 (with
synonymy).

1970 Spathognathodus crassidentatus (Branson and Mehl); Thompson and Fellows, p. Ill, pi. 7,
figs. 8, 14.

EXPLANATION OF PLATE 59

Specimens dusted with ammonium chloride. All x 30.

Figs. 1, 2. Patrognathus variabilis Rhodes, Austin, and Druce. Lateral and oral views of LZA 6005/2
(sample Ma 4).

Figs. 3, 14, 32. Spathognathodus stabilis (Branson and Mehl). 3, LZA 6150/3 (Va 13). 14, LZA 6104/3
(HQ 22). 32, LZA 6118/3 (HQ 35): variant approaching 5. crassidentatus.

Figs. 4, 5. Spathognathodus scitulus (Hinde). Aboral and lateral views of LZA 6129/2 (HQ 46).

Figs. 6, 7, 19, 20. Polygnathus inornatus Branson. 6, 7, Aboral and oral views of LZA 6019/8 (Ma 19).

19, 20, Oral and aboral views of LZA 6012/2 (Ma 11).

Figs. 8, 9, 15-17. Polygnathus communis communis Branson and Mehl. 8, 9, Oral and aboral views of LZA
6019/1 1 (Ma 19). 15-17, Lateral, aboral, and oral views of LZA 6100/3 (HQ 19).

Figs. 10-13, 26. Polygnathus communis carina Hass. 10, 11, Oral and aboral views of LZA 6083/4 (HQ 7).

12, 13, Oral and aboral views of LZA 6083/3 (HQ 7). 26, LZA 6083/1 (HQ 7).

Figs. 18, 24, 29. Spathognathodus crassidentatus (Branson and Mehl). 18, LZA 6098/3 (HQ 18). 24, 29,
Oral and lateral views of LZA 6086/2 (HQ 10).

Figs. 21, 30, 31. Siphonodella isosticha (Cooper). 21, LZA 6019/3 (Ma 19). 30, 31, Oral and aboral views
of LZA 6019/1 (Ma 19).

Figs. 22, 23. Polygnathus symmetricus Branson. Aboral and oral views of LZA 6005/1 (Ma 4).

Figs. 25, 28. Elictognathus laceratus (Branson and Mehl). 25, LZA 6019/12 (Ma 19). 28, LZA 6019/13
(Ma 19).

Figs. 27, 35. Siphonodella sp. (juv?). 27, LZA 6019/7 (Ma 19). 35, LZA 6019/4 (Ma 19).

Figs. 33, 34. Siphonodella obsoleta Hass. 33, LZA 6019/2 (Ma 19). 34, LZA 6014/1 (Ma 15).

Figs. 36, 37. Siphonodella cf. crenulata. Aboral and oral views of LZA 6012/1 (Ma 11).

Fig. 38. Siphonodella cf. S. isosticha. LZA 6019/5 (Ma 19).

Figs. 39, 40. Siphonodella cooperi Hass. Oral and aboral views of LZA 6019/6 (Ma 19).

PLATE 59

jWJ

BUTLER, Carboniferous conodonts

514

PALAEONTOLOGY, VOLUME 16

Remarks. This species was restricted by Klapper (1966, p. 23) to those forms having
two main denticles at the anterior end of the blade and a slightly arched profile.
Specimens with a similar profile but a single major anterior denticle are also in-
cluded in the species here (see PI. 59, fig. 18). A fuller synonymy list is available in
Matthews (in Matthews and Naylor 1973).

Material. 40 specimens, from 20 samples.

Spathognathodus sci talus (Hinde 1900)

Plate 59, figs. 4, 5

*1900 Polygnathus scitulus Hinde, p. 343, pi. 9, figs. 9-11 (only).

1963 Spathognathodus scitulus (Hinde); Rexroad and Collinson, p. 20, pi. 2, figs. 14, 19, 29-31
(with synonymy).

vl969 Spathognathodus scitulus (Hinde); Rhodes, Austin, and Druce, p. 232, pi. 8, figs. 9-11.

Remarks. This spathognathodid has a distinctively flared basal cavity on the outer
side. Few denticles are present and these expand rapidly in size from the posterior
end to the anterior. In some specimens one or two ridges develop on the oral surface
of the flared cavity.

Material. 1 1 specimens, from 6 samples.

Spathognathodus stabilis (Branson and Mehl 1934)

Plate 59, figs. 3, 14, 32

*1934 Spathodus stabilis Branson and Mehl, p. 188, pi. 17, fig. 20.

Remarks. Matthews (in Matthews and Naylor 1973) supplies a full description and
synonymy list for this species, and no attempt will be made to duplicate this here.
Specimens similar to that illustrated as Spathognathodus sp. A by Groessens (1971,
pi. 2, fig. 7) occur in many samples from anchoralis-Zone levels in the Mendips.
They are considered to lie within the range of variation of Sp. stabilis.

Material. 174 specimens, from 49 samples.

Acknowledgements. This paper embodies work carried out while the author was in receipt of a University
of Bristol Postgraduate Scholarship. The facilities provided by the Department of Geology at Bristol are
acknowledged. Special thanks are due to S. C. Matthews, who supervised the project, for his advice and
assistance and to E. W. Seavill and R. Godwin for their photographic work.

REFERENCES

AUSTIN, R. L., CONIL, R., DOLBY, G., LYS, M., PAPROTH, E., RHODES, F. H. T., STREEL, M., UTTING, J. and WEYER, D.

1971. Les couches de passage du Devonien au Carbonifere de Hook Head (Irlande) au Bohlen (D.D.R.).
Cong. Collogues Univ. Liege, 55, 167-177, 2 tables.

RHODES, F. H. T. and STREEL, M. 1970. Conodontes, spores et foraminiferes du Tournaisien in-

ferieur dans la vallee du Hoyoux. Annls. Soc. geol. Beige, 93, 305-315, pi. 1, text-figs. 1-2.

BiscHOFF, G. 1957. Die Conodonten-Stratigraphie des rheno-herzynischen Unterkarbons mit Beriick-
sichtigung der Wocklumeria-Stufe und der Devon/Karbon-Grenze. Abh. hess. Landesamt. Bodenf. 19,
1-64, pis. 1-6, 3 text-figs.

and ZIEGLER, w. 1956. Das Alter der Urfer Schichten im Marburger Hinterland nach Conodonten.

Notizbl. hess. Landesamt Bodenf., Wiesbaden, 84, 138-169, pis. 1-4, 1 text-fig.

BOOGAERT, VAN A. 1967. Devonian and Lower Carboniferous conodonts of the Cantabrian Mountains
(Spain) and their stratigraphic application. Leid. geol. Meded. 39, 129-192, pis. 1-3.

BUTLER: CARBONIFEROUS CONODONTS

515

BRANSON, E. B. and MEHL, M. G. 1934a. Conodonts from the Grassy Creek Shale of Missouri. Univ. Mis-
souri Studies, 8 (1933), 171-259, pis. 13-21.

19346. Conodonts from the Bushberg Sandstone and equivalent formations of Missouri.

Univ. Missouri Studies, 8 (1933), 265-299, pis. 22-24.

1938. Conodonts from the Lower Mississippian of Missouri. Univ. Missouri Studies, 13, 128-

148, pis. 33-34.

1941. New and little known Carboniferous conodont genera. J. Paleont. 15, 97-106, pi. 19.

1944. Conodonts. In shimer, h. w. and shrock, r. r.. Index fossils of North America, pp. 235-

246, pis. 93, 94. New York, Wiley.

BRANSON, E. R. 1934. Conodonts from the Hannibal formation of Missouri. Univ. Missouri Studies, 8
(1933), 301-343, pis. 25-28.

BUDINGER, p. 1965. Conodonten aus dem Oberdevon und Karbon des Kantabrischen Gebirges (Nordspanien).

Inaugural dissertation. University of Tubingen, 1-103, pis. 1-5.

BURTON, R. c. 1964. A preliminary range chart of Lake Valley Formation (Osage) conodonts in the southern
Sacramento Mountains, New Mexico. New Mexico Geol. Soc. Guidebook, 15th Field Conf. 73-75,
1 chart.

BUTLER, M. 1972. Conodont faunas and stratigraphy of certain Tournaisian sections in the Bristol- Mendip
area. Ph.D. thesis. University of Bristol.

CANis, w. F. 1968. Conodonts and biostratigraphy of the Lower Mississippian of Missouri. J. Paleont.
42, 525-555, pis. 72-74, 2 text-figs.

COLLiNSON, c. 1963. Collection and preparation of conodonts through mass production techniques.
Circ. Illinois geol. Survey, 343, 1-16.

SCOTT, A. J. and rexroad, c. b. 1962. Six charts showing biostratigraphic zones, and correlations

based on conodonts from Devonian and Mississippian rocks of the upper Mississippi Valley. Circ.
Illinois geol. Survey, 328, 1-32.

REXROAD, c. B. and THOMPSON, T. F. 1971. Conodont zonation of the North American Mississippian.

In SWEET, w. c. and Bergstrom, s. (eds.), Symposium on conodont biostratigraphy. Mem. geol. Soc.
Amer. 127, 353-394, 8 text-figs., 1 table.

CONIL, R., AUSTIN, R. L., LYS, M. and RHODES, F. H. T. 1969. La limite des etages toumaisien et viseen au
stratotype de I’assise de Dinant. Bull. Soc. beige Geol. Paleont. Hydrol. 77, 39-69.

LYS, M. and MAUVIER, A. 1964. Criteres micropaleontologiques essentiels des formation-types du

Carbonifere (Dinantien) du bassin Franco-Beige. Congres Avanc. Etud. Stratigr. carb., Paris 1963,
325-332, pis. 1, 2.

COOPER, c. L. 1939. Conodonts from a Bushberg-Hannibal horizon in Oklahoma. J. Paleont. 13, 379-
422, pis. 39-47.

DRUCE, E. c. 1969. Devonian and Carboniferous conodonts from the Bonaparte Gulf Basin, northern
Australia. Bull. Bur. Miner. Resour. Geol. Geophys. Aust. 98, 3-157, pis. 1-43.

GEORGE, T. N. 1954. Pre-Seminulan Main Limestone of the Avonian Series in Breconshire. Q. Jl geol. Soc.
Lond. 110, 293-322, pi. 12, 21 text-figs.

1958. Lower Carboniferous palaeogeography of the British Isles. Proc. Yorks, geol. Soc. 31, 227-

318, pi. 15, 27 text-figs.

GREEN, G. w. and WELCH, F. B. A. 1965. Geology of the country around Wells and Cheddar. Mem. geol.
Surv. U.K. 1-225, pis. 1-5, 19 text-figs.

GROESSENS, E. 1971 . Les conodontes du Toumaisien superieur de la Belgique. Prof. Pap. Serv. geol. Belgique,
4/1971, 1-19, pis. 1, 2.

HASS, w. H. 1959. Conodonts from the Chappel Limestone of Texas. Prof. Pap. U.S. geol. Surv. 294-J,
365-399, pis. 46-50.

HIGGINS, A. c., WAGNER-GENTis, c. H. T. and WAGNER, R. H. 1964. Basal Carboniferous strata in part of
northern Leon, N.W. Spain : Stratigraphy, conodont and goniatite faunas. Bull. Soc. beige Geol. Paleont.
Hydrol. 72, 205-247, pis. 1-5, 6 text-figs.

HiNDE, G. 1879. On conodonts from the Chazy and Cincinnati group of the Cambro-Silurian, and from
the Hamilton and Genesee-shale divisions of the Devonian in Canada and the United States. Q. Jl geol.
Soc. Lond. 35, 351-369, pis. 15-17.

1900. Notes and descriptions of new species of Scotch Carboniferous conodonts. Trans, nat. Hist.

Soc., Glasg. 5, 338-346, pis. 9-10.

516

PALAEONTOLOGY, VOLUME 16

KELLAWAY, G. A. 1971. In MATTHEWS, s. c. (ed.), Excursion 1 ; Bristol-Mendip area and south-west England.
Congres. Avanc. Etude Stratigr. Carb., Sheffield 1967, 4, 1645-1651.

and WELCH, f. b. a. 1955. The Upper Old Red Sandstone and Lower Carboniferous rocks of Bristol

and the Mendips compared with those of Chepstow and the Forest of Dean. Bull. geol. Surv. Gt. Br.
9, 1-21, pi. 1.

KLAPPER, G. 1966. Upper Devonian and Lower Mississippian conodont zones in Montana, Wyoming
and south Dakota. Paleont. Contr. Univ. Kansas, 3, 1-43, pis. 1-6.

1971. Patrognathus and Siphonodella (Conodonta) from the Kinderhookian (Lower Mississippian)

of western Kansas and southwestern Nebraska. Bull. St. geol. Surv. Univ. Kansas, 202, 3-14, pis. 1, 2.

KRONBERG, p., PiLGER, A., SCHERP, A. and ZIEGLER, w. 1960. Spuren altvariscischer Bewegungen im nordo-
stlichen Teil des Rheinischen Schiefergebirges. Fortschr. Geol. Rheinld Westf. 3 (1), 1-46, pis. 1-7,
13 text-figs., 3 tables.

MARKS, p. and WENSiNK, H. 1970. Conodonts and the age of the ‘Griotte’ Limestone Formation in the
Upper Aragon Valley (Huesca, Spain). 1. Proc. K. ned. Akad. Wet. B, 73, 238-275, pis. 1-4, 7 text-figs.,
4 tables.

MATTHEWS, s. c. 1969u. A Lower Carboniferous conodont fauna from east Cornwall. Palaeontology, 12,
262-275, pis. 46-50, 1 text-fig.

19696. Two conodont faunas from the Lower Carboniferous of Chudleigh, south Devon. Palaeonto-
logy, 12, 276^280, p\. 51. _

and BUTLER, M. (in press). Siphonodellid conodonts in the Lower Carboniferous. Congres Avanc.

Etud. Stratigr. carb., Krefeld 1971.

and NAYLOR, D. 1973. Lower Carboniferous conodont faunas from south-west Ireland. Palaeontology,

16, 335-380, pis. 35-38.

SADLER, p. M. and SELWOOD, E. B. 1972. A Lower Carboniferous conodont fauna from Chillaton,

southwest Devonshire. Palaeontology, 15, 550-568, pis. 109-111, 1 text-fig.

MEHL, M. G. and THOMAS, L. A. 1947. Conodonts from the Fern Glen of Missouri. J. Scient. Labs. Denison
Univ. 40, 3-19, pi. 1.

MEISCHNER, D. 1971. Conodonten-chronologie des deutschen Karbons. Congres. Avanc. Etude. Stratigr.
Carb., Sheffield 1967, 3, 1169-1180.

MITCHELL, M. 1971. In Annual Report Inst. Geol. Sci., for 1970, 97-98.

1972. The base of the Visean in South-West and North-West England. Proc. Yorks, geol. Soc. 39,

151-160.

PANDER, c. H. 1856. Monographie der fossilen Fische des silurischen Systems der russisch-baltischen Gou-
vernements, 1-91, pis. 1-8. K. Akad. Wiss: St. Petersburg.

REXROAD, c. B. and COLLINSON, c. 1963. Conodonts from the St. Louis Formation (Valmeyeran Series) of
Illinois, Indiana, and Missouri. Circ. Illinois geol. Survey, 355, 1-28, pis. 1-2, 5 text-figs., 1 table.

and SCOTT, a. j. 1964. Conodont zones in the Rockford Limestone and New Providence Shale (Mis-
sissippian) in Indiana. Bull. Indiana Dep. Conserv. Geol. Surv. 30, 1-54, 6 figs., pis. 2, 3.

RHODES, F. H. T. and AUSTIN, R. L. 1971. Carboniferous conodont faunas of Europe. In sweet, w. c. and
BERGSTROM, s. (eds.). Symposium on conodont biostratigraphy. Mem. geol. Soc. Amer. 127, 317-352,
pis. 1-2, 5 text-figs.

and DRUCE, e. c. 1969. British Avonian (Carboniferous) conodont faunas and their value in

local and intercontinental correlation. Bull. Br. Mus. nat. Hist. {Geol.), 5, 4-313, pis. 1-31, text-figs. 1-92.

RICHTER, R. 1948. Einfuhrung in die Zoologische Nomenklatur, 252 pp. Frankfurt-a-M., Kramer.

SANDO, w. J., MAMET, B. L. and DUTRO, J. T. 1969. Carboniferous megafaunal and microfaunal zonation in
the Northern Cordillera of the United States. Prof. Pap. U.S. geol. Surv. 613-E, 1-29, pi. 1, 8 text-
figs.

SCHULZE, R. 1968. Die Conodonten aus dem Palaozoikum der mittleren Karawanken (Seeberggebiet).
Neues Jb. Geol. Paldont., Abb. 130, 133-245, pis. 16-20.

STRAKA, J. J. 1968. Conodont zonation of the Kinderhookian Series, Washington County, Iowa. Univ.
Iowa Stud. nat. Hist. 21, 1-71, pis. 1-7, 5 text-figs.

THOMAS, L. A. 1949. Devonian- Mississippian formations of southeast Iowa. Bull. geol. Soc. Am. 60, 403-
438, pis. 1-4.

THOMPSON, T. L. 1967. Conodont zonation of Lower Osagean rocks (Lower Mississippian) of south western
Missouri. Missouri Geol. Surv. Wat. Resour. Rept. Invest. 39, 88 pp., pis. 1-6.

BUTLER: CARBONIFEROUS CONODONTS

517

THOMPSON, T. L. and FELLOWS, L. D. 1970. Stratigraphy and conodont biostratigraphy of Kinderhookian
and Osagean rocks of south western Missouri and adjacent areas. Missouri Geol. Surv. Wat. Resour.
Kept. Invest. 39, 3-263, pis. 1-9.

ULRICH, E. o. and bassler, r. s. 1926. A classification of the tooth-like fossils, conodonts, with descriptions
of American Devonian and Mississippian species. Proc. U.S. Natn. Mus. 68, 1-63, pis. 1-11.

VAUGHAN, A. 1905. The palaeontological sequence in the Carboniferous Limestone of the Bristol area.
Q. J I geol. Soc. Lond. 61, 181-305, pis. 22-29, 6 text-figs.

VOCES, A. 1959. Conodonten aus dem Untercarbon I und II (Gattendorfia- und Pericyclus-Stufe) des
Sauerlandes. Paldont. Z. 33, 266-314, pis. 33-35, 5 text-figs., 1 table.

1960. Die Bedeutung der Conodonten fiir die Stratigraphie des Unterkarbons I und II (Gattendorfia-

und Pericyclus-Stufe) im Sauerland. Fortschr. Geol. Rheinld Westf. 3 (1), 197-228, 5 text-figs.

WELCH, F. B. A. 1932. The geological structure of the eastern Mendips. Q. Jl geol. Soc. Lond. 89, 14-50,
pis. 1-4, 7 text-figs.

wiRTH, M. 1967. Zur Gliederung des hdheren Palaozoikums (Givet -Namur) im Gebiet des Quinto Real
(Westpyrenaen) mit Hilfe von Conodonten. Neues Jb. Geol. Paldont., Abh. 127, 179-244, pis. 19-23,
15 text-figs.

ZIEGLER, w. 1971. (Review of Rhodes, Austin, and Druce 1969.) Zentbl. Geol. Paldont. II (Heft 5, Jg.
1970), 365-369.

ziKMUNDOVA, J. 1967. Konodontova zona Scaliognathus anchoralis Branson and Mehl v ponikevskych
bfidlicich Nizkeho Jeseniku. Vest, ustfed. Ust. geol. 42, 449-452, pis. 1-4.

MALCOLM BUTLER

Texaco Limited
1 Knightsbridge Green

Typescript received 8 September 1972 London SWIX 7QJ

APPENDIX

List of conodont faunas from isolated Lower Limestone Shale Group samples, (y-kg samples processed.)

1. The Palate Bed, from the Avon Gorge (ST 555 746): lElictognathus sp. (3 specimens), Clydagnathus
spp. (10), Patrognathus variabilis Rhodes, Austin, and Druce (3), Polygnathus inornatus Branson (33),
P. scobiniformis Branson (4), Pseudopolygnathus dentilineatus Branson (7), Siphonodella duplicata
(Branson and Mehl) (3), Spathognathodus plumulus Rhodes, Austin, and Druce (11), Sp. stabilis
Branson and Mehl (1), Bars (24). Slide LZA 6161.

2. Phosphatic lag deposit on Clevedon Foreshore (ST 402 718): Clydagnathus sp. (2), Patrognathus
variabilis Rhodes, Austin, and Druce (1), Polygnathus inornatus Branson (8), Pseudopolygnathus
dentilineatus Branson (1), Spathognathodus plumulus Rhodes, Austin, and Druce (1), Bars (10).
Slide LZA 6164.

3. Bed with Vaughania on Portishead Foreshore (ST 465 775): Spathognathodus plumulus Rhodes,
Austin, and Druce (3). Slide LZA 6165.

4. Small quarry north of road at Asham (ST 717 463). Oolite: Patrognathus variabilis Rhodes, Austin,
and Druce (2), Polygnathus symmetricus Branson (20), Pseudopolygnathus dentilineatus Branson (4),
Spathognathodus stabilis Branson and Mehl (1), Bars (15). Slide LZA 6162.

5. Sample from stream section at Asham (ST 717 464): Pseudopolygnathus dentilineatus Branson (2),
Bars (2). Slide LZA 6163.

THE STRUCTURAL EVOLUTION OF
THE BIVALVE SHELL

by JOHN D. TAYLOR

Abstract. Direct study of the course of evolution of bivalve shell structures has been prevented by the lack of
well-preserved lower Palaeozoic material. The ‘primitive’ molluscan shell structure probably consisted of an outer
aragonitic prismatic layer, the prisms being polygonal in transverse and columnar in longitudinal sections. The
middle and inner shell layers consisted of nacreous structures. Morphologically similar structures are produced
inorganically from the solidification of metals containing impurities. It is suggested that the prism/nacre combina-
tion originally arose spontaneously as a result of the precipitation of calcium carbonate with protein (impurity).
The subsequent elaboration of the shell structure combinations took place along seven major morphological trends.
The main structural changes have been: the modification and loss of the outer prismatic layer; the elaboration of
the middle layer from nacre into various other types of dendritic growth such as calcitic foliated or aragonitic
crossed-lamellar structures; and the loss of organized structure to produce a homogeneous granular structure. In
all the series there has been a progressive loss of layers from the ‘primitive’ three to a more ‘advanced’ two or even one.

In recent years there has been considerable interest in the calcified structures of
invertebrates and the structure of the molluscan shell, particularly that of the Bivalvia,
has received much attention (B^ggild 1930; Taylor et al. 1969, 1973; Wise 1970,
1971). It is now known that the shells consist of a number of distinct structures and
the micro-morphology and distribution of these structures amongst the various
taxa is becoming well known. Although the arrangement of the structures in each
of the bivalve superfamilies is obviously related to their phylogenetic history, it is
difficult to see how the various structures are related to each other and how they
might have evolved. Whilst it has been possible to study the shell structure of some
fossil bivalves, the preservation problems caused by the usually aragonitic shells
have meant that except in very few cases it has not been possible to extend these
studies very far back into the Palaeozoic. This is in contrast to the Brachiopoda,
where the frequently good preservation of the calcitic shell has enabled the shell
structure of Cambrian forms to be examined (Williams 1968). This lack of informa-
tion from the lower Palaeozoic is particularly unfortunate because many bivalve
lineages are of considerable antiquity and it seems that most of the major radiation
of shell structure types took place in the Ordovician (Pojeta 1971).

If it is accepted that the Bivalvia are a monophyletic group, then all of the shell
structure types observed in Recent bivalves must have evolved from a single shell
structure combination. A study of the distribution of shell structures in all living
superfamilies (Taylor et al. 1969, 1973) included the discovery of some transitional
combinations, which, together with evidence of relationships derived from other
available characters makes possible the tentative presentation of an attempt to
demonstrate the course of evolution of bivalve shell structures. The original stimu-
lus for this idea was the discovery of a metallurgical analogy (described below)
which even if not directly applicable to calcification in bivalves, at least provides a
model which (to the author at least) has made the relationships of the various shell
structures comprehensible.

[Palaeontology, Vol. 16, Part 3, 1973, pp. 519-534, pi. 60.]

520

PALAEONTOLOGY, VOLUME 16

THE SHELL STRUCTURES

The various bivalve shell structures have been described in detail by Taylor et
al. 1969, Wise 1970, 1971, and only the main points relevant to this paper are dis-
cussed below.

Prismatic structures

Simple prismatic structures, whether calcite or aragonite have been shown by
various workers (Taylor et al. 1969) to resemble the group growth of spherulites
seen in inorganic samples (Grigor’ev 1965). Recent further work has shown the
spherulitic nature of the first calcification on the periostracum surface (PI. 60, figs.
1 and 2). Although the crystallographic c axis is generally parallel to the long axis
of the prisms, the alignment is not always exact and each of the small crystallites
making up the prism has a slightly different orientation, usually divergent from
the morphological long axis. In some species the arrangement of the crystallites
may be fanlike.

The composite prismatic structure has the longest morphological and the crystal-
lographic c axes aligned more or less parallel to the outside of the shell, but in many
respects closely resembles the simple prismatic structure.

Nacreous structures

Sheet nacre consists of tablet-shaped crystallites laid down in laminae; the crystal-
lographic b axis of the tablets is generally oriented in the growth direction of the
shell and the c axis is normal to the plane of the tablet. Areas of nacre appear to
behave as a single crystal and the crystallites link up to produce large dendritic
growth patterns (text-fig. 1). Often large growth spirals are formed arising from
screw dislocations (Wada 1961). In columnar or lenticular nacre (Taylor et al.
1969; Wise 1970) the tablets are arranged into columns, the growth axis of the
columns corresponding to the c axis of the aragonite. These columns apparently
arise by screw dislocations at the growing tip (see Erben 1971, p. 59, pi. 2, fig. 5)
and are another form of dendritic growth.

Foliated structure

This structure is always composed of calcite and consists of long lath-like crystal-
lites arranged in side-to-side contact and into overlapping sheets. In general the
crystallographic c axis is aligned in the growth direction but local differences in
alignment of areas of crystallites are common. This structure has long been con-
sidered as dendritic growth (Watabe and Wilbur 1961) and as shown in PI. 60,
fig. 4, this interpretation is reasonable.

Crossed-lamellar structure

This is one of the most common shell structures. It consists of elongate needle-
like crystallites which are arranged into lamellae. In adjacent lamellae the morpho-
logical alignment of the crystallites differs by about 98°. The crystallographic c axis
lies within the plane of each lamella, but the orientation of the c axis varies by approxi-
mately 8-10° between adjacent lamellae. Although the structure shows strong

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL

521

TEXT-FIG. 1 . Dendritic growth pattern of aragonite nacre crystallites
on the inner surface of the inner layer of Neotrigonia dubia. Traced
from electron-micrographs, x 9000.

superficial resemblance to twinning there is no
evidence of any twin relationship between the
lamellae. Each of the two orientations shown
by crystallites in the lamellae show complex
branching patterns (text-fig. 2) and it is possible
that some sort of dendritic growth mechanism
is operative. Much work on the mode of secre-
tion of this structure, similar to that on prisms
by Nakahara and Bevelander (1971) is needed.

Complex crossed-lamellar structure

This structure appears genetically related to
crossed-lamellar structure and can probably be
best thought of as being the intergrowth pattern
resulting from crossed-lamellar structure in two
alignments at right angles to each other. As a
result there are four orientations of crystallites.
Occasionally the texture resembles that of the
patellacean gastropods, interpreted as spheru-
litic by McClintock (1967).

Homogeneous structure

This is a name given to a fine-grained struc-
ture with no particular crystal form; it can be

G

TEXT-FIG. 2. Tangential section of the
outer crossed-lamellar layer of Hippopus
hippopus showing the dendritic nature of
each of the lamellar orientations. Traced
from photomicrographs.

522

PALAEONTOLOGY, VOLUME 16

derived from any of the other shell structures by a diminution in grain size and a
breakdown of structural arrangements. Further detailed study will probably reveal
different types of homogeneous structures but these are not yet apparent.

Homologies between layers

Although homologies between shell layers from various taxa should be made
with caution (Taylor et al. 1969) it is reasonable in most cases to use the trace of the
pallial myostracum as a marker horizon. This separates the inner from the middle
and outer layers or just the outer layer, depending on how many layers are present.
In some Pteriomorphia the pallial myostracum is absent and in others it represents
secondary pallial attachment and thus great care must be taken if homologies are
attempted between this group and the rest of the bivalves.

The ''primitive' shell structure

Before discussing the evolution of the bivalve shell structure it is necessary to
attempt to establish the nature of the ‘primitive’ shell structure. McAlester (1965,
1966) has demonstrated, mainly through the evidence of pedal muscle scars, how
the bivalve Babinka may be derived from a monoplacophoran ancestor. At the time,
McAlester thought that this may have been a special character of the Lucinacea
and he suggested a polyphyletic origin for the bivalves. However, it is now known
that several other bivalve groups may be similarly derived from a monoplacophoran
ancestor (N. J. Morris, pers. comm.) and the monophyletic derivation of all bivalves
from this source is a reasonable proposition. It would thus seem reasonable to
regard the structure of the Monoplacophora as being the ancestral structure to
that of the Bivalvia. The work of Schmidt (1959) and Erben et al. (1968) has shown
that, with the exception of Tryblidium, the shell structure of Monoplacophora is, and
was, aragonitic and consisted of an outer simple prismatic layer and nacreous inner
layers (text-fig. 3). The prisms of the outer layer lie with their long axes normal to
the outer shell surface, are polygonal in horizontal section, and bounded by a sheath
of protein matrix (see also figs. 1-8 in Menzies 1968). The inner nacreous layer is
divided by a thin sheet of blocky prisms such as are normally secreted beneath muscle
attachment areas. Another sheet of these prisms occurs on the innermost part of the
shell. Similar sheets of myostracal prisms have been described from the inner shell
layer of some Mytilacea (Taylor et al. 1969, pi. 25, fig. 2) and it was suggested that
they were formed during times of temporary attachment of the mantle to the shell.

Several bivalve superfamilies have a structural combination of aragonite simple
prisms with inner and middle nacreous layers. Although the number of living super-
families having this combination is only five out of thirty-nine, they usually belong
to lineages which extend far back into the Palaeozoic, whereas many of the other
superfamilies not having this structure have arisen in the late Palaeozoic or Meso-
zoic. The superfamilies having this ‘ancestral’ condition are the Pholomyadacea,
Pandoracea, Poromyacea, Unionacea, and Trigonacea. Moreover the combination
of calcite prisms and nacre, as found in the Pteriacea, Pinnacea, Mytilacea, and
Ambonychiacea (extinct) which all originated in the Palaeozoic, is not very dif-
ferent. It has often been stated that the Nuculacea show the most ‘primitive’ struc-
ture (Oberling 1964), but it will be shown later in this paper that it represents an
early modification.

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL

523

TEXT-FIG. 3. Section through shell of Neopilina galalheae
Lemche. Traced from Erben et al. (1968, pi. 3). 1 = simple
prisms; 2 = nacre; 3 = myostracal prisms.

Other evidence of the ‘primitive’ nature of the simple prisms and nacre combina-
tion is seen in its occurrence in the Archaeogastropoda (Wise 1970; Erben 1971;
Taylor unpub.) in Nautilus and ammonites (Gregoire 1962; Mutvei 1964; Erben et
al. 1969). In these groups the simple prisms are not as well defined as in the Bivalvia
but their spherulitic nature is clear. The nacreous layers usually consist of columnar
nacre.

The ‘primitive’ shell structure may thus be fairly reasonably defined. Assuming
a monophyletic origin for the Bivalvia, the problem is how have all the other shell
structure combinations arisen from this ‘primitive’ combination? There is appar-
ently little similarity between the more advanced crossed-lamellar and complex

524

PALAEONTOLOGY, VOLUME 16

crossed-lamellar shells and the primitive nacreo-prismatic forms and after some
years’ consideration of the problem no link could be seen. The situation was changed
by the discovery of a metallurgical analogy which caused a reorientation of thought
resulting in the present tentative proposal of an evolutionary series of shell structure
combinations. This does not mean that the metallurgical analogy is suggested as
the mechanism for calcification in bivalves but merely that its consideration has
been instructive. A model need not be correct to be useful.

THE CELL-DENDRITE ANALOGY

During the solidification of impure melts in quiescent conditions microsegrega-
tions of the impurity may occur. A microstructure which often arises is that of
cellular structure, where at the growth surface most of the impurity segregates into

the walls of a polygonal cell structure
(text-fig. 4) (Chalmers 1958). In section
the cells are columnar and the impurity
appears as a thin line separating adja-
cent cells (Chadwick 1 967). This micro-
structure may arise if there is a zone of
constitutional supercooling (i.e. super-
cooling developed as a result of com-
positional changes in the liquid during
freezing) at the solid/liquid interface.
The usual explanation of the develop-
ment of the cellular structure (Chal-
mers 1958; Tiller 1963; Chadwick
1967) is that small irregularities of the
solid protruding into the supercooled
liquid grow faster than the surrounding
solid; the protuberances reject impuri-
ties in directions both normal to the
tip and laterally. If this process occurs
over the entire solid/liquid interface
then eventually a hexagonal structure
will be produced. The stages in the
development of the cell structure show a progression from a planar interface to a
‘pox’ structure, then elongate cells and finally regular polygonal cells (Tiller 1963).

If the speed of crystal growth is increased, the temperature gradient decreased
(Chalmers 1958) or the degree of constitutional supercooling increased (Chadwick
1967), then the cellular structure may break down into dendritic growth. The criteria
for dendritic growth are that the crystals should be branched or that the axis of the
growing domain should coincide with a crystallographic axis (Chalmers 1958).

It is uncertain how far this analogy can be taken with reference to shell micro-
structure, but there are obvious resemblances between this metallurgical example
and the microstructure of simple prisms (cells) and nacre (dendrites). There does

TEXT-FIG. 4. Cellular impurity structure as found
in metals. Traced from Chadwick (1967,
figs. 4-6(d)).

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL 525

not seem to be any reason why the precipitation of calcium carbonate with an
impurity (organic matrix) should not produce similar structures, perhaps for similar
reasons. The hypothesis put forward here is that the microstructure of simple prisms
and nacre corresponding to cells and dendrites originally arose spontaneously as
a consequence of the precipitation of calcium carbonate contemporaneously with
organic matrix under a certain set of physico-chemical conditions. Subsequently,
because of some selective advantage in this structural combination, perhaps strength
(Taylor and Layman 1972), this arrangement became stabilized. Further elabora-
tion of the depositional conditions resulted in the formation of the other shell
structures, such as foliated or crossed-lamellar structures, both of which appear to
have dendritic growth patterns. These are both in a position homologous with the
middle nacreous layers and could arise by changes in the dendritic growth patterns
of the nacreous structure. The elaboration of shell structures took place at different
rates and in different ways in various groups of bivalves. The details of these changes
are discussed below.

EVOLUTIONARY TRENDS IN SHELL STRUCTURE COMBINATIONS

The evidence for the evolution of the various shell layer combinations was obtained
by superimposing shell structure data upon a phylogeny derived from all available
characters and geological history (Taylor et al. 1973, fig. 33). It has been possible
in this way to demonstrate seven separate trends in shell structure evolution ; these
are shown in text-fig. 5. Examples of living taxa having a particular structural com-
bination in each series are indicated. The trends and examples are not phylogenetic
lineages but represent possible morphological grades arranged in order of increasing
advancement.

As previously stated some superfamilies are relatively little altered from the
ancestral condition; these include the Pholadomyacea, Unionacea, Trigonacea, some
Pandoracea, some Poromyacea, and the early shell of the Clavagellacea.

Trend 1

The first step in this sequence (text-fig. 5) was that the outer, simple prismatic
layer, aragonite in the ancestral condition, became calcite; this state is found today
in the superfamilies Pinnacea, Pteriacea, and in the extinct Ambonychiacea. Although
there has been much research into the calcite-aragonite problem in molluscs (Lowen-
stam 1954, 1964; Dodd 1963; Hall and Kennedy 1967; Kennedy et al. 1969) we still
know very little about how and why an organism can produce aragonite, or calcite,
or both, in the same shell. The temperature effect originally proposed by Lowen-
stam (1954) has so far only been successfully demonstrated in Mytilus (Dodd 1963);
other examples are of doubtful validity (Kennedy et al. 1969; Taylor et al. 1969).
The only generalizations one can make are that bivalves which normally employ
a calcitic outer layer are all epifaunal and the slightly lower solubility of calcite may
have some advantage in this situation. This is not to say that a temperature effect
does not exist but that it is non proven in most cases.

An early divergence of this trend may have been to the Mytilacea where the outer
calcitic prisms are very fine, needle-like, and inclined towards the shell margin.

526

PALAEONTOLOGY, VOLUME 16

A possible mode of formation of similar structures in mammalian enamel has been
proposed by Osborn (1970). A further development in this series was the loss of
the outer calcitic prismatic layer to leave the two underlying nacreous layers. This
condition is found in some tropical Mytilacea (Hudson 1968; Taylor et al. 1969)
and has been related to a temperature effect.

Another offshoot from the calcite prisms/nacre combination gave rise to the
sequence which includes the Pectinacea and Ostreacea. The probable first stage in
evolution was the transformation of the middle nacreous layer to foliated structure
which is calcite, i.e. one type of dendritic growth to another. The changes in micro-
structure accompanying this transformation may merely be a result of the minera-
logical change rather than any direct genetic effect upon shell structure. Although
the initial causes of the change are unknown, there are some mechanical properties
of the foliated shell, such as the resistance to fracture under impact, which may
have some selective advantage (Taylor and Layman 1972). It is uncertain whether
the condition found in oysters of calcite prisms (right valve only) and foliated struc-
ture, is more ‘primitive’ or ‘advanced’ than the combination found today in Pro-
peamussium of calcite prisms, foliated structure and a thin crossed-lamellar layer.
However, the combination found today in the Pectinacea, Limacea, and Anomiacea
of foliated structure and crossed-lamellar structure is a further development from

[ 1 ' f J

1

imiiiii/ii

IxljLiXi

most heterodon
families

1 jj:;;;''"'

rf

bypotheficol

Unionoceo
Trigonacea
Pholadomyaceo
Pandorocea ( port)

nocre

□

foliated

simple prisms composite cros^- complex homoqetseoos

prisms lamellar crossed-lorrvellar

myfilid

prisms

Mya granules

TEXT-FIG. 5. Diagram showing the postulated evolutionary radiation of shell layers into seven major trends
from the ancestral ‘primitive’ condition. The names in the trends represent living taxa having the particular
shell structure combination. The series do not necessarily imply a phylogeny.

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL 527

the ' Propeamussium' condition involving the complete loss of the outer prismatic
layer. In most of the Ostreacea the outer prismatic layer is found only as a very
thin layer on the upper valve; in the Pycnodontidae it is absent altogether. It seems
more likely that the oyster structure is a development from the ' Propeamussium’’
condition by the loss of the inner crossed-lamellar layer.

Trend 2

This is a simple case merely involving the loss of the aragonitic, outer, simple
prismatic layer, leaving nacreous structure as the outer layer. This condition is
found today in the oyster-like unionacean Etheria.

Trend 3

This sequence is again little changed from the ancestral condition and is found
today only in Solemya, in which the outer layer consists of simple aragonite prisms
and the inner of a very thin homogeneous layer. The prisms of the outer layer
are very irregular and have thick interprismatic walls. The prism outlines at the
growth surface are frequently elongate and irregular. The inner homogeneous layer
is very thin and is probably derived from the structural breakdown of nacreous
structure.

Trend 4

In this series the end members are represented today by the Cuspidaridae and
the Thracidae; in both of these families the shell consists almost entirely of homo-
geneous layers. It is fairly certain that both of these groups have descended from
a prismato-nacreous ancestor and certainly in the Cretaceous some ThraciaAxkc
bivalves have a prismato-nacreous shell. A possible transitional stage is seen in the
Recent Poromya granulata in which the outer layer is now homogeneous, but the
two inner layers are nacreous. Perhaps a last vestige of prismatic structure is seen in
Recent Thracia, where although most of the shell is homogeneous, the very outer-
most part of the outer layer shows spherulitic structures resembling the initial stages
of a prismatic layer (PI. 60, fig. 3).

Trend 5

In this sequence the simple prismatic layer changes to the composite prismatic
layer such as found in the Nuculacea. This change is apparently a fairly simple one
involving merely the development of a reflected mantle margin, causing the prisms
to become aligned parallel with the outside of the shell rather than normal to it
(text-fig. 5). This may have happened as a result of the development of the marginal
denticles characteristic of the Nuculacea, which probably allow better value closure
against predators. The Nuculanacea were probably derived from a Nucula-Wkt
ancestor, but at the present day they have a two-layered shell, both layers consisting
of homogeneous structure. However, a Jurassic nuculanid Ryderia graphiea has a
shell structure of eomposite prisms and nacreous structure similar to that of the
Nuculacea (Cox 1959; Taylor el al. 1969). The change probably consisted of the
loss of the outer layer and the breakdown of the two nacreous layers to homogeneous
structure.

528

PALAEONTOLOGY, VOLUME 16

Trend 6

In this sequence the first stage we see is that represented today by Panopea and the
evidence from this genus has been of key importance in the interpretation of the
evolution of shell layers. Panopea has an outer aragonitic simple prismatic layer
(albeit irregular), a middle homogeneous layer, and, within the pallial line, an inner
layer which is sometimes homogeneous but at other times may be built from
spherulite-like structures, resembling complex crossed-lamellar structure. A further
modification of this series along one offshoot is the loss of the outer prismatic layer
leaving a shell consisting of two homogeneous layers as in Hiatella and Panomya.

Another divergent trend resulted in the development of a middle crossed-lamellar
layer with the retention of the outer, simple prismatic layer; this condition is found
in some Pholadacea. The inner layer may consist of homogeneous or complex crossed-
lamellar structure. In other Pholadacea and the Gastrochaenacea, the outer simple
prismatic layer has been lost and the shell consists of crossed-lamellar and complex
crossed-lamellar layers.

In Mya, Platydon, and Zirfaea the outer layer consists of a grey, granular structure
which may be derived from a modification of the outer simple prismatic layer but
this is not certain. The other layers are as in the Pholadacea.

Trend 7

This sequence contains, with the exception of the Myoida, all of the rest of the
heterodont bivalves, which are by far the most numerous of Recent bivalves. How-
ever, more problems are posed in this sequence than in any other, the main difficulty
being that no transitional structures such as those of trend 6 have been found. Thus,
we are left with rather a large discontinuity between the ancestral structures and
what is considered the most ‘primitive’ structural combination in the sequence.
However, as possibly similar changes have taken place in trend 6, there is no reason
to suppose that they may not have taken place in this one. In accordance with the
sequence of events in trends 5 and 6 we may suppose that the three-layered shell
consisting of composite prisms, crossed-lamellar, and complex crossed-lamellar
structures might be the most ‘primitive’ condition now seen in this sequence. This
particular combination is found today in the Lucinacea, Tellinacea, and some Venera-
cea; of these the Lucinacea are an ancient superfamily and can be followed through
the Babinkacea back to the lower Ordovician (McAlester 1965). Included at this

EXPLANATION OF PLATE 60

Fig. 1 . Inner surface of the periostracum and edge of prismatic layer in Anodonta cygnea showing the initial
spherulites (arrowed) on the periostracum surface. The spherulites increase in size and eventually im-
pinge to form the polygonal outlines of prismatic structure. Scanning electron-micrograph, x 550.

Fig. 2. As Fig. 1, but detail of an individual spherulite showing the structure of fine radiating needles.
X 1300.

Fig. 3. Initial spherulites in the outermost part of the outer layer of Thracia phaseolina. This is all that
remains of prismatic structure in this family. Scanning electron-micrograph, x 1100.

Fig. 4. Polished, etched section of foliated structure of Pecten maximus, showing the dendritic nature of
the folia growth. Acetate peel, x 150.

PLATE 60

TAYLOR, structure of bivalve shell

530

PALAEONTOLOGY, VOLUME 16

level is a hypothetical structural combination similar to that described above, but
with an aragonite simple prismatic outer layer rather than a composite prismatic
one. This has never been found in any fossil or living bivalve, but the evidence from
trend 6 suggests that it may once have occurred in trend 7. In trend 5 the derivation
of composite prisms from simple prisms by changes in the shape of the shell margin
has been discussed.

As in the other trends the next stage in the evolutionary sequence from the most
primitive observed is the loss of the outer prismatic layer, whether it be composite
or simple, to form a two-layered shell consisting of crossed-lamellar and complex
crossed-lamellar structures. This structural combination is found in many hetero-
dont bivalve superfamilies such as the Carditacea, Crassatellacea, Chamacea,
Cardiacea, Mactracea, etc., and also in the Arcacea and the Limopsacea, both at
present classified in the Pteriomorphia. A further development is the breakdown of
the crossed-lamellar and complex crossed-lamellar structures to form homogeneous
structure. This may occur in one or both layers and many minor variations (not
illustrated here) are found, particularly in the Veneracea (Taylor et al. 1973, text-
fig. 18). A totally homogeneous shell is found in the Gaimardiacea and some Arctica-
cea {Arctica and Calyptogena).

DISCUSSION

In the evolution of the various structural series three main types of structural
change have probably taken place. These may be: (a) loss of layers, (b) orientation
changes, (c) complete structural changes.

(a) Loss of layers. This is the most commonly occurring of the structural trends and
occurs in all of the recognized sequences, the main consequences being the reduc-
tion in the number of shell layers from a ‘primitive’ three to a more ‘advanced’ two.
This usually involves the loss of the outer prismatic layer, whether formed from
composite or simple prisms. In some sequences inner layers may be lost, for example
in trend 1, but this is apparently associated with changes in the position of pallial
attachment to the shell.

{h) Orientation changes. The only major example of this type of change is the forma-
tion of composite prisms from simple prisms by changes in the shape of the shell
margin as discussed above.

(c) Complete structural changes. Other more major changes may arise as a result of
mineralogical transformation such as the change from aragonitic nacre to calcitic
foliated structure, as seen in trend 1. The differences in structure observed may be
merely a result of the fact that trigonal calcite will not produce the same crystalliza-
tion structures as orthorhombic aragonite, rather than there being a major change
in the secretory regimes. Both nacre and foliated structure are regarded as types of
dendritic growth form.

The origin of crossed-lamellar and complex crossed-lamellar structures and their
appearance in the various morphological trends is much more difficult and no sen-
sible explanation can be provided with present knowledge. What is certain is that
crossed-lamellar structure arose in a position homologous with the middle nacreous
layer of the ‘primitive’ condition. We have no surviving intermediate stages in the

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL 531

transformation, but crossed-lamellar structure appears to be a form of dendritic
growth and might be derived by an elaboration of nacre dendrites. Similarly, the
formation of complex crossed-lamellar structure is also not understood, but it has
evolved in a position homologous with the inner nacreous layer of primitive forms.
(Dr. Donald Boyd (University of Wyoming) has recently found what appear to be
transitional stages in Schizodus, in which a crossed-lamellar layer is present in the
middle of an otherwise prismato-nacreous shell.)

Various mechanisms could be invoked to explain the formation of crossed-lamellar
structure including: type and composition of the organic matrix, physico-chemical
conditions in the extra-pallial fluid, piezo-electric effects, or an alternating electric
charge at the valve margins. None of these suggestions has, as yet, any experimental
support. Workers adhering to the template theory of calcification would stress the
importance of compositional differences in the shell matrix as evidence.

One of the most widely recurring of the morphological changes is the apparent
breakdown of distinct crystalline morphology to form homogeneous structure. This
structure can arise from simple prisms, composite prisms, crossed-lamellar, complex
crossed-lamellar, and nacreous structures, with, in each case, a morphologically
similar result, that is a fine-grained, irregularly granular structure. The reasons for
this structural change are again not readily apparent, but speed of crystallization
may be responsible; growth may be more rapid in homogeneous shells.

One feature obvious from text-fig. 5 is the independent origin of crossed-lamellar
structure in several unrelated groups of bivalves ; it occurs in the Pectinacea-Limacea-
Anomiacea, the Myacea-Pholadacea-Gastrochaenacea, and thirdly in trend 7 where
it occurs in the Arcoida and nearly all families of Heterodonta. The independent
occurrence of this structure presumably arose by the evolution of similar depositional
conditions at the shell-mantle interface, resulting in convergence of depositional
morphologies. This convergence is not really all that surprising, for crossed-
lamellar structure has also been independently evolved in both the Gastropoda and
Scaphopoda.

In a study of various types of nacre, Wise (1970) has attempted to demonstrate
an evolutionary significance in his categories " Vertikalschichtung' , "Treppen, and
' Backsleinbau . The amount of vertical component in the nacreous structure of the
middle shell layers seems to be more closely related to the geometry of the shell,
rather than the antiquity of the lineage. Thus bivalves with a low expansion rate
(high convexity) will have better developed columnar nacre than forms with a
higher expansion rate.

Although this discussion has largely been concerned with the carbonate part of
the shell, the shell is of course a two-phased material, the other phase being the
organic matrix and both phases have evolved together. Ghiselin et al. (1967) and
Degens et al. (1967) have discussed the phylogeny of the amino-acid composition
of the shell matrix proteins in bivalves. Although their conclusions are based upon
very limited sampling of taxa, their groupings are generally similar to the shell
structure trends recognized here.

Many workers consider the shell matrix to have a very active role in molluscan
calcification but, although extensively studied (see Wilbur and Simkiss 1968), the
evidence is extremely ambiguous and often based upon preconceived ideas taken

532

PALAEONTOLOGY, VOLUME 16

from work on calcification in bone (Glimcher 1960). Recent ideas on the role of
the matrix have been well reviewed by Towe (1972). It may be that the matrix has a
more passive role in calcification, such as providing a quiet environment, free of
Brownian movement, in which crystal nucleation and growth can easily occur. It
has been argued that there is a correlation between shell structure type and the
composition of the matrix, thereby suggesting a direct control by the matrix on
structure (see review, Wilbur and Simkiss 1968). Although to some extent this corre-
lation is true, there are important exeeptions. For example, it has been shown by
Degens et al. (1967), that the shell matrix composition of the Arcaeea and Limo-
psacea which have crossed-lamellar shells is more closely similar to that of the
Pteriomorphia having foliated and prismato-nacreous shells, than to other taxa
having erossed-lamellar shells. Although it might be argued that different types of
matrix can produce the same result, the control would seem to lie in other factors.
The evolution of shell matrix composition could possibly be correlated with other
factors such as shell strength and thus have a relation with the mode of life and
shell structure.

HISTORY OF CALCIFICATION IN BIVALVES

Possibly the first calcification in the ancestral molluscs arose as a result of spon-
taneous precipitation into a mucoid coating lying between the mantle and an outer
tanned protein sheet, the periostracum. It is conceivable that those forms with a
thicker, more rigid calcified shell were selectively favoured, with a resultant evolu-
tion towards a much more heavily calcified shell. In this early state, it is probable
that the shell would show no organized structure but consist of an intergrowth of
crystals with no particular orientation. Structures such as this are formed by spon-
taneous precipitation in pulmonate egg shells (into a mucopolysaccharide matrix)
(Taylor, unpub.) and in some foraminifera (Towe and Cifelli 1967, pi. 98). Structures
almost identical to these can be produced inorganically by growing crystals in gels,
by methods similar to those of Henisch (1970). From this early state, how might
the most ‘primitive’ structure we have recognized, simple prisms and nacre in the
Monoplacophora, have arisen? By analogy with the metallurgical example described
above it is suggested that an increase in calcium carbonate saturation, an increase
in crystallization rate, or an increase in impurity (protein) could have collectively or
singly produced the spontaneous precipitation of the cellular, polygonal structure
of simple prisms. Further from the mantle margin an increased crystallization rate
or an increased saturation resulted in the breakdown of the eellular structure to
form laminar dendrites (nacre). This cell/dendrite combination may have originally
appeared spontaneously, but possibly because of some selective advantage conferred
upon the animal became fixed in the population as prisms and nacre. Further elabora-
tion of shell structures mainly concerned the modifications of dendritic growth in
the middle layer and the loss of the outer prisms. Thus both foliated and crossed-
lamellar structures are regarded as varieties of dendritic growth which have been
elaborated from nacreous structure. The mechanisms by which they have arisen are,
however, unknown.

The modifications of shell structure could have arisen spontaneously in a section

TAYLOR: STRUCTURAL EVOLUTION OF THE BIVALVE SHELL 533

of the population and subsequently become fixed in a lineage as a result of some
selective advantage. This advantage could be a more efficient calcification mechan-
ism involving closer control by the animal. The change from aragonitic to calcitic
prisms and from aragonitic nacre to calcitic foliated structure could have originally
arisen as a result of colder environmental temperatures, calcite being easier to
precipitate in colder water. Or perhaps slight biochemical differences in the extra-
pallial fluid may be sufficient to cause the precipitation of the more stable polymorph
(Wilbur and Simkiss 1968).

Although a relation of shell structure with phylogeny is evident, it has been shown
(Taylor and Layman 1972) that there is a strong correlation between structure and
the mode of life of the animal concerned. It can thus be argued that the shell struc-
ture combinations have been evolved in response to functional demands, although
the selective advantages conferred are in most cases unknown. In common with
other anatomical characters of the Bivalvia, shell structures exhibit a mosaic evolu-
tion. In different phylogentic lineages various shell structural combinations are
evolved at different rates presumably in response to increasing specialization to
various diverse modes of life.

Many of the conclusions reached in this paper are speculative, but in view of the
paucity of well-preserved shell structure material in the lower Palaeozoic we must
rely upon circumstantial or indirect evidence to reconstruct the original radiation
pattern of the shell structure combinations.

Acknowledgements. I am grateful to Dr. N. J. Morris for much useful discussion and critically reading
the manuscript. Dr. J. Milledge and Mr. P. Giles of the Department of Chemistry, University College
London, provided important information and discussion on crystal growth in gels and its application to
biological materials. Thanks are also due to Dr. W. J. Kennedy and Dr. A. C. Bishop for discussion.

REFERENCES

B0GGILD, o. B. 1930. The shell structure of the molluscs. K. danske. Vidensk. Selsk. Skr. Copenhagen, 2,
232-325.

CHADWICK, G. A. 1967. Constitutional supercooling and microsegregation. In zief, m. and wilcox, w. r.

(eds,.). Fractional solidification, 1, pp. 113-136. London.

CHALMERS, B. 1958. Growth of crystals of pure materials and of the solvents of solutions. In doremus, r. h.,
ROBERTS, B. w. and TURNBULL, D. (eds.). Growth and perfection of crystals, pp. 291-303. New York,
cox, L. R. 1959. The geological history of the Protobranchia and the dual origin of the taxodont Lamelli-
branchia. Proc. malac. Soc. London, 33, 200-209.

DEGENS, E. T., SPENCER, D. w. and PARKER, R. H. 1967. Palacobiochemistry of molluscan shell proteins.
Comp. Biochem. Physiol. 20, 553-579.

DODD, J. R. 1963. Palaeoecological implications of shell mineralogy in two pelecypod species. J. Geol.
71, 1-11.

ERBEN, H. K. 1971. Anorganische und organische schalenkomponenten bei Cittarium pica (L.). Biomineral-
isation, 3, 5 1 -60.

FLATS, G. and siEHL, A. 1968. Uber die schalenstruktur von Monoplacophoren. Akad. Wiss. Lit.

Mainz, 1968 (1), 1-24.

1969. Die friihontogenetische Entwicklung der Schalenstruktur ectocochleater Cephalo-

poden. Palaeontographica, A, 132, 1-54.

GHiSELiN, M. T., DEGENS, E. T., SPENCER, D. w. and PARKER, R. H. 1967. A phylogenetic survey of molluscan
shell proteins. Breviora, 262, 1-35.

GLiMCHER, M. J. 1960. Specificity of the molecular structure of organic matrices in mineralisation. Am.
Assoc. Adv. Sci. Publ. 64, 421-487.

GREGOiRE, c. 1962. On submicroscopic structure of Nautilus shell. Bull. Inst. Roy. Sci. Nat. Belg. 38, 1-71.

534

PALAEONTOLOGY, VOLUME 16

grigor’ev, d. p. 1965. Ontogeny of minerals. 250 pp. Jerusalem.

HALL, A. and KENNEDY, w. j. 1967. Aragonite in fossils. Proc. Roy. Soc. Loud. B, 168, 377-412.

HENISCH, H. K. 1970. Crystal growth in gels. 1 1 1 pp. Pennsylvania Univ. Press.

HUDSON, j. D. 1968. The microstructure and mineralogy of the shell of a Jurassic mytilid (Bivalvia). Palae-
ontology, 11, 163-182.

KENNEDY, w. J., TAYLOR, J. D. and HALL, A. 1969. Environmental and biological controls on bivalve shell
mineralogy. Biol. Rev. 44, 499-530.

LOWENSTAM, H. A. 1954. Factors affecting the aragonite : calcite ratios in carbonate-secreting marine
organisms. J. Geol. 62, 284-322.

1964. Coexisting calcites and aragonites from skeletal carbonates of marine organisms and their

strontium and magnesium contents. In : Recent researches in the fields of hydrosphere, atmosphere and
nuclear geochemistry, pp. 373-404. Tokyo.

MCALESTER, A. L. 1965. Systematics, affinities and life habits of Babinka, a transitional Ordovician lucinoid
bivalve. Palaeontology, 8, 231-246.

1966. Evolutionary and systematic implications of a transitional Ordovician lucinoid bivalve. Mala-

cologia, 3, 433-439.

MCCLINTOCK, c. 1967. Shell structure of patelloid and bellerophontid gastropods (Mollusca). Bull. Pea-
body Mus. nat. History, 22, 1-1 40.

MENZiES, R. J. 1968. New species of Neopilina of the Cambro-Devonian class Monoplacophora from the
Milne Edwards deep of the Peru-Chile trench, R/V Anton Bruun. In : Symposium on Mollusca, Marine
Biol. Assn. India, Mandapan Camp, 1, 19.

MUTVEi, H. 1964. On the shells of Nautilus and Spirula with notes on the secretions in non-cephalopod
molluscs. Ark. Zool. 16, 221-278.

NAKAHARA, H. and BEVELANDER, G. 1971. The formation and growth of the prismatic layer of Pinctada
radiata. Calc. Tiss. Res. 7, 31-45.

OBERLiNG, J. J. 1964. Observations on some structural features of the pelecypod shell. Mitt, naturf. Ges.
Bern, 20, 1-63.

OSBORN, J. w. 1970. The mechanism of prism formation in teeth: a hypothesis. Calc. Tiss. Res. 5, 115-132.

POJETA, J. 1971. Review of Ordovician Pelecypods. U.S. Geol. Survey Prof. Paper, 695, 1-46.

SCHMIDT, w. J. 1959. Bemerkungen zur Schalenstruktur von Neopilina galathea. Galathea Rept. 3, 73-77.

TAYLOR, J. D., KENNEDY, w. J. and HALL, A. 1969. The shell structure and mineralogy of the Bivalvia: Intro-
duction: Nuculacea-Trigonacea. Bull. Br. Mus. nat. Hist. Zool. suppl. 3, 1-125.

1973. The shell structure and mineralogy of the Bivalvia. Lucinacea-Clavagellacea,

Conclusions. Bull. Br. Mus. nat. Hist. Zool. 22, 253-294.

and LAYMAN, M. 1972. The mechanical properties of bivalve (Mollusca) shell structures. Palaeontology,

15, 73-87.

TILLER, w. A. 1963. Principles of solidification. In oilman, j. j. (ed.). The art and science of growing crystals,
pp. 276-312. New York.

TOWE, K. M. 1972. Invertebrate shell structure and the organic matrix concept. Biomineralization, 4, 1-14.

and ciFELLi, R. 1967. Wall ultrastructure in the calcareous foraminifera : crystallographic aspects

and a model for calcification. J. Paleont. 41, 742-762.

WADA, K. 1961. Crystal growth of molluscan shells. Bull. Nat. Pearl Res. Lab. 7, 703-828.

WAT ABE, N. and WILBUR, K. M. 1961. Studies of shell formation, ix. An electron microscope study of crystal
layer formation in the oyster. J. biophys. biochem. Cytol. 9, 761-772.

WILBUR, K. M. and simkiss, k. 1968. Calcified shells. In florkin, m. and stotz, e. h. (eds.). Comprehensive
biochemistry, 26A, 229-295.

WILLIAMS, A. 1968. A history of skeletal secretion among articulate brachiopods. Lethaia, 1, 268-287.

WISE, s. w. 1970. Microarchitecture and mode of formation of nacre (mother of pearl) in pelecypods,
gastropods and cephalopods. Eclog. geol. Helv. 63, 775-797.

1971. Shell ultrastructure of the taxodont pelecypod Anadara notabilis (Roding). Eclog. geol. Helv.

64, 1-12.

J. D. TAYLOR
Dept, of Zoology
Brit. Mus. (Nat. Hist.)
London SW7 5BD

Revised typescript received 30 August 1972

THE PROBLEMATICAL
PRECAMBRIAN FOSSIL CHUARIA

by TREVOR D. FORD and william j. breed

Abstract. Chuaria circularis from the late Precambrian of the Grand Canyon was regarded by Walcott as a pri-
mitive brachiopod. It has subsequently been referred to as an alga, a chitinous foraminiferid, a gastropod, a hyo-
lithid operculum, a trilobite egg, and an acritarch. It is here suggested that Chuaria is a compressed, unusually large
planktonic organism (generally 2 to 5 mm diameter). Chuaria wimani, Fermoria, and unnamed material from Canada,
Sweden, France, Siberia, India, Iran, and Australia show no systematic differences from C, circularis and are con-
sidered synonymous, Chuaria is compared with Leiosphaeridia and classified with this as an sphaeromorphid acritarch.
All recorded occurrences of Chuaria are in late Precambrian strata, less than 1000 m.y. old: it may be regarded as
a new stratigraphic index fossil.

Chuaria is a small carbonaceous disc-like fossil which has been found in a number
of regions of late Precambrian rocks. It has been assigned to both plant and animal
kingdoms and at different times to several phyla of the latter. Recent collections
from the type-locality in the Grand Canyon allow a more detailed examination and
have resulted in a more definite conclusion regarding its nature.

Chuaria appears to have first been noted in the Kwagunt Valley of the Grand
Canyon by White (in Powell’s 1876 monograph on the Uinta Mountains). Powell
regarded it as a primordial fossil like Lingulella and Obolella, and assigned the
sediments in the floor of the Grand Canyon and some of its tributary canyons to
the Silurian, doubtless using Silurian in the old Murchison sense as pre-Devonian
sediments. No name for the fossil was proposed by Powell at that time.

During the winter of 1882-1883, Walcott investigated what later became known
as the Chuar and Unkar Groups of Powell’s Grand Canyon Series (Walcott 1895).
He spent a considerable part of his time in a search for fossils from these rocks, and
‘but for the discovery of a small Discinoid shell, a couple of specimens of a Pteropod
allied to Hyolithes triangularis, and an obscure Stromatopora-like group of forms,
the two and one-half month’s search for fossils in these groups would have been
without result’ (Walcott 1883, p. 441). These same fossils were later noted by Walcott
(1886, p. 43) who wrote that in the ‘. . . Chuar strata the presence of a fauna is shown
by a minute Discinoid or Patelloid shell, a small Lingula-like shell, a species of
Hyolithes and a fragment of what appears to have been the pleural lobe of the seg-
ment of a trilobite belonging to a genus allied to the genera Olenellus, Olenoides,
or Paradoxides\

In a more definitive study of Precambrian life, Walcott described and figured the
shell-like fossil as Chuaria circularis (Walcott 1899, pp. 234-235, pi. 27, figs. 12 and
13; cf. PI. 61, fig. 1 herein) and referred it to the discinoid type of brachiopod. His
figures were drawings, not photographs, and whilst they could be interpreted as
horny brachiopods there is nothing in them diagnostic of this phylum. He also sug-
gested that they might be opercula of hyolithids. At the same time Walcott also
decided that the Hyolithes of 1 883 and the trilobite fragment of 1 886 were of inorganic
origin though he figured them, again by drawings. The Lingula-like shell of 1886

[Palaeontology, Vol. 16, Part 3, 1973, pp. 535-550, pis. 61-63.]

536

PALAEONTOLOGY, VOLUME 16

was not mentioned and so presumably was included within the specimens referred
to C. circularis. Walcott (1899, p. 235, pi. 27, fig. 9) also noted an enigmatic object
showing some similarity to the brachiopod Acrothele in a limestone 150 ft above
the shale containing Chuaria. The specimen (USNM 33801) has been re-examined
and could be a Chuaria, but it provides too little information to make any further
inquiry profitable. No further specimens have been found. It could just as easily
be a fragment of an oolith in the highly recrystallized and dolomitized oolitic lime-
stone.

Walcott’s type was listed in a catalogue of the U.S. National Museum type fossils
(Schuchert 1905). However, the taxon was not noted in most standard works of
fossils: it was not listed in the Zoological Record, nor mentioned in Zittel or in the
summaries of American fossils by Grabau and Shimer, or Shimer and Schrock.
The name does not appear in Walcott’s classic work on the Cambrian brachiopods.
Neave (1939) listed Chuaria and noted an assignment to Problematica.

In view of the above it is surprising to find that Wenz (1938, pp. 85-86) placed
Chuaria in a new family Chuariidae which was assigned, with some doubt, to the
superfamily Tryblidiacea (Gastropoda), specifically rejecting the possibility that it
might be an orbiculoid brachiopod. Later, Schindewolf (1956, p. 463) dismissed
Chuaria as inorganic, probably a concretion.

Chuaria has been mentioned in several volumes of the Treatise on Invertebrate
Paleontology but in each case only to mention that the genus did not belong to the
group concerned in that volume. In the Gastropoda volume (Knight et al. 1960,
p. I 324) Chuaria was listed as a generic name improperly regarded as Gastropoda
and Monoplacophora and was considered to be a ‘carbon scale’. In the Miscellanea
volume, Hantzschel (1962, p. W 232) followed Schindewolf’s view of 1956 and
classed Chuaria with ‘Fossils probably of inorganic origin’ and claimed it, without
citing supporting evidence, as ‘certainly inorganic’. The Brachiopoda volume
(Williams et al. 1965) surprisingly made no mention of Chuaria but listed the prob-
ably identical fossil Fermoria as a synonym of Protobolella, though this in turn was
regarded as a ‘generic name erroneously attributed to Brachiopoda’. In the Fora-
minifera volume (Loeblich and Tappan 1964, p. C 786) Chuaria was listed as a
‘generic name erroneously applied to Foraminiferida’.

In palaeobotanical literature, David White suggested (1928u, p. 389) that the
genus represented some sort of alga ‘named, though apparently not published, by
Doctor Walcott, Chuaria'. White had evidently missed the 1899 paper and as a
result Andrews (1955, p. 131) listed the genus with the wrong date. White (1928Z?)
also reported finding additional specimens in the upper division of the Chuar Group,
and he was the first to suggest that they were algae or at least alga-like. White’s
specimens have not been located. White also referred to a specimen from the Bass
Limestone (at the base of the underlying Unkar Group) but this has apparently not
been preserved and no further information is available. Recently both Glaessner
(1966) and Cloud (1968) noted Chuaria as an alga. Rowell (1971) did not mention
Chuaria in his recent review of Precambrian brachiopods, but Hofmann (1971)
referred Precambrian fossils from Canada to Walcott’s genus.

In a preliminary review of the situation Ford and Breed (1969, pp. 119-120) left
it uncertain whether Chuaria was ‘Chitinous Foraminifera or algal in nature’ and

FORD AND BREED: CHUARIA

537

required larger samples for further study. These samples have since been collected.
Ford and Breed (1972a) have described the stratigraphy of the Chuar Group, and
together with Mitchell (1972) have demonstrated a probable age of less than 1000 m.y.

COMPARATIVE FOSSILS

1. Chuar ia wimani Brotzen 1941

Wiman (1894) found discs similar to Chuaria circularis in the Late Precambrian
Visingso Formation of Sweden. These were figured but neither described nor named
by Wiman, and it was not until 1941 that Brotzen referred them to Walcott’s genus,
but, on account of their smaller size, erected a new species C. wimani. He regarded
them as chitinous foraminiferids. Eisenack (1951, p. 192), however, referred them
to Leiosphaera, a well-known early acritarch. {Leiosphaera was later emended to
Leiosphaeridia by Eisenack.) Subsequently Eisenack (1966) revised this opinion on
the basis of chemical tests and supported Brotzen’s interpretation. On the basis of
colour, composition, and wall-thickness Eisenack compared C. wimani with the
chitinous foraminiferid Archeochitina gotlandica from the Silurian Visby Marl of
Gotland. Meanwhile Timofeev (1960) had examined the type material by crushing
and dissolving it for a study of nanno-plankton, and later (1966) compared some of
the smaller ‘sporomorphs’ to those described from the Brioverian of France (Roblot
1964). Some of the fragments of pellicles of Laminarites described by Timofeev
(1960) could well be pieces of Chuaria, as noted by Eisenack (1966). Eisenack hinted
that the smaller objects might well be young C. wimani. Both Wiman and later
Regnell (1955, p. 555) considered a possibility that the fossils might be trilobite
eggs but this suggestion seems to have found little favour.

One of us (T. D. F.) has been able to study the remaining specimens of the type
C. wimani, mounted on three glass slides (PI. 62, figs. 2, 3, 5, 6). One of these is a
serially sectioned specimen which may be that prepared for Eisenack (1966, p. 52).
The findings are reported below in the description of Chuaria.

Timofeev (1970) has drawn attention to the existence of giant sphaeromorphid
microplankton, similar to C. wimani, which he has found whilst dissolving rock
samples from the Riphean (Upper Precambrian) of Siberia. No description has been
published, but he included photographs of two specimens of C. wimani from Sweden
renamed Kildinella magna.

Timofeev (1969, pi. 6, fig. 3) also figured and briefly described a further rather
indeterminate specimen from the Visingso Series as Trachysphaeridium vetterni sp.
nov., though Eisenack (1966, p. 53, fig. 1) had previously figured it as C. wimani.

2. Chuaria sp. Hofmann 1971

Small round to oval ‘brachiopod-like shells’ were found by Allan (1913, pp. 174,
192) in a 50 cm shale layer 16 m below the top of the Hector Formation (Late Pre-
cambrian) of Banff National Park in Canada. These show, very poorly, irregular
creases in the centre and more strongly developed concentric wrinkles around the
margin.

Hofmann (1971, p. 24, pi. 11, figs. 5-7) briefly described topotype material (GSC
types 24409, 24410) and referred them to Chuaria, though remaining uncommitted

H

538

PALAEONTOLOGY, VOLUME 16

as to their nature, ‘perhaps compressed planktonic spheroids, Foraminifera ... or
small medusoids similar to ones illustrated by Wade (1969)’.

Further specimens have been collected from the type locality and the present
writers support Hofmann’s assignment to Chuaria.

3. Fermoria Chapman 1935

The only other named fossil which seems to be comparable to Chuaria is Fermoria,
first described from the late Precambrian of India and more recently from Iran
(PI. 63, figs. 1,2).

Small carbonaceous disc-like fossils were found in the Suket Shales of the Vind-
hyan System of India by Jones (in Holland 1909, p. 66), who commented that they
might be compared with either Obolella or C. circularis. Other suggestions (see
Pascoe, 1959, p. 498) were that they belonged to Acrothele, known to occur in the
Cambrian of the Salt Ranges. However, Chapman (1935) assigned the specimens
to two new genera and four new species, Protoholella jonesi, Fermoria minima, F.
granulosa, and F. capsella.

Sahni (1936) thought that there was insufficient evidence for the separation of
these and placed them all in the synonymy of F. minima, though at the same time
erecting a new generic name Vindhyanella for one of the specimens figured as Proto-
bolella jonesi by Chapman (1935, pi. 2, fig. 1), though he admitted that the specimen
was lost!

In 1954 Sahni and Shrivastava briefly described and named a single, larger, new
fossil found with Fermoria as Krisimania acuminata. Their illustration (1954, fig. 4)
is entirely unconvincing regarding the filaments they claim to be attached, and the
writers support Glaessner (1962) in regarding it simply as a large Fermoria.

Misra and Dube (1952) recorded new material with Fermoria which they regarded
as mostly inorganic pellets. Misra (1957) restated this, noting that some alleged
Fermoria were chlorite aggregates in schist, and others were haematite spots in
sandstone. Misra’s plate 7, however, shows forms which could easily be badly pre-
served algal bodies like Chuaria.

Pascoe (1959, facing p. 498) figured specimens up to 4 mm diameter. He also
commented that Fermoria left a white ash when incinerated and was therefore a
plant, but at the same time he felt it possible that Fermoria could be an archaic form
of brachiopod though with ‘no reliable feature definitely attributable to this class’.

A few specimens of Fermoria from the Geological Survey of India collections have
been examined. They are from Neemuch, Madhya Pradesh (24° 24' north, 74° 54'
east), in the Vindhyan System. They occur either isolated or as small clusters of
smooth carbonaceous discs on fissile olive-coloured shale. Taking these in con-
junction with the various descriptions of other specimens, the writers have no
doubt that Fermoria should be regarded as synonymous with Chuaria.

Fermoria has also been found in Iran, apparently in large numbers at several
localities (PI. 63, figs. 1, 2). Assereto (1963, pp. 507-508, fig. 2) and Stocklin et al.
(1964, p. 14, pi. 1, figs. 3-5) have recorded Fermoria in the Chapoghlu Shales (late
Precambrian) of northern Iran. They figured specimens up to 3 mm diameter crowded
together.

A few specimens of Fermoria from Iran have been examined, and a number of

FORD AND BREED: CHUARIA

539

unpublished photographs by R. Assereto of other specimens have been available
for comparison. Though mostly lacking in carbonaceous matter, the impressions
on fine-grained olive-grey shale are so close to Chuaria as to leave no doubt that
here again, organisms identical to Chuaria were present. Whole surfaces of chips
of shale are covered with impressions, and clusters of at least fifty are indicated.
They are commonly 2-3 mm in diameter. There is little indication of overlap, but
concentric wrinkles are frequent particularly near the margins.

Recent interpretations of Fermoria have either been non-committal or that it is
algal. In the Treatise volume on Brachiopoda (Williams et al. 1965, p. H 864) Fermoria
is noted only as a synonym of Protobolella, which in turn is listed among the generic
names erroneously ascribed to Brachiopoda. Hantzschel (1962, p. W 240) listed
Fermoria amongst unrecognizable genera.

Glaessner (1966, p. 41) was non-committal and noted both Fermoria and Chuaria
under the heading of ‘other algae’, thus supporting Howell (1956, p. 1 10), who also
included Corycium enigmaticum in this group of uncertain algae. Ohlson (1961),
though, regarded the latter as mud-pellets armoured with aegagropilous algal debris.
Cloud (1968) also listed Fermoria as ‘possibly algal but needs restudy’.

4. Unnamed fossils

In describing medusoids from the Central Mt. Stuart Beds of the Central Aus-
tralian Late Precambrian, Wade (1969, p. 356, pi. 69, figs. 5-7) noted ‘numerous
minute unidentifiable organisms’ in maroon sandstones with minor shales. Latex
casts have been examined (PI. 63, fig. 3) and the impressions clearly show the con-
centric wrinkles characteristic of Chuaria, though they are somewhat larger, ranging
between 5 and 8 mm. As noted above, Hofmann (1971, p. 24) compared them with
the Canadian specimens of Chuaria.

SYSTEMATIC DESCRIPTION
Group ACRiTARCHA Evitt 1963

Subgroup SPHAEROMORPHITAE Downie, Evitt, and Sarjeant 1963
‘group’ MEGASPHAEROMORPHiDA Timofeev 1969
Family leiosphaeridae Eisenack 1959
Genus chuaria Walcott 1899

Diagnosis. Flattened carbonaceous spheroids, now discs, from 0-5 mm to 5 mm in
diameter, commonly 2-2-5 mm, showing wrinkles and cracks irregularly or con-
centrically arranged owing to crushing; no surface ornament; no pores; openings
restricted to gaps where spheroid burst open in a few specimens; translucent resinous
yellow in prepared specimens.

Chuaria cireularis Walcott 1899

Plates 61-63

1899 Chuaria circularia Walcott, pp. 234-235, pi. 27, figs. 12, 13.

1932 Neobolus minima Chapman, p. 29 (nom. nud.).

1933 Obolella jonesi Chapman, p. 20 (nom. nud.).

1933 Fermoria minima (Chapman), p. 20 (nom. nud.).

540

PALAEONTOLOGY, VOLUME 16

1933 Fermoria granulosa Chapman, p. 20 (nom. nud.).

1933 Fermoria tripartita Chapman, p. 20 (nom. nud.).

1935 Fermoria minima Chapman, pp. 114-116, pi. 1, figs. 1, 3.

1935 Fermoria granulosa Chapman, p. 116, pi. 1, figs. 2, 4; pi. 2, fig. 5.

1935 Fermoria capsella Chapman, p. 117, pi. 2, figs. 3, 4.

1935 Protobolella Jonesi Chapman, pp. 117-118, pi. 1, figs. 5, 6; pi. 2, fig. 1.

1936 Vindhyanella jonesi (Sahni), p. 467.

1941 Chuaria wimani Brotzen, p. 260.

1954 Krishnania acuminata Sahni and Shrivastava, p. 40, fig. 4.

1963 Problematica Assereto, pp. 502-503, fig. 2.

1969 minute unidentifiable organisms, Wade, p. 356, pi. 69, fig. 7.

1969 Kildinella magna Timofeev, p. 14, pi. 6, figs. 4, 5.

1970 Kildinella magna Timofeev, pi. 1, figs. A, B, D.

1971 Chuaria sp. Hofmann, p. 24, pi. 11, figs. 5-7.

Nomenclatorial notes. The name C. circularis was published by Walcott in 1899 and thus has Linnean
priority. All remaining names have been placed in synonymy as the present writers do not feel that the
known fossils show sufficient features for consistent diagnosis of separate species, let alone genera. Further-
more, there has been an element of doubt in that most writers have compared their material with C. circularis
and have distinguished it only on the basis of either size or on features which are mostly diagenetic. Rowell
(1971, pp. 72-73) has discussed the confused nomenclatorial history of Fermoria, and it need not be re-
peated except to note that he overlooked the fact that Chapman had introduced the names without diag-
noses in 1933, two years before the formal descriptions (Chapman 1935).

Species diagnosis. As for genus.

Type specimens. Walcott’s (1899) type material was catalogued under U.S. National Museum no. 33800
and consists of six flakes of shale, each with one or more specimens. One flake is unfossiliferous, and there
is also one small bottle that contains indeterminate fragments, again without any observable fossil. The
original of Walcott’s figure 13 cannot be identified in the collection. The original of his figure 12 is prob-
ably the specimen illustrated here as PI. 61, fig. 1; this specimen is now designated lectotype and is still
catalogued under USNM 33800. A number of specimens have been selected from the collections made
by Ford and Breed (1969, 1972) and these have been added to the U.S. National Museum reference col-
lection and are catalogued under USNM catalogue 36, no. 181859.

Type locality. See under Stratigraphic Occurrence below.

Dimensions. Individuals are commonly 2 to 2-5 mm in diameter. Walcott (1899, p. 234) recorded speci-
mens ranging from 2 to 5 mm diameter though the largest illustrated (pi. 27, fig. 12) is barely 3 mm.

The Indian specimens of "Fermoria’’ seen by the authors range between 2 and 3 mm though Chapman
(1935, p. 115) noted the largest as 4-5 mm diameter.

The Iranian specimens oi ""I Fermoria' according to Stocklin et al. (1964) ranged from 1 mm to ‘several
mm’, but the specimens seen by the authors range only between 2 and 3 mm. Assereto’s unpublished
photographs indicate specimens up to 6 mm being common though the scale on the photos raises some
doubts. Wade’s specimens from Central Australia range up to 8 mm whilst those from the Hector Forma-
tion only reach 4 mm. Brotzen (1941), Eisenack (1966), and Timofeev (1969) noted a size range from
0-5 mm upwards to about 2-5 mm and the last two have indicated that there is a continuous range down-
wards to much smaller sizes. Eisenack noted his smallest specimens as being only 62 p.m but expressed
the feeling that there should be still smaller juveniles. This extreme range down into the sizes normally

EXPLANATION OF PLATE 61

Figs. 1-7. Chuaria circularis from Awatubi Member, Kwagunt Formation, Chuar Group, late Precambrian,
of Nankoweap Butte, Grand Canyon, Arizona. 1, Lectotype, X 10, U.S. Nat. Museum, 33800. 2, x 25,
Univ. Leic. 49398b. 3, x25, Univ. Leic. 49375. 4, x25, Univ. Leic. 49495. 5, Small cluster, x25,
Univ. Leic. 49392. 6, Peel of small cluster, x40, Univ. Leic. 56744. 7, Peel showing infolded margins,
x25, Univ. Leic. 56743a.

PLATE 61

FORD and BREED, Chuaria

542

PALAEONTOLOGY, VOLUME 16

expected for nanno-plankton raises considerable difficulties both of nomenclature and definition, since
nanno-plankton of this general nature have been placed in a number of genera and species by palynolo-
gists, chiefly in Russia (see Downie 1967; Timofeev 1965, 1969, 1970). Surface ornament and textures
present in juveniles are not necessarily present in the mature forms, so that Chuaria may include adults
of several smaller species. The definition of Chuaria is thus arbitrarily restricted to forms larger than
0-5 mm. Roblot (1964) figured a ‘sporomorph’ 256 jiim in diameter from the Brioverian of Normandy
which could well be a small Chuaria.

Remarks. Topotype specimens of C. circularis occur as black carbonaceous discs
on the bedding laminae of dark blue-black shale. The discs may be solitary or in
clusters, of which the largest so far seen numbered twenty-three. Individuals in the
clusters never show overlap, though some lateral crushing may be seen. This is
taken to indicate that the individuals were spheroidal, or at least inflated discoidal,
when deposited, since spheroids do not normally pile up on top of one another.
The clusters do not suggest that there was any direct connection between individuals
but rather that they came to rest washed together in random fashion. The same
observation can be made regarding the specimens from Iran, where clusters of fifty
or more have been seen.

Chuaria shows no surface ornament, except wrinkles and cracks due to crushing.
Examination with the scanning electron microscope simply shows the grain size
of the enclosing sediments impressed on the fossil. Frequent mud-cracked and
ripple-marked surfaces indicate intermittent desiccation, and the wrinkles are thus
probably due to shrinkage on drying. They show no regular behaviour except for
the concentric wrinkle or wrinkles near the margin of many specimens. This is par-
ticularly well seen in the Iranian specimens. Serial sections of C. ‘‘wimanV show, as
Eisenack (1966) noted, two simple walls with a thickness of 50 to 70 |U,m. The pre-
servation of C. circularis from the Grand Canyon has not permitted successful
serial sections as yet. The walls of C. "wimanV meet about 0-25 mm from the margin,
giving the false impression of a narrow marginal flange.

A number of diagenetic effects have been observed on both C. circularis and
C. 'wimani', and they appear to have counterparts in some descriptions and photo-
graphs of 'Fermoria'. Most obvious is the effect of minute cubes of pyrite in the
shale, which give a granulose effect, presumably the cause of Chapman distinguish-
ing F. granulosa as a separate species. A solitary cube can be seen in some photo-
graphs of C. 'wimanV . Otherwise cracking and distortion seem to be the main effects.
Many C. circularis show radial cracks in the margin, commonly three or four but
sometimes many more. This may be the reason why Chapman distinguished F.
tripartita. Some specimens show openings in the young stages according to Eisenack,

EXPLANATION OF PLATE 62

Figs. 1-6. Chuaria circularis from the late Precambrian of the Grand Canyon and Sweden. 1, Peel with
marginal crushing simulating a flange, X 25, from Grand Canyon, Univ. Leic. 56745. 2, C. 'wimani',
separated and mounted, showing wrinkling of flanks, x 30, Visingso Series, Sweden. Univ. Uppsala
Visingso Colin. 3, C. "wimanP, separated and mounted, showing burst open appearance, x 30, Visingso
Series, Sweden ( Wiman 1894, pi. 5, fig. 2). Univ. Uppsala Visingso Colin. 1-9. 4, Peel, x25, from
the Grand Canyon. Univ. Leic. 56743b. 5, 6, C. 'wimani', two serial sections, showing thickened or
duplicated flange on one side, x 35, Visingso Series, Sweden. Univ. Uppsala Visingso Colin. 1-9.

PLATE 62

FORD and BREED, Chuuria

544

PALAEONTOLOGY, VOLUME 16

but the only one observed in the ‘mature’ specimens is quite clearly where the spheroid
burst open, either on initial crushing or perhaps during life (PI. 62, fig. 3).

The specimens from the Hector Formation show more relief than any of the
others, but this is also thought to be a diagenetic effect in the rather more indurated
argillite matrix. Small round slickensided marks are probably crushed gas bubbles
and indicate considerable compaction. Some Chuaria from the Hector are convex
and others concave on the same lamina and both show wrinkles due to shrinkage
or compaction. The impressions in the Australian material are sufficiently strongly
developed in fine-grained sandstone to indicate that the globular bodies were not
fully compressed when the sandstone was indurated.

Chemical composition. As Eisenack (1966) observed, the fossils are easily incinerated
and leave a white ash. Eisenack found this was silica in his specimens, but some clay
minerals and pyrite appear to be present in others. He regarded the silica, clay, and
pyrite as of diagenetic origin. The substance is insoluble in KOH, concentrated HCl,
HE, and H2SO4 even after heating. Cold concentrated HNO3 has no effect, but
boiling HNO3 bleaches the substance. Schulze solution bleaches the substance more
quickly, but can be completely destructive. Eisenack examined some specimens of
C. ‘‘wimanf for phosphate by chemical means, with negative results. Examination
of C. circularis from the Grand Canyon by electron microprobe by Mr. Jarosewich
in the Smithsonian Institution also failed to reveal phosphorus. The last test indicates
that Chuaria is not of brachiopod nature, whilst the remainder suggest only that it
is organic, largely carbon. Eisenack also noted pronounced shrinkage on treatment
with KOCl which, he claimed, distinguished Chuaria from pollen, spores, and
hystrichospheres (and acritarchs presumably), as it was a characteristic reaction of
fossil chitinous foraminiferids. However, this test is not considered valid by all paly-
nologists. Further tests by electron microprobe are thought to be pointless as they
would only determine elements that were likely to be present in the enclosing shale.

Biological affinities. As Eisenack (1966) and Timofeev (1970) have noted, there is
a size range from nanno-plankton up to 2 mm. Timofeev has applied the name
Kildinella (a sphaeromorphid acritarch) with a new trivial name magna (1969, 1970)
to a specimen previously named C. 'wimanf by Brotzen, without discussion of
reasons for so doing. He has further placed this genus in two separate ‘groups’ of
acritarchs, Sphaeromorphida and Megasphaeromorphida, without making it clear
what status his ‘groups’ have in relation to the Group Acritarcha of Evitt (1963)
and the Subgroup Sphaeromorphitae of Downie, Evitt, and Sarjeant (1963).

Assignment to higher taxa also presents problems. The ‘group’ Megasphaero-
morphida erected by Timofeev (1969) is little more than a convenient grouping
for large planktonic organisms, and seems to be broadly equivalent to superfamily
status. Alternatively the present authors feel that there is some merit in placing

EXPLANATION OF PLATE 63

Figs. 1-4. Chuaria circulari.s from the late Precambrian of Iran, Australia, and Grand Canyon. 1, 2,
From the Chapoghlu Shale, W. Elburz, Iran, x 10, Univ. Leic. 58123. 3, From Central Australia,
X 10, latex cast of University of Adelaide, Geology Department spec. F16472. 4, Cluster from the
Grand Canyon, x 10, Univ. Leic. 49398a.

PLATE 63

FORD and BREED, Chuaria

546

PALAEONTOLOGY, VOLUME 16

Chuaria with Leiosphaeridia in the family Leiosphaeridae. Comparison with the
descriptions of Leiosphaeridia and Tasmanites provided by Wall (1962) and by
Schopf (in Tschudy and Scott 1969) shows that Chuaria is similar to both but much
larger. Tasmanites, however, has a punctate wall and has been compared with the
modern Pachysphaera pelagica and Halosphaera minor of the Class Prasinophyceae,
Phylum Chlorophyta. Chuaria has no visible punctation, so that it is more appro-
priately referred to the family Leiosphaeridae of the Acritarcha.

All current writers (e.g. Glaessner 1966; Cloud 1969; Timofeev 1970) have noted
Chuaria as a fossil alga. The writers support this assignment and, following Glaessner
(1966) and Timofeev (1970), regard Chuaria as an unusually large acritarch-like
organism, or organisms, comparable with Leiosphaeridia.

It may be noted that many late Precambrian and Cambrian micro-plankton with
diameters from 0-1 to 0-25 mm have been recorded (e.g. Roblot 1964; Timofeev
1965, 1969, 1970). Perhaps only a few of these forms grew to the size of Chuaria.
Obviously more palaeopalynological research needs to be done on these rocks to
extract the full range of nanno-plankton as well as the larger forms. Downie (in
Ford and Breed 1969) provided preliminary notes on the nanno-plankton. In spite
of Chuaria being placed in a group separate from Pachysphaera, the observation
by Parke (in Wall 1962, p. 359) that the latter releases a flagellated stage raises the
possibility that Chuaria may have done so. Few specimens show any sign of an
opening but they may either have split equatorially and so show no opening whilst
lying in this plane, or they may have been immature when fossilized. One specimen
of C. "wimanV (PI. 62, fig. 3) has obviously split open but this may have been during
burial or extraction.

STRATIGRAPHIC OCCURRENCE

Walcott (1899, p. 234) noted that his specimens were collected 730 ft (219 m)
beneath the summit of the Chuar terrane in the Kwagunt Valley of Grand Canyon.
On referring to his measured section of the Chuar (1894, pp. 508-512) this appears
to be close to the Cherty Pisolite in the Walcott Member of the Kwagunt Formation
as defined by Ford and Breed (1972a, 1973), but Walcott was ambiguous in that the
measurement of 730 ft (219 m) could have been either by altitude or by stratigraphic
thickness. However, Ford and Breed (1969, 1972a, 1973) found Chuaria in shales
over a thickness of about 100 ft (30 m) on Nankoweap Butte, overlooking Kwagunt
Canyon, with the greatest abundance in two beds about 30 and 80 ft (9 and 24 m)
below the Flaky Dolomite, a horizon at the base of the Walcott Member and the
top of the Awatubi Member. Rare specimens were also found some 5000 ft (1500 m)
lower in the Chuar Group, in shales about 100 ft (30 m) below the top of the Tanner
Member. The cherty pisolite near the base of the Walcott Member has recently
yielded a flora of microscopic filamentous and spheroidal algae (Schopf, Ford, and
Breed 1973).

The age of the Chuar Group has been discussed by Ford et al. (1972) and by
Ford and Breed (1972a, 1973) and appears to be less than 1000 m.y., but definitely
Precambrian, i.e. Upper Riphean.

The Swedish Visingso Formation containing C. 'wimanC is now regarded as

FORD AND BREED: CHUARIA

547

belonging to the Varegian Formation, which is the younger part of the Eocambrian,
deposited less than 950 m.y. ago (Magnusson 1965).

The Iranian specimens come from the Chapoghlu Shale regarded by Stocklin et al.
( 1 964) as in the lower part of a series of Upper Precambrian to ? Lower Cambrian age.

The Hector Formation of Canada is unconformably covered by Cambrian and
rests unconformably on Beltian, and is thus part of the Windermere Series, recently
dated by Harrison and Peterman (1971) as between 570 and 850 m.y. Licari and
Cloud (1968) reported the discovery of nanno-plankton resembling the modern
green algal family Oocystaceae in these beds.

Timofeev (1970) noted that Megasphaeromorphida occurred in the Upper
Riphean of Eastern Siberia, and Shatsky (1952) also recorded Chuaria in the Upper
Riphean, though he gave no details.

The occurrences of Fermoria are more difficult to place owing to the confusion of
records of truly organic remains with those of inorganic substances. It must suffice
to say that the Vindhyan rocks are regarded by most writers on India as being of
late Precambrian age (Howell 1956; Pascoe 1959).

The Central Australian occurrence is in beds assigned to the topmost division of
the Upper Precambrian. The presence of numerous medusoids suggests a correla-
tion with the Ediacaran of South Australia, but none of the medusoids is common
to both localities and the Ediacaran fauna has not been found in association with
Chuaria elsewhere.

Sporomorphs up to 256 /xm diameter have been recorded from the Brioverian of
Normandy by Roblot (1964, pi. 11, fig. 12) which could well be a small Chuaria. The
Brioverian is generally regarded as late Precambrian.

Thus all known occurrences of Chuaria and fossils here regarded as synonymous,
are in late Precambrian rocks, broadly falling within the Upper Riphean division
of Precambrian time, though the lack of radiometric age data on most of the sedi-
ments concerned allows no placing more accurate than between 1 000 and 570 m.y. ago.

CONCLUSIONS

It is concluded that Chuaria is of plant origin, most probably being a large leio-
sphaerid acritarch. An arbitrary lower size limit is adopted of forms larger than
0-5 mm. They are generally preserved as flattened hollow spheroids, with cracks
and wrinkles owing to crushing and diagenesis. Forms previously named C. wimani,
Kildinella magna, and Fermoria minima are thought to be at present indistinguishable
from C. circularis. The stratigraphic range seems to be limited to the Upper Riphean,
roughly from 1000 m.y. ago to the beginning of the Cambrian. Occurrences are now
known in Arizona, Canada, Sweden, France, Siberia, Iran, India, and Australia,
and it seems clear that these carbonaceous spheroids provide a stratigraphic index
fossil for late Precambrian rocks.

Acknowledgements. Thanks are due to the Grand Canyon National Park authorities for their assistance
in many ways, and to Hatch River Expeditions and Arizona Helicopters Inc. for assistance in reaching
the field area. A grant from the Grand Canyon Natural History Association facilitated field work, and
one of us (T. D, F.) received a Fulbright Travel Grant and help from the University of Leicester Research
Board.

548

PALAEONTOLOGY, VOLUME 16

Thanks are also due to the following for their assistance in procuring specimens and photographs:
Dr. Ellis Yochelson (U.S. Geological Survey) for loan of specimens and supplying notes and photos of
type specimens; Professor Preston Cloud (University of California) for the gift of specimens from the
Hector Formation and for copies of Assereto’s photographs of Iranian material; Professor Martinsson
(University of Uppsala) for loan of Wiman’s specimens; the Director-General of the Geological Survey
of India in Calcutta for the gift of specimens of Fermoria \ Dr. Khadem, Managing Director of the Geo-
logical Survey of Iran, for the gift of Iranian Fermoria; to Professor Glaessner (University of Adelaide)
for supplying latex casts of Dr. M. Wade’s Australian material; and to Dr. W. S. Gussow for assistance
with details of the Hector Formation locality.

Mr. Michael Hayles (University of Leicester) has taken most of the photographs.

REFERENCES

ALLAN, J. A. 1913. Rocky Mountains: Bankhead to Golden. Geol. Surv. Canada Guidebook No. 8. Trans-
continental Excursion Cl, Pt. 2, 167-201.

ANDREWS, H. N. 1955. Index of generic names of fossil plants, 1820-1950. U.S. Geol. Surv. Bull. 1013, 262 pp.

ASSERETO, R. 1963. Palaeozoic Formations in Central Elburz, Iran. Riv. Ital. Palaeont. 69 (4), 503-543.

BROTZEN, F. 1941. Nagra bedrog till Visingsoformationens stratigrafi och Tektonik. Geol. Foren. Fork.
Stockholm, 63, 245-261.

CHAPMAN, F. 1932. In Palaeontological Dept. Report. Rec. Geol. Surv. India for 1931, 66, 29.

1933. In Palaeontological Dept. Report. Rec. Geol. Surv. India for 1932, 67, 20.

1935. Primitive fossils, possible atrematous and neotrematous brachiopods, from the Vindhyans

of India. Rec. Geol. Surv. India for 1935-1936, 69, 109-120.

CLOUD, p. E. 1968. Pre-metazoan evolution and the origins of the Metazoa. In drake, e. t. (ed.), Evolution
and Environment, pp. 1-72. New Haven, Yale Univ. Press.

DOWNiE, c. 1967. The geological history of the microplankton. Rev. Palaeobot. Palynol. 1, 269-281.

EViTT, w. R. and sarjeant, w. a. s. 1963. Dinoflagellates, hystrichospheres and the classification of

the acritarchs. Stanford Univ. Publns. Geol. Sci. 7, No. 3, 16 pp.

EiSENACK, A. 1951. Uber Hystrichosphaerideen und andere Kleinformen aus Baltischem Silur und Kam-
brian. Senckenbergiana, 32, 187-204.

1966. liber Chuaria wimani Brotzen. Neues Jahrbucb Geol. Paldont. Stuttgart, 1, 52-56.

EVITT, w. R. 1963. A discussion and proposals concerning fossil dinoflagellates, hystrichospheres and acri-
tarchs. Proc. Nat. Acad. Sci. 49, 158-164, 298-302.

FORD, T. D. and BREED, w. J. 1969. Preliminary Geologic Report of the Chuar Group, Grand Canyon, Ari-
zona. In BAARS, D. L. (ed.). Four Corners Geol. Soc. Grand Canyon Guidebook, 1 14-122.

1972u. The Chuar Group of the Proterozoic, Grand Canyon, Arizona. Proc. 24th Internat.

Geol. Cong., Montreal, Sect. 1, 3-10.

19726. The problematical Precambrian fossil Chuaria. Ibid. 11-18.

1973. The late Precambrian Chuar Group, Grand Canyon, Arizona. Bull. Geol. Soc. Amer.

84, 1243-1260.

and MITCHELL, J. G. 1972. Name and Age of the Upper Precambrian Basalts in the Eastern Grand

Canyon. Bull. Geol. Soc. Amer. 83, 223-226.

GLAESSNER, M. F. 1962. Precambrian fossils. Biol. Revs. 37, 464-494.

1966. Precambrian Palaeontology. Earth Sci. Rev. 1, 29-50.

HANTZSCHEL, w. 1962. Trace fossils and Problematica. In moore, r. c. (ed.). Treatise on Invert. Palaeont.,
Part W (Miscellanea), W 177-W 245. Univ. Kansas.

HARRISON, J. E. and PETERMAN, z. E. 1971. Windermere Rocks and their Correlation in the western United
States. Geol. Soc. Amer. Abstracts and Program for 1971, 592-593.

HOFMANN, H. J. 1971. Precambrian fossils, pseudofossils and problematica in Canada. Bull. Geol. Surv.
Canada, 189, 146 pp.

HOLLAND, T. H. 1909. General Report. Rec. Geol. Surv. India, 38, 66.

HOWELL, B. F. 1956. Evidence from fossils of the age of the Vindhyan System. Jour. Pal. Soc. India 1(1),
108-112.

FORD AND BREED: CHUARIA

549

KNIGHT, J. B, 1960. Mollusca 1. In moore, r. c. (ed.). Treatise on Invert. Palaeont., Part I. Univ. Kansas.
LiCARi, G. and CLOUD, p. E. 1968. Eucaryotic nannofossils in kerogen from the Pre-Paleozoic Windermere
Series of Alberta. Geol. Soc. Amer. Spec. Paper, 121 (Absts. for 1968), 174-175.

LOEBLICH, A. R., JR. and TAPPAN, H. 1964. Protista 2. In moore, r. c. (ed.). Treatise on Invert. Palaeont.,
Part C. Univ. Kansas.

MAGNUSSON, N. H. 1965. The Precambrian history of Sweden. Quart. Jl. Geol. Soc. London, 121, 1-29.
MiSRA, R. c. 1949. Organic remains from the Vindhyan (Precambrian). Current Sci. No. 18.

1957. Fermoria, the enigma of Indian palaeontology. Jour. Pal. Soc. India, 2, 54-57, pis. 7-8.

and DUBE, s. N. 1952. A new collection and a restudy of the organic remains from the Suket Shales

(Vindhyans), Rampura, Madhya Barat. Sci. and Culture, 18, 46-48.

NEAVE, s. A. 1939. Nomenclator Zoologicus, 1 (A-C), 957 pp. London, Zoo. Soc.

OHLSON, B. 1961. Observations on recent lake balls and ancient Corycium inclusions in Finland. C. R.
Soc. Geol. Finland, 33, 377-390.

PASCOE, E. H. 1959. A Manual of the Geology of India and Burma, 2 (3rd edn.), 1344 pp. Delhi.

POWELL, j. w. 1876. Report on the Geology of the Eastern Portion of the Uinta Mountains. U.S. Geol.
Survey, 218 pp.

REGNELL, G. 1955. Leiosphaera (Hystrichosph.) aus unterordovizischem Kalkstein in S. O. Schonen,
Schweden. Geol. Foren. Stockholm Fork. 77, 546-556.

ROBLOT, M. M. 1964. Sporomorphes du Precambrian Armoricain. Ann. Palaeont. 50 (2), 105-1 10, pis. 11-14.
ROWELL, A. J. 1971. Supposed Precambrian brachiopods. Smithsonian Contrib. Palaeobiology, 3, 71-79.
SAHNi, M. R. 1936. Fermoria minima : a revised classification of the organic remains from the Vindhyans
of India. Rec. Geol. Surv. India, 69, 458-468.

and SHRiVASTAVA, R. N. 1954. New Organic remains from the Vindhyan System and the probably

systematic position of Fermoria Chapman. Current Sci. 23, 39-41.

SCHINDEWOLF, o. H. 1956. Uber Prakambrische Fossilien. In lotze, f. (ed.), Geotekton. Sympos. zu Hans,
Stille, 455-480. Stuttgart.

SCHOPF, J. w., FORD, T. D. and BREED, w. J. 1973. Micro-organisms from the late Precambrian of the Grand
Canyon, Arizona. Science, 179, 1319-1321.

SCHUCHERT, c. 1905. Catalogue of the type specimens of fossil invertebrates in the Department of Geology,
U.S. National Museum. Bull. U.S. Nat. Mus. 53, 1-704.

SHATSKY, N. s. 1952. On the beds between the Palaeozoic and the Proterozoic on the Russian Platform.
Proc. Acad. Sci. U.S.S.R. Geol. Ser. 5, 36-49.

STOCKLiN, J., RUTTNER, A. and NABAVi, M. 1964. New Data on the Lower Palaeozoic and Precambrian of
North Iran. Geol. Surv. Iran, Rep. 1, 29 pp.

TIMOFEEV, B. V. 1960. Sur la characteristique micropalaeontologie de la formation de Visingso. Geol.
Foren. Fbrh. Stockholm, 82 (1), 28-42.

1965. Phyto-plankton of the late Proterozoic and early Palaeozoic seas. Palaeontological Conference,

Novosibirsk, 112-114. (In Russian.)

1966. Micropalaeontological investigations of the older series. Akad. Nauk. U.S.S.R. Lab. Geol.

Dokl. Izdat. Nauk. Moscow-Leningrad, 240 pp. (In Russian.)

1969. Proterozoic Sphaeromorphida. Acad. Nauk. U.S.S.R. Inst. Precamb. Geol. Geochron. Lenin-
grad, 146 pp., 39 pis. (In Russian.)

1970. Sphaeromorphida geants dans le Precambrian Avance. Rev. Palaeobot. Palynol. 10, 157-160.

TSCHUDY, R. H. and SCOTT, R. A. 1969. Aspects of Palynology, 510 pp. New York and London, Wiley.
WADE, M. 1969. Medusa from uppermost Precambrian or Cambrian sandstones. Central Australia. Palae-
ontology, 12, 351-365, pis. 68-69.

WALCOTT, c. D. 1883. Pre-Carboniferous strata in the Grand Canyon of the Colorado, Arizona. Amer.
J. Sci. 26, 437-442.

1886. Second contribution to the studies on the Cambrian faunas of North America. U.S. Geol.

Surv. Bull. 30, 369 pp., 31 pis.

1895. Precambrian igneous rocks of the Unkar Terrane, Grand Canyon of the Colorado, Arizona.

U.S. Geol. Surv. 14th Ann. Rep. (for 1894), 503-519.

1899. Precambrian Fossiliferous Formations. Geol. Soc. Am. Bull. 19, 199-244.

WALL, D. 1962. Evidence from recent plankton regarding the biological affinities of Tasmanites Newton
1875 and Leiosphaeridia Eisenack 1958. Geol. Mag. 94, 353-362.

550 PALAEONTOLOGY, VOLUME 16

WENZ, w. 1938. Gastropoda. In schindewolf, o. h. (ed.), Handbuch der Paldozoologie, 6 (1), 240 pp.
Berlin, Borntraeger.

WHITE, D. 1928a. Study of the fossil floras in the Grand Canyon. Carnegie Inst. Washington, Yearbook,
No. 27, 389-390.

19286. Algal deposits of Unkar Proterozoic Age in the Grand Canyon, Arizona. Proc. Nat. Acad.

Sci. 14, 597-600.

WILLIAMS, A. et al. 1965. Brachiopoda. In MOORE, R. c. (ed.). Treatise on Invert. Palaeont, Part H, 927 pp.
in 2 vols. Univ. Kansas.

wiMAN, c. 1894. Palaeolontische Notizen, 1 und 2. Bull. Inst. Geol. Univ. Uppsala, 2, 109-117.

T. D. FORD

Department of Geology
University of Leicester
Leicester LEI 7RH

W. J. BREED

Museum of Northern Arizona

Revised typescript received 6 December 1972 Flagstafif, Arizona, U.S.A.

Addendum. Since writing the above, the authors have heard that Dr. W. C. Gussow has prepared a note
on Chuaria from the Hector Formation, of Banff National Park, Canada. This will be published in Journal
of Palaeontology, 1973, 47, no. 6.

VISEAN TRILOBITES FROM
HOLWELL, SOMERSET

by GERHARD and renate hahn

Abstract. Three species of trilobites from the Clifton Down Limestone (Visean) at Holwell Quarry in the Mendip
Hills (Somerset) are described: Linguaphillipsia matthewsi sp. nov., Phillipsia (Phillipsia) holwellensis sp. nov., and
Cummingella jonesi jonesi (Portlock 1843). This is the first record of the genus Linguaphillipsia Stubblefield 1948
from the Carboniferous Limestone of western Europe.

The stratigraphical relationships and age of trilobites from the Carboniferous
Limestone facies of the Lower Carboniferous are still poorly known, certainly so
when compared with what is known about trilobites in the Culm facies (G. and R.
Hahn 1971).

A recently discovered fauna from low in the Clifton Down Limestone (Visean)
at Holwell Quarry, Somerset (ST 727 452), includes three species of trilobite, the
subject of this paper. Dr. S. C. Matthews (University of Bristol) is working on
other aspects of the rich fauna. The thirty trilobite specimens are well preserved
and partly silicified. They belong to three species: Linguaphillipsia matthewsi sp.
nov., Phillipsia (Phillipsia) holwellensis sp. nov., and Cummingella jonesi jonesi
(Portlock 1843). The majority of the specimens belong to L. matthewsi, five (pygidia
only) to P. (P.) holwellensis, one (a pygidium) to P. (P.) cf. holwellensis, and two
(one pygidium, one fragment of a free cheek) to C. jonesi jonesi.

The specific composition of the trilobite fauna is surprising as only one of the
three species, C. jonesi jonesi, belongs to a characteristic western European genus. L.
matthewsi is the first record of this genus in western Europe and specifically is closest
to L. ohmorensis (Ohkubo 1951) from Japan. Although Phillipsia (Phillipsia) is a
common subgenus in the Lower Carboniferous of western Europe P. (P.) holwellensis
is closest to P. (P.) conserrata Weber 1937 from the ? Visean of the southern Urals.

Acknowledgements. We are grateful to Dr. S. C. Matthews (University of Bristol) for making available
to us material originally discovered by Dr. N. H. Trewin (University of Aberdeen) and to Mr. R. G.
Godwin for the excellent photographs.

Repository. The material is deposited in the Geology Museum, University of Bristol, and bears the numbers
BU 21000-21030. Indeterminable fragments are unnumbered.

DESCRIPTIONS

Family proetidae Hawle and Corda 1847
Subfamily linguaphillipsiinae G. and R. Hahn 1972
Genus linguaphillipsia Stubblefield 1948

Type species. Linguaphillipsia terapaiensis SX.\xhh\efit\d 1948.

Diagnosis. Treatise, p. 401, and G. and R. Hahn 1972, p. 360 (‘Beziehungen’)-

[Palaeontology, Vol. 16, Part 3, 1973, pp. 551-561, pi. 64.]

552

PALAEONTOLOGY, VOLUME 16

Linguaphillipsia matthewsi sp. nov.

Plate 64, figs. 1-6; text-figs. 1-2
Derivation of name. After Dr. S. C. Matthews.

Holotype. Cephalon BU 21000; PI. 64, fig. 1 ; text-fig. 1.

Type locality. Holwell Quarry, Somerset, England (National Grid Reference ST 727 452).

Type horizon. Clifton Down Limestone, Visean.

Distribution. Known only from type locality and type horizon.

Paratypes. Four fragments of cranidia (BU 21001-21004), four free cheeks (BU 21005-21008), 2 frag-
mentary free cheeks (BU 21009-21010), 4 pygidia (BU 21019-21022), 7 thoracic segments, complete or
broken (BU 2101 1-21018).

Diagnosis. A species of Linguaphillipsia with the following characteristics. Glabella
long, partially overhanging the anterior border, markedly constricted between y-y,
between S-S only slightly broader than between )3-/S. Palpebral lobes situated pos-
teriorly, long facial suture without a straight portion between e-^. Eyes large, genal
spines medium in length. Pygidium slightly triangular in outline, with 17-18 rings
and 10-12 ribs on each pleura. Border of medium breadth, separated by a well-
impressed border furrow. Relief and sculpture (nodes) on the cephalon and pygidium
well marked.

Description. Cephalon (holotype, external mould, BU 21000). Lateral view (text-fig. Ifi). Glabella rising
above anterior border, continuously curved to a point near occipital furrow; occipital furrow well incised.
Occipital ring somewhat higher than glabella, reaching highest point in occipital node, situated near pos-
terior margin. Border vertical on cranidium, overlapped by glabella, ornamented along its total length
by 7-8 parallel lines which extend on to genal spine. Free cheek strongly vaulted, eye large, rising some-
what posteriorly.

Dorsal view (PI. 64, fig. 1 ; text-fig. la). Outline oval, only slightly broader than long (without genal spines).
Glabella long, tongue-like in shape, extending forward almost to anterior margin, markedly constricted
at y. Anterior part well rounded and arched, posterior part flatter and broader than anterior part, greatest
breadth between S-8. Glabellar furrows (lp-3p) deeply incised. Ip strongly curved backward, cutting
preoccipital lobes (LI) out of glabella, and extending to occipital furrow. 2p-3p short; 2p directed slightly
backward, but 3p directed slightly forward. Occipital furrow nearly straight (tr.), with only slight back-
ward curvature behind LI on each side. Occipital ring heavily arched in its centre, but immersed on each
side behind LI. Dorsal furrows deeply incised. Fixed cheek with very long, moderately diverging anterior

EXPLANATION OF PLATE 64

Figs. 1-6. Linguaphillipsia matthewsi sp. nov. 1, Cephalon, holotype, x5-9 (BU 21000). 2, Cranidium,
(broken during transport after photographing, therefore without number and scale). 3, Large pygidium,
x7-7 (BU 21020). 4, Free cheek, x6 0 (BU 21005). 4a, Dorsal view; 4b, Lateral view. 5, Large
pygidium, X 5-5 (BU 21021). 6, Medium sized pygidium, x4-8 (BU 21019).

Fig. 7. Cummingella jonesi jonesi (Portlock 1843). Pygidium, right pleural lobe (broken during transport
after photographing), x 5-2 (BU 21029).

Figs. 8-9. Phillipsia (Phillipsia) holwellensis sp. nov. 8, Pygidium, x2-9 (BU 21024). 9, Pygidium, right
pleural lobe (broken during transport after photographing), x 3-2 (BU 21025).

Fig. 10. Phillipsia {Phillipsia) cf holwellensis sp. nov. Pygidium, left pleural lobe (broken during transport
after photographing), x6-8 (BU 21028).

All specimens from Visean at Holwell, Somerset.

PLATE 64

HAHN, trilobites from Holwell

554

PALAEONTOLOGY, VOLUME 16

part (y-|3), long palpebral lobe, and short posterior part. |3 situated anteriorly, rounded; y, S, and e also
gently rounded, no straight portion evolved. Posterior part of fixed cheek short (exsag.), medium in
length (tr.). |S situated slightly inward of S. Border of the free cheek relatively broad, limited by a distinct
border furrow. Striations of the border only partly visible. Free cheek ascending rapidly from inner side
of border to eye; eye socket separated by a well-marked eye furrow. Eye large, strongly curved, lenses not
preserved. Posterior border of free cheek separated by deeply incised border furrow, which extends later-
ally to meet outer border furrow, then curves backwards into shallow spinal furrow. Genal spine of medium
length, sharply pointed, ornamented with some striations, divided longitudinally by spinal furrow into
two portions. Internal portion rounded, larger and more elevated than external portion, which diminishes
towards tip of spine. Surface of glabella covered by coarse nodes; nodes on occipital ring finer; only few
nodes on cheek region. Doublure as wide as border on free cheek; narrowed on the genal spine, slightly
broadened anteriorly. Front part occupied by rostral plate which lies exactly beneath the glabella. Doublure
covered with striations similar to those seen on the border in lateral view and dorsal view. Doublure of
occipital ring not preserved.

Measurements (in mm). Lengths: cephalon (without genal spine) 8 0, cephalon (with genal spines) 11-7,
glabella 6-5, |3-y 2-75, palpebral lobe (y-e) 2-9, e-w 1-25, eye 3-3; breadths: cephalon (at the base of the
genal spines) 11-2, cranidium between /3-|3 51, cranidium between 8-S 5-8, glabella between S-S 4-6.

Thoracic segment (BU 21011). Dorsal view (text-fig. 2a). Axis broad (tr.) and narrow (sag.), subdivided
into a short (sag.) anterior part that disappears laterally, and a longer (sag. and exsag.) posterior part,
which is separated from anterior part by distinct furrow; anterior part situated somewhat lower than
posterior portion. Articulating half-ring short, not as long (sag.) as anterior part of axis; axial furrow well
pronounced, extending laterally to meet furrow that divides axis. Dorsal furrows slightly incised. Inner
(adaxial) part of each pleuron situated horizontally, outer (abaxial) part bent downward (broken on BU
21011, but visible on BU 21012). Pleural furrow beginning at dorsal furrow, extending nearly to outer
border. Anterior flange well pronounced, axial process small, fulcral process not differentiated; posterior
flange not well separated from the pleura, axial socket and fulcral socket indistinct. Exact shape of lateral
pleural border not preserved in any of specimens at hand. Ventral view (text-fig. 2b). Posterior doublure
very prominent, covering nearly whole axis (sag.). Apodemata well evolved, node-like. No vertical rim
between inner and outer part of pleura as seen in Treatise, figs. 49b, 50.

Anterior view (text-fig. 2c). Axis strongly arched; apodemata relatively short, situated beneath the dorsal
furrow on each side. Inner part of pleuron horizontally disposed, outer part bent downward at an angle
of about 45°. Axial process on each side visible as a little node. Posterior view (text-fig. 2d). Arching of
axis as in anterior view. Posterior wall of the segment slightly grooved above the apodemata, grooves
continuing on each side of posterior wall of pleuron, visible ventrally also (see text-fig. 2b). Dorsal furrows
distinct, but nearly obsolete in anterior view. (As a whole the thoracic segments are similar to those of
Paladin (Paladin) helmsensis (Whittington 1954, pi. 3, figs. 7-16).) Number of thoracic segments in L.
matthewsi unknown (only isolated segments are present), but presumably totalling nine as in nearly all
Carboniferous Proetidae. Affiliation of the described thoracic segments to L. matthewsi not proven, but
probable because of rarity of all other trilobite species in the Holwell fauna in comparison with L. matthewsi.

Pygidium (BU 21019). Side view. Rhachis nearly as high as pleural lobe, curving gently back to last ring,
then falling down vertically to short, convex postaxial portion which is part of pygidial border. Rings
prominent, strongly arched, with steep posterior slopes. Pleural ribs also prominent, well separated by
pleural furrows. Anterior part more nearly horizontally disposed than behind rhachis. Posterior view.
Rhachis well arched, curved semicircularly. Inner half of pleura directed horizontally, outer half steeply
sloping downward. Anterior part of border sloping parallel to outer half of pleura.

Dorsal view (PI. 64, figs. 3, 5-6). Outline slightly triangular in shape, somewhat broader than long. Rhachis
long, nearly as broad as a pleural lobe, gently tapering posteriorly, with termination well rounded. 17-18
rings; rings 1-10 very prominent, last rings indistinct. Each ring divided (trans.) into a spinous central
portion and on each side a smooth portion (which covers the region where the nearly invisible impressions
of the pygidial muscles are situated). Central part ornamented with 5-7, slightly backwards directed, short
spines, one of which covers sagittal line. Axial furrows directed straight (tr.) centrally, showing slight

HAHN AND HAHN; VISEAN TRILOBITES

555

r':

' ‘ \ "iSi

\

1

pi

axs

ar

1b

TEXT-FIGS. 1-2. Linguaphillipsia matthewsi sp. nov. Visean, Holwell, Somerset, England.

1. Cephalon, holotype, BU 21000 (see PI. 64, fig. 1). a. Dorsal view; b. Lateral view.

2. Thoracic segment, BU 21011. a. Dorsal view; b, Ventral view; c. Anterior view; d. Posterior view.
Abbreviations, afl anterior flange; ahr articulating half-ring; ap apodeme; ar axial ring; axp axial process;
axs axial socket; db doublure; df dorsal furrow; af axial furrow; p pleura; pf pleural furrow; ts transversal

furrow, dividing axial ring in an anterior and a posterior portion.

backward curvature against dorsal furrow; anterior axial furrows well incised, posterior furrows only
slightly incised. Dorsal furrows distinct. Pleural lobe with 10 ribs and place for one more. Ribs well sepa-
rated by narrow, deeply incised pleural furrows. Rib furrows vestigial, visible as weak line on anterior ribs
only: on first rib grooved against border, separating clearly anterior and posterior branch of rib. Anterior
branch of ribs distinctly broader (exsag.) than posterior branch; ribs rounded in cross-section, ornamented
with small node at half-length (tr.). Ribs terminating at well-incised border furrow which separates rela-
tively narrow, convex curved border from remaining part of pleural regions. First rib furrow continuing
on to border. Terminal part of border in contact with posterior slope of rhachis. Connecting half-ring and
connecting half-ribs (sag., exsag.) narrow; connecting half-ribs marked by slight process which is directed
anteriorly and is situated somewhat inward of rib nodes. Doublure (visible in cross-section on left pleura)
as wide as border, pressed against it, ornamented with striations.

556

PALAEONTOLOGY, VOLUME 16

Measurements. See Table 1.

TABLE 1 . Measurements (in mm) of three pygidia of Linguaphillipsia matthewsi sp. nov.

Pygidium

Rhachis

Number

Number

Length

Breadth

Length

Breadth

of rings

of ribs

BU 21019

7-9

100

7-0

3-5

17

10( + 1)

BU 21020

5-75

6-75

5-2

2-6

16( + 1)

9(+l)

BU 21021

9-3

11-3

8-5

40

18

12

Variations. Among the free cheeks significant variation is found only in the structure of the border region.
In BU 21005 the border furrow is relatively shallow and the border itself only slightly arched (PI. 64, fig. 4),
but in BU 21007 and BU 21008 the border furrow is more deeply incised and the border well arched. These
two latter free cheeks are both smaller than BU 21005 (approximately half as long as BU 21005), so that
these differences of the border regions may be interpreted as having arisen during postlarval ontogeny.

Among the pygidia the number of rings and ribs differs only slightly (see Table 1^ The rib furrows are
somewhat better expressed in BU 21020 (PI. 64, fig. 3) and BU 21021 (PI. 64, fig. 5) than in BU 21019
(PI. 64, fig. 6). In all other respects the available pygidia are very similar, accentuating the specific charac-
teristics of L. matthewsi.

Discussion. Among the characteristic features of L. matthewsi the most important
one is the way in which the glabella encroaches on the anterior cephalic border. In
L. terapaiensis Stubblefield 1948, L. paczoltovicensis (Jarosz 1914) and related species
the glabella is separated from the anterior border by a deeply incised, prominent
border furrow; in L. silesiaca (Scupin 1900), which is closely related to L. tera-
paiensis, the glabella is lengthened anteriorly and presses against the anterior border,
but does not encroach on it. In L. longicornuta (Leyh 1897) and its allies the anterior
border is broad (sag.) and plane, not arched, and the border furrow is vestigial. In
these species the glabella also terminates at the border, and does not encroach on
it. Only in L. ohmorensis (Ohkubo 1951) from the lowermost Carboniferous of Japan
does the glabella encroach on the posterior part of the border in the manner seen
in L. matthewsi. As both these species are similar in most of their other features, it
seems probable that L. matthewsi might have descended from the older Japanese
species. The main differences between L. matthewsi and L. ohmorensis are found in
the structure of the pygidia ; in L. ohmorensis the pygidium is more elongate and its
border is broader. The number of rings and ribs is similar in the two species (17-18
rings and 10-12 ribs in L. matthewsi, 17 rings and 11 ribs in L. ohmorensis). Other
differences between the two species are of minor interest; they involve the more
slender glabella, the rather lesser arching of the cephalic border, the smoother
surface, lacking granulation, of L. ohmorensis. It is possible that some of these
differences are simply due to post-mortem influences, because in L. ohmorensis it
is the inner mould that is known and in L. matthewsi the outer.

The geographical distribution of L. ohmorensis and L. matthewsi is perplexing,
the one being found in Japan, the other in England. But Linguaphillipsia was a genus
typical of the Tethys region during the Lower Carboniferous and has been found in
nearly all parts of the Eurasiatic Tethys; in Austria, Turkey, Central Asia, SE. Asia,
Japan, and Australia. From this central pool it seems that species occasionally
invaded the border regions of Tethys: Germany, Poland, the Moscow Basin, and
the Urals. It may be merely an accident of preservation that one of the two related
species is found in the eastern part of the Tethys region, in Japan, and the other

HAHN AND HAHN: VISEAN TRILOBITES

557

near its western end, in England. Probably species closely related to L. ohmoremis
and L. matthewsi would have spread out through most parts of the Tethys, as was
the case with the L. terapaiensis-gvonp, whose distribution is known from Poland
to SE. Asia.

Subfamily phillipsiinae Oehlert 1886
Subgenus phillipsia (phillipsia) Portlock 1843

Type species. Phillipsia kellyi Portlock 1843.

Diagnosis. Osmolska 1970, p. 79, and G. and R. Hahn 1972, p. 391 (‘Beziehungen’).

Phillipsia {Phillipsia) holwellensis sp. nov.

Plate 64, figs. 8-9; text-fig. 3

Derivation of name. After Holwell Quarry, Mendip Hills, England.

Holotype. Pygidium BU 21023i; text-fig. 3.

Type locality. Holwell Quarry, in the Mendip Hills, Somerset, England.

Type horizon. Clifton Down Eimestone, Visean.

Distribution. Known only from the type locality and the type horizon.

Paratypes. 6 pygidia, partly broken, BU 210232 3, 21024-21027.

Diagnosis. A species of Phillipsia {Phillipsia) with the following characteristics:
pygidium with 13-14 rings and 11-12 ribs. Posterior ribs sharply curving backwards.
Posterior branches of the ribs suppressed, forming only the vertically directed
posterior slope of each rib. Border very narrow, covered by the terminal portions
of the ribs. End of rhachis peculiarly constructed (see below).

Description (holotype, BU 21023i). Side view (text-fig. hb). Rhachis not nearly as high as pleural lobes,
curving down gently in its posterior half; postaxial portion inclined at about 45°. Rings prominent, rhachis
furrows deeply incised. Pleural ribs also prominent, well separated by pleural furrows. Border horizontally
disposed.

Posterior view. Rhachis well arched in a nearly semicircular curve. Dorsal furrows deeply incised. Pleural
lobes transversely arranged where they first arise from dorsal furrow, then curving downwards very

2mm

TEXT-FIG. 3. Phillipsia {Phillipsia) holwellensis sp. nov. Visean, Holwell,
Somerset, England. Holotype pygidium, BU 21023i. a. Dorsal view;
b. Lateral view.

558 PALAEONTOLOGY, VOLUME 16

steeply in their outer parts but returning to horizontal disposition at border; anterior ribs sloping nearly
vertically.

Dorsal view (text-fig. 3a). Outline oval, nearly as broad as long. Rhachis long, relatively narrow, clearly
divided laterally into well-arched central portion and less well-arched lateral portions which cover region
of impressions of pygidial muscles. Number of rings 13; rings prominent, separated by deeply incised,
relatively broad (sag.) rhachis furrows; posterior rings also well differentiated. Terminal part of rhachis
blunt, peculiar in its construction: behind the last ring there follows a very broad and shallow furrow
that changes into a slight elevated rim at the very end of the rhachis. Postaxial portion separated from
the rhachis by a distinct step. Dorsal furrows well incised. Pleural lobes sub-divided by 12 ribs on each
side. Anterior ribs nearly transverse; in their direction posterior ribs directed clearly backwards. Rib
furrows suppressed, visible only on the anterior ribs, rather more clearly seen on BU 21024 and BU 21025
(PI. 64, figs. 8-9). Posterior branches of ribs confined to the vertically orientated posterior slope of each
rib. Pleural furrows deeply incised, broad (exsag.). Ribs continuing on very narrow border, which is
distinct from rest of pleural lobes only by its curving back into a horizontal position. Connecting half-
ring (broken medially in holotype) and connecting half-ribs narrow (sag., exsag.), projecting only slightly.
Surface of holotype nearly smooth, but on BU 21024 and BU 21025 with a row of very small nodules on
each ring and rib (PI. 64, figs. 8-9). Doublure (visible on BU 2\023^ narrow, with 7-8 soft striations.

Measurements (in mm). Length of pygidium 9-7, length of rhachis 8-5, breadth of pygidium 10 0, breadth
of rhachis 4 0.

Variations. See under Phillipsia (Phillipsia) sp., below.

TABLE 2. Comparison of pygidia of best-known species of Phillipsia (Phillipsia) Portlock 1843

Ph. (Ph.) gemmulifera (Phillips 1836)

Rings

17

Ribs

13

Post, branches
of ribs

Not suppressed

Extension
of ribs

Not on border

Ph. (Ph.) ornata Portlock 1843

19-21

16-18

,,

On border

Ph. (Ph.) truncatula (Phillips 1836)

18

16

Not on border

Ph. (Ph.) magnoculata Osmolska 1970

16

13

,,

,,

Ph. (Ph.) kellyi Portlock 1843

17

13

Suppressed

,,

Ph. (Ph.) moelleri Osmolska 1970

20

15

Ph. (Ph.) conserrata Weber 1937

16

10

On border

Ph. (Ph.) holwellensis sp. nov.

14

11-12

,,

Discussion. As shown by Table 2, Ph. {Ph.) holwellensis has a peculiar combination
of morphological features such as is not seen in any other species of Phillipsia {Phil-
lipsia). These features are: 1, the low number of rings and ribs; 2, the fully posteri-
orly directed posterior ribs; 3, the suppression of the posterior branches of the ribs;
4, the prolongation of the ribs on the border; and 5, the very weak sculpture.
A similar number of rings and ribs is seen only in certain Russian and Australian
species, e.g. Ph. {Ph.) conserrata Weber 1937 (16 rings, 10 ribs), Ph.l {Ph.l) dungo-
gensis Mitchell 1918 (14 rings, 12 ribs), and Ph. {Ph.l) rockhamptonensis Mitchell
1918 (12 rings, 8 ribs). The latter two Australian species differ markedly from Ph.
{Ph.) holwellensis in that the posterior branches of their ribs are not suppressed and
by the lesser backward curvature of their posterior ribs (as far as can be judged from
the photographs given by Mitchell). Rather more closely similar to Ph. {Ph.) holwel-
lensis is Ph. {Ph.) conserrata from the ?Visean of the southern Urals, which never-
theless differs in two characteristics: there are more rings, but fewer ribs, and the
surface is ornamented by a coarse granulation (see Osmolska 1970 and G. and R.
Hahn 1972).

HAHN AND HAHN: VISEAN TRILOBITES

559

Ph. (Ph.) holwellensis differs from all described species of Phillipsia (Phillipsia)
so far known in western Europe not only by its lesser number of rings and ribs and
its soft ornamentation, but also in its combination of suppressions of the posterior
branches of the ribs and prolongation of the ribs on to the border. Suppression of
the posterior branches of the ribs occurs in Ph. {Ph.) kellyi Portlock 1843 and Ph.
(Ph.) moelleri Osmolska 1970, but in neither of these species do the ribs invade the
border. On the other hand, Ph. (Ph.) ornata Portlock 1843, the ribs continue on to
the border, but the posterior branches of the ribs are not suppressed. Finally, in
Ph. (Ph.) gemmulifera (Phillips 1836), Ph. (Ph.) truncatula (Phillips 1836), and Ph.
(Ph.) magnoculata Osmolska 1970 the posterior branches of the ribs are not sup-
pressed, nor do the ribs invade the border. In both these respects they differ from
Ph. (Ph.) holwellensis.

Ph. (Ph.) holwellensis appears as an isolated occurrence in the Visean of England,
and is most nearly related to Ph. (Ph.) consenata from the southern Urals.

Phillipsia (Phillipsia) cf. holwellensis sp. nov.

Plate 64, fig. 10

Among the collection from Holwell there is one pygidium (BU 21028) which is
similar to Ph. (Ph.) holwellensis in many features, especially in its low number of
rings (13) and ribs (1 1) and in the peculiar structure of the terminal part of its rhachis,
but which differs in three respects: its outline is more rounded, its last ribs have a
lesser backward inclination, and (the most important) its border is somewhat ele-
vated and is not invaded by the ribs (see PI. 64, figs. 8-10).

Although this pygidium looks rather different from that of Ph. (Ph.) holwellensis,
it is probable that it belongs to this species. The elevation of the border may have
been due to post-mortem factors; it has developed to a relatively high degree on the
right pleural lobe but is not seen in the anterior part of the left pleural lobe. The
other differences may be interpreted as due to biological variation.

Subfamily cummingellinae G. and R. Hahn 1967
Genus cummingella Reed 1942

Type species. Phillipsia jonesii Portlock 1843.

Diagnosis. Treatise, p. 401, and G. and R. Hahn 1972, pp. 341-342 (‘Beziehungen’).

Cummingella jonesi jonesi (Portlock 1843)

Plate 64, fig. 7

*1843 Phillipsia jonesii Portlock, 308, pi. 11, figs. ha d.

1970 Cummingella jonesi jonesi, Osmolska, 55-56, pi. 5, figs. 3-4, text-figs. 5a, g, p, s.

1972 Cummingella jonesi jonesi, G. and R. Hahn, 348-351, 351 (with full synonymy).

Type, type locality, type horizon, distribution. See G. and R. Hahn 1972, pp. 350, 351.

Remarks. Among the Holwell trilobites are two specimens which can be referred
to Cummingella jonesi jonesi. One is a pygidium (BU 21029, PI. 64, fig. 7), originally
complete, now broken along the fracture visible on the rhachis in the photograph,
the other is the posterior part of a left free cheek, BU 21030.

560

PALAEONTOLOGY, VOLUME 16

Specimen BU 21029 compares very well with the pygidium of the lectotype figured
by Stubblefield (1952, pi. 1, figs, la-c) in the shape of the pygidium, the shape of
the rhachis, the number of rings (12 + ) and ribs (9), the breadth of the border and
its arching. As in C. jonesi jonesi only the four anterior rib furrows continue dis-
tinctly on the border, whereas in C. jonesi laticaudata (Woodward 1 884) (see Osmolska
1970, pi. 5, figs. 8-9) and in C. jonesi orleiensis Osmolska 1970 (see Osmolska 1970,
pi. 5, fig. 1) the posterior rib furrows also encroach on the border. This feature
determines the subspecific identity of the pygidium found at Holwell.

BU 21030 shows the posterior part of a free cheek, with an elevated lateral and
posterior border, both of which are separated from the rest of the cheek by well-
incised furrows. A genal spine is not present. The character of the free cheek is
therefore as is found in C. jonesi jonesi.

These specimens of C. jonesi jonesi, although they add no new morphological
information, do nevertheless contribute something to our stratigraphical know-
ledge of the subspecies. In all earlier recorded occurrences (including the type speci-
mens) the exact stratigraphical location has been unclear. Stubblefield (1952) cites
only ‘Carboniferous Limestone’, and Osmolska (1970, p. 55) states ‘(?Middle)
Visean’ without more precise information. The first indication of a more exact
stratigraphical attribution was given in G. and R. Hahn (1968), where specimens
from Heiligenhaus (Germany) were reported to be confined to cuIIS, Lower Visean.
The Holwell specimens seem to occur at a comparable stratigraphical horizon. It
is possible that C. jonesi jonesi was already extant in cullj8-y (see the discussion in
G. and R. Hahn 1968, pp. 441-442), but this is not yet confirmed, neither morpho-
logically nor stratigraphically. Occurrences younger than the cull/culll boundary
and genuinely referring to C. jonesi jonesi are not known to us.

REFERENCES

ENDO, R. and MATSUMOTO, E. 1962. Permo-Carboniferous trilobites from Japan. Sci. Rep. Saitama Univ.
Ser. B,4(2), 149-172, pis. 8-10.

HAHN, G. and R. 1968. Cummingella (Tril.) im mittel-europaischen Unter-Karbon. Senckenberg. leth.
49 (5-6), 439-463, 1 pi.

1971. Trilobiten. In Arbeitsgemeinschaft fiir Dinant-Stratigraphie. Die stratigraphische Gliederung

des Dinantiums und seiner Ablagerungen in Deutschland. News!. Stratigr. 1 (4), 7-18, 1 pi.

1972. Trilobitae carbonic! et permici III. Fossilium Catalogus. I. Animalia, 120, 335-531.

1973. Zur Evolution von Linquaphillipsia (Trilobita, Unter-Karbon). Senck. leth. 53, 479-514, 2 pis.

MITCHELL, J. 1918. The Carboniferous trilobites of Australia. Proc. Linn. Soc. N.S.W. 43, 437-494, pis.
46-53.

MOORE, R. c. (ed.). Treatise on Invertebrate Paleontology, Part O, Arthropoda 1 (Trilobitomorpha) 1959,
i-xix, 1-560. Geol. Soc. Amer. Univ. Kansas Press.

osmPlska, h. 1970. Revision of non-cyrtosymbolinid trilobites from the Tournaisian-Namurian of
Eurasia. Palaeont. Polon. 23, 1-165, pis. 1-22.

PORTLOCK, j. E. 1843. Report on the geology of the county of Londonderry, and of parts of Tyrone and
Fermanagh, i-xxxi, 1-784, pis. 1-38. Dublin.

STUBBLEFIELD, c. J. 1948. Carboniferous trilobites from Malaya. In muir-wood, h. m., Malayan Lower
Carboniferous fossils and their bearing on the Visean palaeogeography of Asia. Bull. Brit. Mus. (Nat.
//lit. ), 97-102, pis. 13 14.

HAHN AND HAHN; VISEAN TRILOBITES

561

STUBBLEFIELD, c. J. 1952. Proposed use of the Plenary Powers to vary the Type Species of the genus "Cum-
mingella Reed 1942 (class Trilobita) (Carboniferous). Bull. zool. Nomencl. 6, 150-154, 1 pi.

WEBER, V. N. 1937. Trilobites of the Carboniferous and Permian system of the U.S.S.R. 1. Carboniferous
trilobites. Monogr. Paleont. S.S.S.R. 71 (1), 1-159, 11 pis.

WHITTINGTON, H. B. 1954. Two silicified Carboniferous trilobites from West Texas. Smithson. Misc. Coll.
122(10), 1-16, pis. 1-3.

GERHARD and RENATE HAHN
Freie Universitat Berlin
1 Berlin 33
Altensteinstr. 34a

Revised typescript received 18 October 1972 Germany

Cl

<

Ci

1

i

c

Jk

SYMBIOTIC RELATIONSHIPS BETWEEN
ECTOPROCTS AND GASTROPODS, AND
ECTOPROCTS AND HERMIT CRABS
IN THE FRENCH JURASSIC

by T. j. PALMER and c. d. Hancock

Abstract. Certain gastropods from the Pierre Blanche de Langrune (Upper Bathonian) at Lion-sur-Mer, Calvados,
France, have been encrusted by a succession of ectoproct zoaria. After the death of the gastropods, the vacated
shells, with their encrustations, have been occupied by hermit crabs. Abrasion of the shell during locomotion of
the crabs produces flat areas near the shell aperture, and discontinuities in the ectoproct colony growth. The rela-
tionship between the ectoproct and the inhabitant of the shell is considered to be one of true symbiosis.

In the course of work by one of us (T.J.P.) on the palaeoecology of Upper Bathonian
faunas in England and northern France, collections were made from the upper
caillasse (= shell bed) within the Pierre Blanche de Langrune (Clydoniceras discus
Zone), exposed on the foreshore at Lion-sur-Mer, Calvados. The caillasse rests on
a hardground 9 m beneath the top of the Pierre Blanche. Among the more striking
objects to be found in this bed are colonies of encrusting ectoprocta, whose over-all
shapes approximate to that of a trochiform gastropod. These colonies are relatively
common, and it is a feature of all of them that part of the colony, usually next to the
aperture, is worn flat (PI. 65, fig. 1). A section through the colony confirms the opinion
that it has been built up by successive layers of ectoproct, of the genus Berenicea,
growing on a gastropod shell (probably Ataphrus sp.) (PI. 65, figs. 2, 3).

We postulate that the following sequence of events has given rise to the objects
as we find them in the caillasse;

1. In certain, but not all, of the colonies, colonization by the ectoproct has started
before the completion of growth of the gastropod. In these cases, the outer whorl
of the shell may be seen to enclose a layer or two of the ectoproct zoaria between
itself and the adjacent inner whorl (PI. 65, fig. 4). Whilst the gastropod is alive, the
shell is supported by the foot during locomotion. In this event, the whole of the
upper surface of the shell is available for colonization by the ectoproct, and none
of the shell is being dragged along the substrate and abraded. The first few layers
of zoaria thus cover the shell uniformly (PI. 65, fig. 3).

2. After the death of the gastropod, the shell is occupied by a hermit crab. No longer
being supported by a fleshy foot, it is dragged along the substrate. Any zoaria cover-
ing the area of contact with the substrate are worn off, whilst growing zoaria are
prevented from expanding to cover this area. Therefore the subsequent layers grow
asymmetrically around the shell, and the area in contact with the substrate remains
flat (PI. 65, figs. 3, 5). Even when the crab retracts into the shell, the shell rests on the
flat area, and is therefore not available as substrate to the spreading ectoproct.

3. The layers of zoaria continue to accrue whilst the shell is inhabited by the hermit
crab, or, more likely, a succession of hermit crabs. The crabs have therefore to main-

[Palaeontology, Vol. 16, Part 3, 1973, pp. 563-566, pi. 65.]

564

PALAEONTOLOGY, VOLUME 16

tain for themselves an opening from the shell to the outside. The genetic programme for
the shell aperture to be laid down in a trochospiral arrangement has been lost with the
death of the gastropod. Therefore the aperture maintained by the pagurid leads straight
to the outside (PI. 65, fig. 5), rather than spirally, as when the gastropod was alive.

4. The colony now grows only by addition of successive layers of ectoproct zoaria
over the surface, and not according to a trochospiral pattern as when the gastropod
was alive. These zoaria tend to round off the angulations of the trochiform shell,
and the shape of the colony changes from being dominantly conical to dominantly
ovoid. This effect is enhanced by the continued abrasion on one side (compare the
over-all shape of the colony in PI. 65, fig. 5 with that of the original gastropod).
This change in shape results in a change in the location of the area which is dragged
along the substrate, and consequently in the stable resting position of the colony
when the crab is withdrawn into the shell, from the base of the trochospire to the
long side of the ovoid, where the convexity is least. A section through the colony
shows how the flattened area migrates with successive layers of zoaria as the over-all
shape of the colony changes from coniform to ovoid (PI. 65, fig. 6).

DISCUSSION AND CONCLUSIONS

Of the twenty-six examples of the encrusted gastropods collected by the authors
from this bed (Oxford University Museum catalogue numbers J. 40001-40026),
all but one show the pagurid wear marks, and in none is the hole which allowed
emergence of the crab overgrown by ectoprocts. It would seem, then, that demand
for housing by pagurids was high, and that only very seldom did a dwelling remain
vacant for any significant period of time. Such a conclusion is in keeping with observa-
tions made independently by the two authors, in the Canary Islands (C.D.H.), and
off the Florida Keys (T.J.P.), that empty gastropod shells are virtually never found
on the sea bottom. Either they contain a gastropod, or they contain a hermit crab.
Considering that the crabs rely on being able to change their shells each time that

EXPLANATION OF PLATE 65

Fig. 1 . Colony of ectoproct zoaria encrusting a trochiform gastropod. The lower left comer of the colony
has been worn away (arrowed) due to abrasion against the substrate during locomotion of the hermit
crab which occupied the shell, OUM, J. 40001, x 2-2.

Fig. 2. Section through colony along plane sub-parallel to worn area, showing successive layers of ecto-
proct zoaria, OUM, J. 40002a, x 3 0.

Fig. 3. Section through colony cutting across worn area (arrowed). Encrustation of this specimen started
before death of the gastropod (see fig. 4). Consequently, the first few layers of zoaria at the point of
wear, are unabraded, OUM, J. 40003a, x2-8.

Fig. 4. Part of fig. 3 showing four layers of zoaria (arrowed) enclosed between the last whorl of the gastro-
pod and the preceding whorl. This indicates that colonization by the ectoproct started whilst the gastro-
pod was still alive, x 8 0.

Fig. 5. Section through colony showing the aperture (arrowed) maintained by the hermit crab. It leads
to the outside in a straight line, OUM, J. 40004a, x 3 0.

Fig. 6. Details of the change in location of the area of abrasion: the arrows point along the planes of con-
tact with the substratum, for successive growth stages of the colony. These planes of contact, where
abrasion occurred, change location as the over-all shape of the colony becomes less trochospiral, and
more ovoid, OUM, J. 40004b, x4-6.

PLATE 65

PALMER and HANCOCK, symbiotic relationships

566

PALAEONTOLOGY, VOLUME 16

they outgrow their previous one, it seems not unlikely that any change by a larger indi-
vidual is rapidly followed by the reoccupation of the recently vacated shell by a slightly
smaller individual, and so on until one very small shell is vacated and left vacant.

Associations between gastropods, encrusting organisms, and hermit crabs are
known from periods other than the Jurassic. Wear marks ascribed to hermit crabs
have been described on Recent gastropods encrusted by hydractinians (Schafer 1962).
Similarly worn, but unencrusted, gastropods have been described from the Pliocene of
Belgium by Boekschoten (1967), and an unworn example encrusted by the ectoproct
Cellepora sp., from the Pliocene of Britain, has been figured by Pinna (1972, p. 33).

Busk (1857, pi. 9, fig. 6c) also figures Cellepora edax Busk encrusting a gastro-
pod, and Wood (1872, p. 55) discusses the encrustation of Turritella crassicostata
by Edax. The Oxford University Museum Pleistocene collection contains five
specimens showing wear marks on gastropod shells encrusted by hydractinians
(OUM, Q. 1512-1516); there is also a Natiea sp. from the Coralline Crag (Pliocene),
which is encrusted by Cellepora edax and similarly worn.

Busk (loc. cit.) considers that Cellepora was parasitic upon the gastropod shell,
since it frequently appears, both in Pliocene and Recent examples, to have effected
the solution of the underlying shell. However, in these examples there is no evidence
that there was any occupant of the shell at the time during which the solution occurred,
and the inference of a parasitic relationship is not justified. In the Bathonian speci-
mens, the ectoproct colony was both active and growing during the life of the gastro-
pod, as well as during the subsequent occupation of the shell by the hermit crab. In
this case, there was no destruction of the shell by the ectoproct, and the relationship
between gastropod and ectoproct would appear to have been one of true symbiosis.
The shell provides a stable substrate, and the behaviour of the gastropod prevents it
constantly being rolled around and abraded by current activity. In turn, the ectoproct
offers the gastropod protection by disguise, and it also strengthens the shell against
attack by mollusc-eating vertebrates and predatory decapods. This reasoning applies
equally if the shell is occupied by a hermit crab; in this case, however, the ectoproct
probably gains further by gathering food particles released by the scavenging be-
haviour of the crab, as well as those suspended in the crab’s respiratory currents.

Acknowledgements. The authors would like to thank W. J. Kennedy, A. Hallam, and F. Fiirsich for their
useful comments, and J. R. McAvoy, S. M. Baker, and Miss Helen Birch for technical assistance.

REFERENCES

BOEKSCHOTEN, G. J. 1967. Palaeoecology of some Mollusca from the Tielrode Sands (Pliocene, Belgium).

Palaeogeog. Palaeoclim. Palaeoecol. 3, 311-362.

BUSK, G. 1857. A Monograph of the Fossil Polyzoa of the Crag. Palaeontogr. Soc. [Monogr.].

PINNA, G. 1972. The Dawn of Life. London.

SCHAFER, w. 1962. Aktuopaldontologie nach Studien in der Nordsee. Frankfurt, Kramer.

WOOD, s. V. 1872. Supplement to the Mollusca from the Crag. Palaeontogr. Soc. [Monogr.].

T. J. PALMER C. D. HANCOCK
Department of Geology and Mineralogy
University of Oxford, Parks Road

Typescript received 1 December 1972 Oxford 0X1 3PR

Note added in press. Further discussion, particularly on the identity of the ectoproct, may be found in:
Fischer, J.-C. and Buge, E. 1970. Atractosoecia incrustans (d’Orbigny) (Bryozoa Cyclostomata) espece
bathonienne symbiotique d’un Pagure. Bull. Soc. geol. France, 12, 126-133.

PALYNOLOGIC CORRELATION OF THE
DORSET ‘WEALDEN’

by N. F. HUGHES and c. a. croxton

Abstract. Using the Cicatricosisporites group of palynomorphs, events raised from ten selected samples from
the ‘Wealden’ of Worbarrow Bay, Dorset, are bracket-correlated with events of similar nature from the Warlingham
Borehole, Surrey. Although sedimentation rates appear to have differed, deposition at Worbarrow seems to have
continued throughout much of Berriasian to Aptian time as at Warlingham. The constituent data of the events
comprised graded comparison records with the use of twelve new biorecords and some of those published by Hughes
and Moody-Stuart (1969).

In this paper we attempt to correlate by means of a selection of palynologic data,
a section of about 1400 ft (425 m) of ‘Wealden’ strata exposed in the cliffs of Wor-
barrow Bay, Dorset. These beds overlie conformably some slightly dubious ‘Upper
Purbeck’ beds north of Worbarrow Tout (Arkell 1947, and earlier authors) pre-
sumably of Berriasian age; they are overlain more or less conformably by ‘Lower
Greensand’, with marine bivalves, which is presumably of Aptian age. These ‘Weal-
den’ strata may therefore be of any age from Early Berriasian to Late Aptian.

The attempted correlations are with the Wealden section of the Warlingham
Borehole, Surrey, which is itself not yet firmly correlated with an international
scale. The correlations are therefore not scale-dated as it would be premature to
do this; it should, however, be possible to date Worbarrow automatically as soon
as dates for parts of the Warlingham reference are agreed.

The appropriate rock samples and preparations have been deposited in the
Sedgwick Museum, Cambridge.

Method. The method is explained in Hughes and Moody-Stuart (1969, pp. 86-87)
and correlation is based on palynologic events raised from rock samples, in this
case restricted to palynomorphs of the Cicatricosisporites group. The events are
composed of graded comparison records (Hughes and Moody-Stuart 1967), based
on taxa described as biorecords which in this case all come from Warlingham or
Worbarrow rocks. Some previously described events and biorecords are taken from
Hughes and Moody-Stuart (1969); the numerous new events and their constituent
comparison records, plus twelve new biorecords, are presented systematically in a
condensed tabular form. All these taxa are fully employed in the stratigraphic
correlation; those not so required are omitted from the paper.

The handling of taxa is as described in Hughes (1971). No comparison is made
with published Linnean taxa as there is no stratigraphic purpose in doing so. Com-
parison records and biorecords may be reassembled subsequently into taxa under
the Rules of Botanical Nomenclature (Stafleu et al. 1972) if required for some gross
palaeoecologic synthesis.

Notation. Each biorecord bears a unique number which may be quoted with author
initials and date if referred to outside the paper. The accompanying letter and num-
ber in italics is the observer’s working identifier but is subsequently a non-search item.

[Palaeontology, Vol. 16, Part 3, 1973, pp. 567-601, pis. 66-77.]

568

PALAEONTOLOGY, VOLUME 16

Each event is numbered in a similar way with the working sample (field) number
in italics as a non-search item; several events of different taxal origin may be raised
from one sample. Comparison records may be uniquely referred to either as 105
event W128 cf. 6 B5, or as 105 event cf. 6 cicatr.

SYSTEMATIC DESCRIPTIONS OF BIORECORDS

\

Description common to all biorecords below. All trilete miospores; amb shape,
equatorial shape, general distribution of muri, and mural profile may be taken from
photographs. Measurements and other data are given on Tables 1 and 2, diagram-
matic mural profiles to scale on text-fig. 1. Lips are simple, low, membranous unless
otherwise stated. The ratio of radial to interradial exine thickness is given on Table 2,
and no further reference is made to it in descriptions. On Table 2 measurements of
exine include the murus in spores of negative sculpture, and exclude it in those of
positive sculpture unless otherwise stated.

No comment is made under preservation concerning the frequent folding of thin-
walled spores.

For preparation details see the appropriate event preparation Table 5, and for
sample sediment details see Appendix.

17 CICATR B20

Plate 66; text-fig. 1

Description. Laesura may be sinuous. Proximal face sometimes shows a small
smooth triangle or reduced muri (figs. 1, 3, 5). Proximal muri: three interradial sets
of approximately 4. Polar view: 0-4 muri (in profile) cross radial margin (figs. 1, 4).
Distally three sets of 4-12 muri form an asymmetrical pattern (fig. 9).

Preservation. 21% torn, often radially (figs. 6, 7).

Local distinction. 25 cicatr B21 is larger and has more and narrower muri which are more closely spaced
and have the characteristic ‘swirling’ pattern. 28 cicatr DG is larger with a thicker exine and negative
sculpture.

EXPLANATION OF PLATE 66

Magnification, x 1000.

Figs. 1 12. Biorecord 17 CICATR Slide KOI 8/7. 1, Proximal aspect; OR 28-7 1 17-8. 2, Distal aspect,
low focus; OR 291 117-9. 3, Proximal aspect; OR 35 0 125-3. 4, Proximal aspect; OR 49-7 110-0.
5, Proximal aspect; OR 54-4 1 18-0. 6, Distal aspect, low focus; OR 37-4 1 10-6. 7, Distal aspect, low
focus; OR 53-4 123-6. 8, Oblique aspect; OR 30-0 121-3. 9, Proximal aspect, low focus; OR 41-1
127-3. 10, Equatorial aspect; OR 35-0 119-0. 11, Equatorial aspect; OR 38-3 116-4. 12, Oblique
aspect; OR 39-6 126-4.

PLATE 66

HUGHES and CROXTON, biorecord 17 CICATR B20

570 PALAEONTOLOGY, VOLUME 16

TABLE 1 . Sample, preparation, and diameter information for twelve biorecords.

CICATR

Biorecord

Record

Sample

Preparations

Diameter
100 specs.

/Um

Factors possibly
influencing measurem’ts

Aspect %

Fern spore
size index

Limits8iMean

O’

Pol. Equ.Obl

<30 30-50 >50

17. B20

WM 1655

KOI8/2,3,5,6,7

(22) 31-4 (44)

4-5

54 22 24

13 67 20

18. C3

WM 1749/8

V 411/6,7,8,9

(30)43-0(61)

5-8

84 10 6

26 47 27

19. A6

WM 1681/6

V 500/4,5,6,7

(27)43-9(63)

6-6

45 30 25

10 61 29

20. DD

WM 1415/3

V 963/4

(39)60-8 (92)

10-6

58 23 19

18 51 31

21. C4

WM 1415/3

V963/2-4.WI03/I-3

(32)45-9(58)

5-3

77 18 5

18 51 31

22.DB

WM 1415/3

(36) 59-0 (80)

10-5

72 19 9

18 51 31

23.DCE

WM 1415/3

V 963/2-4. W 103^3

(47)70-1 (95)

10-4

75 13 12

18 51 31

24. C 5

W III

W 190/1,2,3,4,5,6,7

(28)43-1 (58)

63

73 21 6

29 61 10

25.B2I

Will

W 190/1,2,3,4,5,6

(25) 38-4(55)

7-1

49 34 17

29 61 10

26. AST

W 9

W 197/4,5,6

(31) 40-5(55)

5-3

56 26 18

16 63 21

27. C6

W 14

V 198/1-4. W262/1, 3

(28) 42- 1 (63)

6-2

54 39 7

19 43 38

28. DG

WM 1217/6

W058/l,2,3,8,7

(28) 37-2(50)

4-6

66 22 12

23 51 26

18 CICATR O
Plate 67 ; text-fig. 1

Description. Width of one lumen (0-8) 1-7 /xm (4-5) (86). Proximal muri: three inter-
radial sets of 1 or 2. Distal mural pattern either three sets of 2 (occasionally 3) muri
forming a central tri-radiate lumen (figs. 4, 6) or a set of sub-parallel muri (fig. 7).
Radial equatorial features are extensions of the coalesced outer muri from adjacent
interradial sets ; ratio length/width at half-length : (0-6) 1 -2 (2-7) (82) ; some are parallel-
sided and others cone-shaped. These features may not extend beyond the periphery
of the amb (fig. 6).

Preservation. Characteristically split immediately adjacent to the radial equatorial feature. Corrosion of
muri takes the form of cross striations (fig. 9).

Local distinction. 1 cicatr Cl has narrower and more numerous muri, a thinner exine, and positive sculp-
ture. 20 CICATR DD is larger and has a radial lumen. 27 cicatr C6 has narrower muri with a different mural
profile and variable radial equatorial extensions.

explanation of plate 67
Magnification of figs. 1-9, x 1000; fig. 10, x2000.

Figs. 1-10. Biorecord 18 cicatr C3. 1, Proximal aspect; V411/6, OR 56-5 117 0. 2, Proximal aspect;
V411/6, OR 50 0 119-5. 3, 4, Distal aspect, low and high focus; V411/6, OR 35-0 1210. 5, Proximal
aspect; V411/7, OR 28-8 123-5. 6, Distal aspect, low focus; V411/6, OR 25-0 124-0. 7, Distal aspect;
V41 1/8, OR 26-5 126-0. 8, Equatorial aspect; V411/7, OR 43-1 109-9. 9, Distal aspect; V411/8, OR
24-7 1 19-9. 10, Part of oblique aspect, showing mural profile; V41 1/7, OR 25-3 129-4.

PLATE 67

HUGHES and CROXTON, biorecord 18 CICATR C3

572

PALAEONTOLOGY, VOLUME 16

17 CICATR B20

1

18 CICATR C3 I

19 CICATR A6|

20CICATR DD

L U U U U U U ^

21 CICATR C4

I

[~nj~n-r~Lr~ij —

I

22 CICATR DB

U

23CICATR DCE

V

V

■V

24CICATR C5

25CICATR B21

1

n^j-iTLh

' ^

26CICATR A

j

5T

1

W V/ V V 1

27CICATR C6

^_r~L_rn_j

28CICATR DG

prun-Tir

TEXT-FIG. 1 . Mural profile diagrams of twelve biorecords, all to same scale. Left, four muri and lumina
constructed from measurements made on spores, x 2000. Right, some sketched profiles.

TABLE 2. Exine and sculpture measurements for twelve biorecords. Certain measurements are omitted from the table when they are believed to lack
significance, e.g. for four adjacent muri when the mean would exceed the spore radius; other comments in text.

Sculpture Aim

Height of muri

6

z

62

1

50

19

2

9

21

3

28

35

7

4

LimitsSMean

(0-4) 07 (1-3)

- 2-5 -

(07) f2 (2 0)
(1-5) 2 6 (3-5)

- 10 -

(10) 1-5 (2 0)

(10) 1-3 (2 0)

- 10 -
(0-4) 07 (10)
(0-5) N (1-5)
(15) 2-4 (35)
(07) 0-9 (10)

4 adjacent muri
and lumina

d

z

93

98
58
79
55
100
58

99
96

88

b

O, _0 — m— m — ro 1 O

— cUin — if)c\j — — — 1 —

Limits 8i Mean

(48) 76(100)

(70) 11-9(200)
(16 0)27 1(400
(7-0) 9-9 (130)
(200)30-5 (43-0)
(9-0) 12-5(200)
(8 0) 10-7 (150)
(4 0) 5-7 (9-0)
(8-0) 10-8 (15-0)

(4-0) 6-2 (9 0)

Width of muri

LimitsSMean No.

95

97
100
100
100
100
100
93

99

100

98
100

(0-5) 0-9 (1-5)
(20) 4-4 (6-5)
(1-0) 1-6 (2-7)
(3-0) 5-6 (9-0)
(10) 1-8 (3-0)
(4-0) 7-0 (ll-O)
(1-5) 2-4 (4-0)
(1-3) 1-9 (3-0)
(0-3) 0-6 (1-0)
(1-3) 1-9 (2-5)
(1-0) 3-2 (70)
(0-5) 1-0 (1-7)

+ve

or

-ve

■f 1 1 1 1 1 1 + 1 _^o 1

Laesura

Long

or

Short

Long

Long

Long

Long

Long

Short

Short

Long

Long

Short

Long

Long

e

(U

c

X

LU

Ratio Radial/
Interradial

d

z

48

41

51

71

71

72

33

31

75

LimitsSMean

(1-0) 1-9 (2-0)

(1-0) 1-5 (2-2)
(10) 1-4 (2-7)
(1-3) 2-3 (4-3)
(1-0) 1-5 (3-3)
(1-0) 1-3 (2-3)

(10) 1-1 (1-5)

- 1-0 -

(1-0) 1-3 (2-2)

Radial or
Equatorial Extens.

d

z

49

82

41

54

77

71

72
90
33

56

75

LimitsSMean

(0-6) 1-5 (2-5)
incM

(2-5) 5-6(8 0)

(1-5) 2-3 (3-5)
exM

(3-0) 5-6 (9 0)

(2-0) 4-3 (6-5)

(3-0) 6-0(12-0)

(3-0) 6-4(13-0)

(3-5) 7-4(15-0)

(1-0) 1-7(20)
incM

2-5 (4-4) &0
2 0 (3-6) 5-5

Interradial

d

z

58

82

41

68

71

71
96

72

96
87
77

97

LimitsSMean

(0-6) 1-0 (2-Q)
incM

(15) 39 (5-5)

(1-0) 1-5 (2-0)

(20)4-1 (60)

(1-0) 1-9(30)

(2-0) 4 0 (8 0)

(3 0) 4-6 (8-0)

(1-0) 2-1 (40)

(10) 1-6 (2-0)
incM

(1-0) 2-0(30)

(1-5) 2-8 (5 5)
incM

(2-0) 2-8 (4-5)

CICATR

Biorecord

17. B20

18. C3

19. A6

20. DD

21. C4

22. DB

23. DCE

24. C 5

25. B2I

26. AST

27. C 6

28. DG

574

PALAEONTOLOGY, VOLUME 16

19 CICATR A6
Plate 68 ; text-fig. 1

Description. Prominent lips (figs. 1, 2, 3). Proximal face: some specimens have a
small smooth contact area. Proximal muri: three interradial sets of 3-5. Polar view:
(0) 3 (5) muri (in profile) cross the radial margin (figs. 1, 3). Distal muri form an
asymmetrical pattern of three sets of 1-15 muri (figs. 6, 7). Sub-parallel distal muri
rare. Muri are sinuous and often bifurcate (fig. 7).

Local distinction. 1 cicatr AT has more muri which are straight not sinuous and a thicker exine.

20 CICATR DD
Plate 69; text-fig. 1

Description. Contact area smooth and often encompasses all of the laesura (figs.
1, 2). Lips thick (fig. 2). Proximal muri: three interradial sets of (1) 2 (3). The distal
mural pattern, of three sets of 1-5 (fig. 6) or a sub-parallel set of 4-9 muri (fig. 8),
is distinguished by radial lumina each flanked by two muri which project beyond
the radial margin (figs. 1, 2, 3, 5, 6, 9).

Local distinction. 18 cicatr C3 is smaller with no radial lumen. 5 cicatr A2 is smaller with thicker posi-
tive muri. 22 cicatr DB has no radial lumen.

21 CICATR C4
Plate 70; text-fig. 1

Description. Convexity of amb shape distinctive. Small smooth contact area (figs. 3, 5).
Proximal muri : three interradial sets of commonly 3-4. Distal mural pattern norm-
ally 12-20 sub-parallel muri (fig. 9), rarely three sets. Radial equatorial extensions
conical in shape, ratio length/width at half-length: (0-5) 0-9 (1-7) (74). Proximally
the extensions are unsculptured but on the distal side the muri extend across the
extensions (figs. 1, 7).

Local distinction. 24 cicatr C5 has denser and more elongate extensions. The amb has straighter sides
and there are a lower number of muri.

EXPLANATION OF PLATE 68

Magnification of figs. 1-9, X 1000; fig. 10, x2000.

Figs. 1-10. Biorecord 19 cicatr A6. 1, Proximal aspect; V500/6, OR 33-7 113-5. 2, Proximal aspect;
V500/6, OR 34-5 124-5. 3, Proximal aspect; V500/5, OR 63-6 127-2. 4, Proximal aspect; V500/6, OR
24-4 120-4. 5, Distal aspect, low focus; V500/6, OR 41-9 112-3. 6, Distal aspect; V500/5, OR 36-0
115-0. 7, Distal aspect; V500/5, OR 49-8 113-3. 8, Equatorial aspect; V500/5, OR 53-5 1 17-8. 9, Equa-
torial aspect; V500/5, OR 52-5 114-0. 10, Part of oblique aspect, showing mural profile; V500/6, OR
35-4 110-7.

PLATE 68

HUGHES and CROXTON, biorecord 19 CICATR A6

576

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 69

Magnification of figs. 1-9, x 1000; fig. 10, x2000.

Figs. 1-10. Biorecord 20 cicatr DD, Slide V963/4. 1, Proximal aspect; OR 40-3 1210. 2, Distal aspect,
low focus; OR 27-5 108-5. 3, Distal aspect; OR 37-6 110-3. 4, Equatorial aspect; OR 30-2 120-1.
5, Proximal aspect; OR 38-7 127-8. 6, Distal aspect; OR 39-6 124-0. 7, 8, Proximal aspect, high and
low focus; OR 38-8 1 19-3. 9, Distal aspect; OR 37-6 110-1. 10, Part of equatorial view, showing mural
profile; OR 38-0 1 12-0.

PLATE 69

HUGHES and CROXTON, biorecord 20 CICATR DD

578

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 70

Magnification, x 1000.

Figs. 1-10. Biorecord 21 cicatr C4. 1, Distal aspect, low focus; V963/2, OR 24 0 108-4. 2, Distal aspect,
low focus; V963/3, OR 42-2 117-5. 3, Proximal aspect; V963/4, OR 52-7 123-2. 4, Proximal aspect;
V963/2, OR 33-6 129-2. 5, Distal aspect, low focus; V963/4, OR 50-8 122-7. 6, Distal aspect, low focus;
V103/1, OR 38-4 113-3. 7, Proximal aspect, low focus; V963/4, OR 51-5 115-5. 8, Distal aspect;
V963/4, OR 28-6 111-0. 9, Distal aspect; V963/4, OR 53-0 124-1. 10, Equatorial aspect; V963/4,
OR 34-6 127-4.

PLATE 70

HUGHES and CROXTON, biorecord 21 CICATR C4

580

PALAEONTOLOGY, VOLUME 16

22 CICATR DB

Plate 71 ; text-fig. 1

Description. Proximal muri: three interradial sets of 2-4. Distal muri: 4-8 sub-
parallel muri or three sets of muri in an asymmetrical pattern (figs. 4, 7). This taxon
includes specimens with conical radial equatorial extensions grading into those
with no extension (figs. 2, 7). All specimens in this case have drillings in the muri
and they are therefore thought to be primary.

Local distinction. 9 ciCATR A Pis a similar size but the muri are narrower and more numerous. 20 cicatr DD
differs in mural profile and the distinctive radial lumen.

23 CICATR DCE
Plate 72; text-fig. 1

Description. No lips have been distinguished. No unsculptured area. Proximal muri:
three interradial sets of 3-15. In the radial equatorial area the lumina are discontinu-
ous forming an interlocking pattern beyond the laesura (fig. 7). Distal face strongly
convex and (13) 20 (30) muri bifurcate to accommodate this (figs. 3, 8. 9).

Local distinction. 10 CICATR ASS is smaller with fewer muri and a different radial equatorial and distal
mural pattern.

24 CICATR C5
Plate 73; text-fig. 1

Description. Prominent lips (fig. 2). Proximal muri : three interradial sets of approxi-
mately 3. Distal mural pattern normally three sets of up to 14 muri (fig. 7). Radial
equatorial extensions are unsculptured and are long, narrow, and tapering. Ratio
length/width at half-length: (IT) 1-8 (4 0) (78).

Preservation. Some specimens have ‘drillings’ that appear to be corrosion (fig. 9).

Local distinction. 8 cicatr C2 has ‘knob-like’ rather than tapering extensions and is smaller. 21 cicatr C4
has shorter, more conical extensions which are sculptured distally; the amb is more convex-sided and
there are more muri.

EXPLANATION OF PLATE 71
Magnification of figs. 1-6, X 1000; fig. 7, x500.

Figs. 1-7. Biorecord 22 cicatr DB. 1, Proximal aspect; J035/3, OR 49-6 11 5 0. 2, Proximal aspect;
V963/1, OR 401 116-8. 3, Distal aspect, low focus; V103/1, OR 40-5 112-2. 4, Distal aspect; V103/3,
OR 36-7 113-8. 5, Equatorial aspect; V963/3, OR 55-8 110-6. 6, Oblique aspect; V963/2, OR 33-3
117-1. 7, Distal aspect; V963/2, OR 37-4 117-7.

PLATE 71

HUGHES and CROXTON, biorecord 22 CICATR DB

582

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 72
Magnification of figs. 1-4, x 1000; figs. 5-10, x500.

Figs. 1-10. Biorecord 23 cicatr DCE. 1, Proximal aspect; W103/2, OR 32-5 106-8. 2, Proximal aspect;
V963/2, OR 26-6 125-7. 3, Distal aspect; V963/3, OR 47-9 124-9. 4, Equatorial aspect; V963/3, OR
55-0 124-8. 5, Proximal aspect; V963/4, OR 49-8 118-9. 6, Proximal aspect; V963/4, OR 43-7 120-3.
7, Proximal aspect; W103/1, OR 31-7 111-6. 8, Distal aspect; V963/3, OR 40-3 129-0. 9, Distal aspect ;
V963/3, OR 47-4 121-2. 10, Oblique aspect; V963/3, OR 30-2 129-2.

PLATE 72

HUGHES and CROXTON, biorecord 23 CICATR DCE

584

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 73

Magnification, x 1000.

Figs. 1-11. Biorecord 24 cicatr C5. 1, Proximal aspect; W190/3, OR 35-2 123-6. 2, Proximal aspect;
W190/4, OR 48-2 116 0. 3, Distal aspect, low focus; W190/2, OR 52-2 104-5. 4, Distal aspect, low
focus; W190/5, OR 38-7 119-6. 5, Distal aspect, low focus; W190/6, OR 44-6 122-6. 6, Distal aspect,
low focus; W190/2, OR 25-1 116-9. 7, Distal aspect; W190/5, OR 38-7 1 19-6. 8, Distal aspect; W190/5,
OR 40-0 111-8. 9, Distal aspect; W190/5, OR 42-1 122-5. 10, Distal aspect; W190/6, OR 29-4 122-7.
1 1, Equatorial aspect; W190/6, OR 30-7 109-4.

PLATE 73

HUGHES and CROXTON, biorecord 24 CICATR C5

586

PALAEONTOLOGY, VOLUME 16

25 CICATR B21
Plate 74 ; text-fig. 1

Description. Proximal face has a small smooth apical area (figs. 2, 3). Proximal muri
(12-16) are in three sets oblique to the edge of the amb and meet the laesura at an
angle (anticlockwise swirl) (figs. 1, 3, 6). Distal mural pattern is three asymmetrical
sets of 8-12 muri or 20-25 sub-parallel bifurcating muri. Distal muri may also show
clockwise swirl.

Local distinction. 17 cicatr B20 is smaller, has larger and fewer muri. 28 cicatr DG has wider negative
muri and a thicker exine. In neither of the above have ‘swirling’ proximal muri been observed.

26 CICATR A5T
Plate 75; text-fig. 1

Description. Some specimens have a smooth contact area (figs. 2, 3). Proximal muri:
three interradial sets of 3-7. Outer proximal muri continuous round ends of laesura
(figs. 1, 2, 3, 4). Distal muri: 12-20 in a sweeping ‘parabolic’ pattern (figs. 8, 9).

Local distinction. 10 cicatr A5S is smaller, has narrower lumina, rectangular mural profile, thicker exine,
and circular lumina occur in the distal mural pattern.

27 CICATR C6
Plate 76; text-fig. 1

Description. Interradial exine thickness may vary in the same specimen (fig. 6).
Smooth contact area covering most of the proximal face (fig. 3). Sculpture may be
positive or negative, width of one lumen: (1-0) 2-9 fxm (8 0) (92). Distal muri: three
asymmetrical sets of 1-7. Radial equatorial extensions variable: conical, tapering,
or ‘flattened’; in some specimens they are not very prominent. Ratio length/width
at half-length: (0-6) 1-0 (1-7) (48).

Local distinction. 18 cicatr C3 has wider negative muri with a rectangular mural profile and more pro-
minent and uniform radial extensions. 7 cicatr Cl has more numerous narrower muri.

EXPLANATION OF PLATE 74
Magnification of figs. 1-11, x 1000; fig. 12, x2000.

Figs. 1-12. Biorecord 25 cicatr B21. 1, Proximal aspect; W190/4, OR 26-2 116-3. 2, Distal aspect, low
focus; W190/4, OR 35-5 116-7. 3, Proximal aspect; W190/2, OR 53-1 115-2. 4, Proximal aspect;
W190/6, OR 30-3 109-3. 5, Proximal aspect; W190/1, OR 59-6 124-4. 6, Proximal aspect; W190/5,
OR 33-5 123-9. 7, Distal aspect; W190/6, OR 53-9 118-0. 8, Equatorial aspect; W190/2, OR 27-1
123-3. 9, Equatorial aspect; W190/4, OR 36-6 125-3. 10, Equatorial aspect; W190/4, OR 51-6 123-1.
11, Oblique aspect; W 190/4, OR 29-0 125-1. 12, Part of oblique aspect, showing mural profile; W190/4,
OR 33-2 121-8.

PLATE 74

HUGHES and CROXTON, biorecord 25 CICATR B21

588

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 75

Magnification, x 1000.

Figs. 112. Biorecord 26 cicatr AST. 1, Proximal aspect; V197/5, OR 29T 1210. 2, Distal aspect, low
focus; V197/5, OR 41-9 125-3. 3, Proximal aspect; V197/4, OR 32-6 117-7. 4, Proximal aspect; V197/6,
OR 55-0 113-2. 5, Distal aspect; V197/7, OR 37-9 117-0. 6, Proximal aspect, low focus; V197/4, OR
58-4 104-5. 7, Distal aspect; VI 97/4, OR 33-1 112-6. 8, Distal aspect; V197/4, OR 26-8 122-3. 9, Distal
aspect; VI 97/4, OR 54-7 112-3. 10, Equatorial aspect; VI 97/6, OR 26-0 123-0. 1 1, Equatorial aspect;
V197/6, OR 28-2 105-7. 12, Oblique aspect; V197/4, OR 57-3 127-0.

PLATE 75

HUGHES and CROXTON, biorecord 26 CICATR AST

590

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 76
Magnification of figs. 1-11, x 1000; fig. 12, x2000.

Figs. 1-12. Biorecord 27 cicatr C6. 1, Proximal aspect; V198/3, OR 30-2 109-8. 2, Proximal aspect;
V198/2,OR421 117-4. 3, Proximal aspect; V198/2, OR 37-1 121-0. 4, Distal aspect; W262/1, OR 28-6
110-8. 5, Proximal aspect, low focus; V198/2, OR 14-8 1 10-5. 6, Distal aspect; V198/3, OR 57-1 108-9.
7, Distal aspect; V198/4, OR 51-3 126-6. 8, Distal aspect; V198/2, OR 37-5 112-3. 9, Oblique aspect;
W262/3, OR 60-7 125-7. 10, Distal aspect; V198/1, OR 43-3 125-3. 1 1, Equatorial aspect; V198/1, OR
33-1 122-0. 12, Part of oblique aspect, showing mural profile; VI 98/4, OR 48-0 115-1.

PLATE 76

HUGHES and CROXTON, biorecord 27 CICATR C6

592

PALAEONTOLOGY, VOLUME 16

EXPLANATION OF PLATE 77

Magnification, x 1000.

Figs. 1-12. Biorecord 28 cicatr DG. 1, Proximal aspect; W058/2, OR 56-6 1191. 2, Proximal aspect;
W058/7, OR 45-5 113 0. 3, Proximal aspect; W058/1, OR 31-4 115-2. 4, Distal aspect, low focus;
W058/2, OR 46-9 114-5. 5, Distal aspect, low focus; W058/1, OR 35-8 102-4. 6, Distal aspect, low
focus; W058/1, OR 51-6 106-9. 7, Distal aspect; W058/3, OR 54-3 107-2. 8, Distal aspect; W058/7,
OR 51-6 123-9. 9, Equatorial aspect; W058/8, OR 34-7 112-2. 10, Equatorial aspect; W058/7, OR
41-6 117-3. 11, Equatorial aspect; W058/8, OR 39-3 121-9. 12, Oblique aspect; W058/2, OR 40-6
110-6.

PLATE 77

HUGHES and CROXTON, biorecord 28 CICATR DG

594

PALAEONTOLOGY, VOLUME 16

28 CICATR DG

Plate 77 ; text-fig. 1

Description. Laesura may be sinuous (fig. 1). No smooth contact area distinguished.
Proximal muri: three interradial sets of 15-16. Distal mural pattern: 18-25 sub-
parallel muri.

Local distinction. 17 cicatr B20 is smaller with a thinner exine and positive sculpture. 25 cicatr B21 has
narrower positive muri and a thinner exine, also ‘swirling’ proximal muri.

SYSTEMATIC DESCRIPTION OF THE EVENTS

The palynologic events are composed of comparison records graded A and B,
or ungraded either if the number of specimens used was below twenty-five or if
preservation was imperfect. The details are given in Tables 3-5.

Revised event. In the case of sample WM 174016 from which Event 53 was raised
(Hughes and Moody-Stuart 1969), further preparations have been studied and the
constituent taxa reviewed; a new event number (81), which in effect bears the date
1972, has therefore been allotted for the new information although it has been taken
from the same rock sample. In contrast the Events 43 ( WM1819j5) and ?>6{WM1843)
have been used again; the details of these two have been presented in Table 3 for
convenience of use with new events.

Unworked taxa. The column headed ‘others’ on the right of Tables 3 and 4 contains
records, in addition to those of rare unplaceable specimens, of spores which we
would previously have placed in comparison cfC. to an existing biorecord. Such
records implied presumed new taxa which were not, however, made into biorecords
as they were not needed for comparison in the present project, thus saving consider-
able time. Percentages in this category are occasionally high as in Event 100 (Table 3)
or 112 (Table 4) in which in each case there were high numbers of forms not sub-
sequently seen elsewhere. Unpublished details are filed. We feel that this principle
could be applied with saving elsewhere in palynologic publication.

EVENT CORRELATION

The selected Worbarrow events are correlated individually with the reference
scale section at Warlingham (text-fig. 2) by means of a bracket in each case; the
bracket consists of the two statements ‘After reference event X’ and ‘Before ref-
erence event Y’. In theory such a statement may be refined subsequently up to the
limits of rock sampling. No attempt is made to equate a Worbarrow event with a
reference event as this is logically impossible.

The points considered to be critical in each correlation are set out below but the
list should be read in conjunction with Tables 3-5.

105 EVENT W128: between events 36 WM1843 and 72 WM1795; occurrence of
both cf. 6 B5 and ef. 8 C2 (suggesting proximity to event 43 WM1819I5); high
percentages cfA. 1 AT, cfA. 4 AW, and cfA. 7 Cl ; absence of cf. 10 A5S and cf.
17 B20.

TABLE 3. Event composition data for Warlingham Borehole reference scale. In comparison tabulation, larger numerals indicate biorecord from that

sample. CON = ‘Contignisporites' .

SJ3MI0

=

so

in

20

o

ro

to

o

in

-

rO

ro

7

C

)

—

—

—

O

ODO

6

cf

_

o<

PJ Q

U

fv- iD
OJO

7

cf

u

VP

01

as

oo2

“o

rot3

5

cf.

2

cf.

~ u

26

AST

o<

t£)0

o

ro*^

17

CfB.

ioT3

9

CfB

9

CfB

CDh-

M-T3

o£

lOo

-o2

ro'^

cmT3

- CD

o

c

0>

rT m
<M (J

4

cf

a.<

o

2

cf.

cf

rOo

2

cf

tnT3

3

cf

CMT3

“ u

<f>

(D

(/)

c

toy

M’o

o

O

Q.

(O

CVJ CD

.. •

a.

o

CM Q

*“ o

“ o

fO o

*“ u

— u

3

o

Q.

E

_ ^
CMO

ro <j

fO*^

10

cf

lo'o

3

cf

to

~ o

~ o

O'

o

OQ

ro

2

cf

(/>

CM O

to

,

(D

0)

19

A6

2

cf.

7

cf.

4

. cf-

.n£
~ o

8

cf

II

CfB.

(Do

to

O

CL

(/>

to

o

18

C3

8

cf

3

cf.

“■*o

3

cf

cf.

ro<!

o<

o

r-o

4

cf.

M-73

cmT5

in

ro^J

T5

C

h- CM
GO

9

cf

5

cf

lO*o

GQ

13

8

cm's

*“ o

-<

r-<

■^b

33

to2

o

00*0

c^<
“ o

ro'o

“o

o

o

O in
“ <r

o“

gjCD

“o

<JV*-

o

18

cfB.

CM CD

m'5

roS

o

o

"’T3

o

PJ<

i^b

=

o

”o

cmT3

(/>

o

o

■O

“*o

“*0

roT)

7

cf

ro

cf

■^O

u

■ —

o

o

o

0)

8

C2

'«o

4

cf

roT3

2

cf

5

cf.

M-o

cf.

6

cf

>«b

p

9

3

cf.

O

m'o

ro^

u

o

CD

7

Cl

4

cf

in^

2

cf.

ro'^

cmTS

25

cfA.

18

cfA.

O

pjb

O.S

o

o<

“ o

99

9

piS

u

tn~

■O'C

OOo

“TS

t^TS

cmTj

roT3

<j><
“ o

Cf

“o

U

20

CfA.

\£y

3

cf.

(M<

o

o

17

cf A.

-<

CMh-

u

P

6

ro^

ro"^

>P>0

46

CfA.

47

.CfA.

3

AR

■”^3

(M^j

2

cf.

oS

a

_<

'o

1 1
AT

II

cf A

2

cf.

II

cf A.

7

CfA.

<

2>t3

7

cf.

9

cf.

coS

o

tos

“ o

in

0)

a

to

X

<D

TD

C

!=

m

ID

m

o

ro

to

CM

in

CD

ro

o

CM

fo

CO

0

h-

0^

m

CM

in

CM

7 20

1 29

7 39

7 27

to

r--

CD

'£

rO

~

6

in

in

to

in

to

m

in

to

to

to

m

in

«:r

in

m

o

o

0>

M

ro

O

r-

to

00

00

to

ro

00

ro

o

to

to

to

m

CM

u.

Li_

(/)

rO

ro

CM

CM

CM

CM

ro

CM

CM

CM

ro

ro

u

20

28

20

CM

50

o3

26

(T>

ro

M"

20

eg

34

CJ*

o

S!i|odossD|0

CM

1

-

to

-

CM

00

O

in

ro

ro

OO

m

in

m

iD

in

o

c

>>

Sd(DDDDSj9

CO

25

25

28

45

23

CM

=

ro

'tr

r«-

o

23

ro

30

37

30

\£

CM

CM

o

CL

SUJ8J

47

in

34

33

34

36

32

54

32

55

'd-

37

45

00

49

38

22

43

47

24

S0|3

m

o

CM

C2

00

CM

CM

r--

o

-

-

-

CM

cn

(D

CM

'£

1

Q

o;

c
■ O

CM^

in

ro

CM

\

m

tD

CM

'V.

CO

m

\

o

r--

ro

\

r-

o

CM

\

m

in

M*

\

ro

to

fO

\

00

in

CM^

~

ro

\

ro

ro

\

rO

M-

v,

O

CO

\

00

5;

o

o

to

in

\

O

CO

lO

\

m

ro

CM^

\

ro

ro

ro

'v

to

CL

O

CM

O

o

o

0^

(D

(T>

in

C7>

m

o

m

ro

in

in

o

O

s

5

$

s

5

>

>

>

>

>

>

>

>

>

>-

>

>

>

<D

CD

\

O

CO

C<J

\

ro

in

to

f:::

to

\

m

CM

\

cn

to

m

\

to

<7^

to

■\

o

m

m

to

\

to

\

o

CO

\

CD

in

ro

Q.

iO

m

rO

to

m

OO

m

0^

M-

c

e

o

CM

CM

ro

ro

in

m

to

to

to

to

CD

00

0)

o

(/)

s

5

5

2

s

s

s.

5

5

5

5

5

s

s

5

s

>

LU

5

5

5

5

5

s

5

s

5

5

5

g

$

5

S

$

5

5

5

d

ro

O

O

(T>

(T)

1^

(D

O

cn

00

1^

CO

ro

cn

CM

ro

CD

o

o

CD

<T>

CD

(T>

(T>

00

CO

CO

CO

CO

CD

1^

p

ro

SJ8MJ0

H

c

"5

c

cd

■s.

X

PQ

O

o

.2

CQ

'5

XJ

c

o

o

a

S

o

>

u

.J

50

CM CD

Cl

E

o

-o

o

o

Q)

O

(D

^ O
r-cvj
”CD

o!n

<

^Cl

CT^<

^CM

®L

CD

29

g

-

23

20

-

00

CM

CM

M-

ro

3

cf.

cdTS

II

CfA,

<£

9

cfA.

*^0

co<

o

VP

II

14

CfA,

2

cf

ro

4

cf

6

cf

cSo

2

cf

3

cf

*(3

i-£

O

4

cf

«><

o

m*t3

3

cf

>"o

~ u

”73

cf.

2

cf

6

cf

5

cf

o<

”o

10*0

“*0

8

cf

’^73

“"o

oj73

7

cf

C£>o

cf

in29

o

^OD

■^O

3

cf

o£

O

II

cfB

<fi<

csj73

7

cf

” o

”73

<M't5

cm73

2

cf

” o

m2

u

9

cf

7

cf

2

cf

ro'tS

2

cf

21

CfA.

C7><

fO<

»^73

C£)73

9

CfB

<Mo

»f>75

2

cf

M-73

II

CfA

7

cf.

18

CfA.

m-73

47

cfA.

ro2

o

II

CfA

'J-s

o

•«o

18

CfA

13

cfA

14

CfA

O

CM

29

CO

ro

27

29

CM

38

34

43

5

63

53

43

54

56

64

00

in

m

42

29

C£>

00

CJ^

cn

IT)

If)

-

in

44

33

22

r-

00

CM

27

1

-

M-

-

CM

64

1

68

24

ro

40

ro

g

30

ro

fO

38

CM

fO

38

35

36

39

ro

M*

rO

44

23

38

lO

o

fO

-

ro

CD

-

CD

in

CO

CM

\

o

<r»

m

\

C7>

CM

CD

O

ro

\

00

<T>

CM

\

CM

<T>

\

rO

CM

\

CM

\

O

o

CM

>

to

•s.

O

00

CM

>

>

>

5

5

5

<r>

CD

CM

CM

IT)

CM

oo

CM

00

CM

i

5

5

5

s

5

5

5

5

lO

2

—

o

601

108

107

106

105

js jaijio
S!||odossD|3
saiDOODSig
suiej J9LH0
so|o

<^o

0) .—

i O

TABLE 5. Preparation data for events.

It

COO
c d.

Q>

2(0

^ I

^ ^ ^

OX
'•*- c.
o o

T3 U

o o
o

£E — ^E — EE
e E E G E e E E
OgOOgOgg

E E E E E

O O O o g

— — — to rO

&.9

CL o

ii-' — CM

s s >

oi CM — — :

5 5

)±L

5 $ $

5 5

o

o

<v

o

m

E

o

CD

2 g

3 CT) ^ 2.

$ $ $ $

CM — —

ID rO
O O

9 8

LZ r- l:

— CM CM

$ 5 ^ $

5 $

i2

■c

O O
^ 0)

->

1

1

l

o

coo

"S

Q

Min

Sep

>>

>>

%

%

%

c

o

o

' —

cZ

«o

CO

<o

t/i

(/>

tn

(ft

t/i

o

C

c

c

c

c

c

c

C

c

c

c

c

C

— c

E

E

E

E

E

G

E

E

E

E

E

E

E

E

E

E

E

CL

o

O

O

o

in

O

O

o

in

o

in

in

in

O

m

in

o

<D

rO

CM

ro

ro

CM

CM

CM

rO

CM

CM

CM

CM

CD

CM

CM

c£

O o
o

&o

in

fO

ro

in

00

h-

CD

O

CO

O

in

ro

CD

CO

m

in

o

o

00

r^)

ro

a

. o

o

0'>

O

o

<T>

in

cn

in

O

in

ro

in

in

O

o

5

>

:5

■:>

>

>

>

>

>

>

>

>

>

>

>

>

</>

LiJ

o

8

o

m

o

o

CM

CM

CM

CO

ro

d"

m

GD

<

O

o

CD

o

o'

CM

Q

CM

CM

S

oS

go

c

CO

O

m

00

CO

ro

r-

CM

ro

CD

>

o>

CJ>

o

cr>

CO

CO

CO

CD

00

00

r-

N

ro

UJ

rO

in

CD

CD

CD

00

in

0)

Q.

M"

(T>

\

in

CO

m

r-

ro

<y>

CO

\

o

m

M-

in

m

\

o5

o

M"

<T>

r-

in

m

<T>

\

o>

ro

M*

E

fO

m

m

CO

CO

CO

CO

1^

r--

r-

00

00

o

(T)

5

5

s

s

s

s

s

5

$

5

5

5

5

5

5

s

s

5

5

5

5

5

(S)= Schulze’s solution

598

PALAEONTOLOGY, VOLUME 16

106 EVENT W124 and 107 event W18\ technically also between events 36 WM1843
and 72 WM1795, but probably after event 43 WM 18 191 5 on superposition and
on absence of cf. 6 B5 in spite of comparable ‘fern spore size index’ with event
105 W^72<^ (see above).

108 EVENT W121\ between events 72 WM1795 and 83 WM1681I6; occurrence of
cf. 18 C3, but not of cf. 19 ^<5; high percentages of cfA. 10 A5S and cfA. 4 AW.

109 EVENT W15: between events 81 WM 174016 and 87 WM1645; occurrence of
cfA. 19 A6, but not cf. 24 C5, cf. 25 B21, or cf. 26 AST; high percentages of cfA.
10 ASS, and cfA. 17 B20.

110 EVENT W2: between events 86 WM16SS and 90 WM1S37; occurrence of cf.
24 C5, cf. 25 B21, and cf. 26 AST, but not of cf. 20 DD and cf. 23 DCE; last occur-
rence of cf. S A2.

1 1 1 EVENT W14 : between events 89 WM1S69 and 97 WM 139411 ; occurrence of
cf. 22 DB, cf. 23 DCE, and cf. 27 C6.

112 EVENT W12B: between events 89 WM1S69 and 97 WM 139411; occurrence of
cf. 20 DD, cf. 22 DB, cf. 23 DCE, and cf. 27 C6 ; last occurrence of cf. 9 AP.

114 EVENT W9: between events 96 WM141SI3 and 99 WM121Sj6; high percentage
of cfA. 26 AST; occurrence of cf. 24 CS and cf. 27 C6; no record of cf. 20 DD,
cf. 22 DB, or cfA. 28 DG.

115 EVENT Will : between events 98 WM 13 1912 and 103 WM1060j9; occurrence
of cf. 28 DG, cfA. 26 AST; relatively high percentage of biorecord 24 CS.

In several cases above, we could guess at a narrower bracket but the proof could
not be satisfactorily expressed at this stage, e.g. 105 Event correlation 42/44, 108
Event correlation 77/81, 109 Event correlation 83/86, 112 Event correlation 96/97,
114 Event correlation 97/98, and 115 Event correlation 99/101. It is to be expected
that the application of further data from other spores would supply proof, and this
process would comprise the progressive refinement or narrowing of the brackets
for which we intend to provide.

As will be seen from the event numbering system there are many intermediate
events in the reference scale but they are not quoted because they are not rich enough
in Cicatricosisporites group spores to bear on the current problem ; in most of these
cases the fern spore size index shows low percentages of spores over 50 ^Ltm diameter.

STRATIGRAPHIC CONCLUSIONS

The ‘Coarse Quartz Grit’ of Worbarrow Bay (Arkell 1947, Bed 14) appears to
fall between events 81 and 83 on the Warlingham scale approximately around
1700 ft depth in the borehole. Worssam and Ivimey-Cook (1971) of the Institute of
Geological Sciences consider this depth to lie in the Upper Tunbridge Wells Sand-
stone, based on general lithologic grounds and partly on ostracod correlations.
Hughes and Moody-Stuart (1969) made palynologic correlations that suggested
that the 1700 ft depth correlated with rocks rather lower in the outcrop succession of
the Hastings Beds. Further work will be necessary to reconcile these views, but we sug-
gest that it may not be entirely valid to identify outcrop formations such as the Wad-
hurst Clay at such a distance from the Central Weald, and near to the basin margin.
The age of the ‘Coarse Quartz Grit’ is still not better known than late Valanginian/

WARLIN6HAM

BOREHOLE

Depth
in feet

1100 -

1200-

1300-

1400-

1500-

1600-

1700-

1800-

1900-

2000-

LGS

-1060/9

1153/2-8

-1217/6

1275/6

1319/2

1394/1

1415/3

- 1456/5

■1537

1569

- 1610/6

1645

1655

-1681/6

.1740/6
; 1749/8
1757

1795

■1819/5

■1843

Reference

Events

Correlation

Brackets

WORBARROW BAY
SECTION (DORSET)

Events

115

114

112

108

107

106

105

Will

W9

WI2B-

17

Ill

WI4 -

no

WI2 -

15

109

WI5 -

W2I

WI8-

W24

WI28-

32

Height in
feet above
Marble Bed
1-1300

30

28

■1200

-1100

-1000

26

23

-900

10

Bed

Arkell

1947

800

-700

-COG

-600

-500

■400

300

-200

-100

Purbeck
Marble
I Bed I

TEXT-FIG. 2. Diagram to show correlation brackets of ten Worbarrow Bay Cicatricosisporites events, on
a reference scale of events of similar origin from Warlingham Borehole. CQG = the prominent ‘Coarse

Quartz Grit’.

600

PALAEONTOLOGY, VOLUME 16

early Hauterivian (Hughes 1958). Although this bed is only 20 ft (6 m) thick, it
presumably represents some distinct tectonic event in south-west England or perhaps
in the then adjacent Iberia (Allen 1972).

COMMENTS ON METHOD

Starting work in a new section. We found it advisable to make provisional biorecords
of twenty-five specimens and brief comparison records in the first place, so that the
considerable time and effort necessary to make 100 specimen biorecords, etc., could
be saved when these were not required for any specific correlation. Observer-time
is the only real bottleneck in the procedure.

Assemblage-types (Batten 1973). Samples have not yet been fully analysed in this
way, and no mathematical expression has been devised to allow for the effect of the
‘Fern Spore Size Index’, both on biorecords and on events.

Future work. Supplementation of the present work by use of several other promising
groups of palynomorphs in the same way will strengthen and refine these correlation
brackets.

REFERENCES

ALLEN, p. 1972. Wealden detrital tourmaline; implications for north-western Europe. Jl geol. Soc. Land.
128, 273-294.

ARKELL, w. J. 1947. Geology of the country round Weymouth, Swanage, Corfe and Lulworth. Mem.
Geol. Surv. Engl, and Wales.

BATTEN, D. J. 1973. Usc of palynologic assemblage-types in Wealden correlation. Palaeontology, 16, 1-40,
2 pis.

HUGHES, N. F. 1958. Palaeontological evidence for the age of the English Wealden. Geol. Mag. 95, 41-49.

1971. Remedy for the general data-handling failure of palaeontology. In cutbill, j. l. (ed.). Data

processing in biology and geology. Syst. Assoc. Spec. Vol. 3, 321-330.

1973 (in press). Towards effective data-handling in palaeopalynology. Proc. 3 Internal. Palynol.

Conference, Novosibirsk {1971), Sec. 2.

and MOODY-STUART, J. c. 1967. Proposed method of recording pre-Quatemary palynological data.

Rev. Palaeobotan. Palynol. 3, 347-358, 1 pi.

1969. A method of stratigraphic correlation using early Cretaceous miospores. Palaeontology,

12, 84-111, 10 pis.

STAFLEU, F. A. ct al. 1972. International Code of Botanical Nomenclature as adopted by the XI International
Botanical Congress, Seattle 1969. Internal. Assoc. Plant Taxonomy, Regntm Vegetabile, 82, 426 pp.
WORSSAM, B. c. and ivimey-cook, h. c. 1971. The stratigraphy of the Geological Survey borehole at Warl-
ingham, Surrey. Bull. Geol. Surv. G.B. 36, 178 pp.

N. F. HUGHES
C. A. CROXTON
Department of Geology
Sedgwick Museum
Downing Street

Typescript received 27 July 1972 Cambridge

HUGHES AND CROXTON: CORRELATION OF WEALDEN

601

APPENDIX OF SAMPLE DESCRIPTIONS

WARLINGHAM BOREHOLE

Unregistered samples, by courtesy of the Geological Survey in 1956,

WM 1060/9
WMl 153/2-8
WM1217/6

Mudstone, light olive grey (5 Y 6/1); mica. Plant fragments present.

Mudstone, light olive grey (5 Y 6/1), laminated; mica. Plant fragments abundant.
Siltstone, mottled light olive grey (5 Y 6/1) and yellowish grey (5 Y 7/2), unsorted; mica.
Plant fragments common.

WM1275/6

WM1319/2

WM1394/1

Mudstone, greenish grey (5 GY 6/1), laminated; mica. Plant fragments present.
Mudstone, light olive grey (5 Y 6/1); mica. Plant fragments present.

Siltstone, yellowish grey (5 Y 7/2), wavy laminations. Plant fragments common. Pyritized
plant fragments.

WM1415/3

Siltstone, light olive grey (5 Y 6/1) and pale yellowish brown (10 YR 6/2), laminated;
carbonate ; mica. Plant fragments present.

WM1456/5
WM1537
WMl 569
WM1610/6
WMl 645
WM1655
WM1681/6
WMl 740/6

Banded medium grey (N5) mudstone and yellowish grey (5 Y 7/2) siltstone; mica.
Siltstone, greenish grey (5 GY 6/1), wavy laminations; mica.

Mudstone, yellowish grey (5 Y 8/1); mica.

Mudstone, light olive grey (5 Y 6/1); mica. Plant fragments common.

Mudstone, light olive grey (5 Y 6/1); mica. Plant fragments common.

Siltstone, banded very light grey (N8) and light grey (N7), wavy laminations; mica.
Siltstone, pinkish grey (5 YR 8/1), laminated; mica. Plant fragments present.

Siltstone, banded yellowish grey (5 Y 8/1) and medium light grey (N6); mica. Plant
fragments present.

WMl 749/8

Mudstone, banded light olive grey (5 Y 6/1) and very light grey (N7); mica. Plant frag-
ments present. Ostracods abundant.

WMl 757

Siltstone, yellowish grey (5 Y 8/1), wavy laminations; mica. Ferruginous staining, grey-
ish orange (10 YR 7/4).

WMl 795

Mudstone, medium light grey (N6), laminated; calcareous; mica. Plant fragments present.
Ostracods abundant.

WM1819/5
WMl 843

Shale, banded light grey (N7) and medium grey (N5); mica.

Siltstone, light olive grey (5 Y 6/1), wavy laminations; mica. Plant fragments common.

WORB ARROW BAY

Will Fine sandstone, light olive grey (5 Y 5/2), unsorted; semi-consolidated. Abundant plant

W9

fragments.

Unsorted siltstone with larger pebbles, light brownish grey (5 YR 6/1). Abundant plant
fragments.

W12B

Banded fine sandstone and siltstone, pale yellowish brown (10 YR 6/2); semi-consoli-
dated. Plant fragments present.

W14

W2

W15

W121

Mudstone, medium light grey (N6). Plant fragments abundant. Ostracods.

Medium sandstone, light brownish grey (5 YR 6/1), unsorted. Plant fragments common.
Medium sandstone, pale yellowish brown (10 YR 6/2), unsorted. Plant fragments common.
Siltstone, dark yellowish brown (10 YR 4/2), unsorted. Large plant fragments abundant.
Ferruginous stains, light brown (5 YR 5/6).

W18

W124

Mudstone, medium light grey (N6). Plant fragments present.

Banded siltstone, greyish olive (10 Y 4/2) and greyish yellow (5 Y 7/2), unsorted. Small
plant fragments present. Ferruginous staining, dark yellowish orange (10 YR 6/6).

W128

Fine sandstone, dark greenish grey (5 GY 4/1), unsorted. Large plant fragments. Fer-
ruginous stains.

M

[Owing to Authors" revision there are no pages 602-606]

^-":r

'■" -" ■» '■■ ■^. ■'■ ‘v^i- ■„

■ < ; « ,.■•?.■ '’>ri(V •f."' .rs

"'^ ■■ •■ ■ U ■ ■•'■'"■ V . V . ,

t ,- ,!'• • M» J ■ M SH!.'''** >‘'H>:> i4^.1 ^ • ‘ A'..'

t ■ • *i%' •’•

i*.-

t. . « : K. "'^'■;> •:.• ;|ii'

,,.**1V .1 » * V.! % ' ■ '’.V^'^vW

.V »;'{■ 'kt >'•'., S' *'’” ■

> ' ' '■' . ■’S'’";'' ''

' ■'. , - , *.*■(%'* '\'J^ ;'t ,". -' ;;i' '>■■'•’

V'' ■ '”■ ’.' '*.’ V

m

" " ,Wv/ . ' ^^'''' '.f ' ' ' r ', ‘fM# ij' »

*• ’ ,.;■ !•■■ _ ' . *f -. '- i.-fM _' ,

t >,n< ‘ '.';ii','f-' tf -. '■ ■

■ i ••-, ■‘V5_ri*!; i'jli-V"' f,K V/. :

V ir-’ ,irr-*wj*iW,

.; 'C ai ■ '*'.

...'iJ!. -:^. ,i'' /..Mfriu;) ■'' ■ '

'•;»■ '-'a/..

'■ !' v'ift' ■' '■

YW

■ i

i **••“• '\ fi'v" 'i-i.*., *.t

■<■' ^ >(i'iv.':''*'^j*^

■'. ,' '''.li) 'h'

-. ♦ .

'• i< '■'■ a'! w ^

^ ^ i a‘

. ■»

■It: 1-1 .1 I '

'• - , f .1 V

\.v'*‘ :■... T; »«-’t!S'4»''.4|S«

■■ ■ ,^. nv. > i,-

I M , 4* " I ; ' • ■=^'!«Tli^.||^ -f .J (,■ HM(
”. . ft -' '■ " L’jUlt‘

" ..j,#;. ■ r''j4 ,

.' '■ . .,t:'t 1, »T<t-,V'- ■.

(.f/r*l ~. ,( , , ,. > r i.'iii'i

V ' I- ■,'r.lM

• <r<^ } ■ di’ ■• ■,',1 1 f . ^11 '■ I

'i'l' .-Jit;;:-'

ISOTOPIC RATIOS AND WEALDEN
ENVIRONMENTS

by P. ALLEN, M. L. KEITH, F. C. TAN, and P. DEINES

Abstract. Isotopic methods of assessing Wealden palaeoenvironments are reviewed. Emphasis is placed on the
importance of using only primary, skeletal, untransported carbonates. Recrystallization and cementation generally
reduce the and ratios of Wealden carbonates. Hence, while those with high are likely to be

marine, those with low cannot be assigned to a depositional environment unless known to be primary.

Ratios from aragonitic shells of one species each of Cassiope, Neomiodon, and Filosinal indicate ‘marine’ condi-
tions. Less certainly, marine conditions may be indicated by calcitic shells of pelecypods (one species each of Lio-
slrea and Neomiodon), gastropods (one species of Paraglauconia), and ostracods (two species of Cypridea and one
of Mantetliana). Forms not assignable to ‘marine’ or ‘fresh’ water are three species of Cypridea, one species each
of Theriosynoecum, Damonella, Filosina, Unio, Viviparus, and a corbulid, and two species of Equisetites. The new
data support the transgressive models for the major Wealden clay formations.

Interpretation of transitional water temperatures from ratios is unreliable. Nevertheless, values from

primary aragonites in the Hastings Beds (Neomiodon) and Weald Clay (Filosina, Cassiope) lie in the range of modern
marine molluscs from warm temperate waters. This is consistent with current opinion based on other palaeontological
evidence.

An earlier reconnaissance (Allen and Keith 1965) suggested that, in the Wealden
clay formations, broadly reflects the variable depositional salinities inferred
from palaeontological evidence. Subsequent work on body-fossils and trace-fossils
confirms the fluctuating salinities, particularly for the Weald Clay, and the marginal
environments postulated (Shephard-Thorn et al. 1966; Smart et al. 1966; Anderson
1967; Anderson et al. 1967; Worssam and Thurrell 1967; Thurrell et al. 1968;
Kilenyi and N. W. Allen 1968; Dines et al. 1969; Kennedy and MacDougall 1969;
Batten 1969; Watson 1969).

Such environments yield isotopic ratios that are so far impossible to interpret in
detail (Keith and Parker 1965). This, combined with problems of Wealden dia-
genesis and our previous use of whole-rock samples, led us to investigate the limita-
tions of the isotopic method more fully. In particular Neomiodon, with its widely
ranging and varying degrees of recrystallization (Allen and Keith 1965, table 1),
merited further examination.

LIMITATIONS OF ISOTOPIC METHODS

General. The isotopic basis for interpreting ancient depositional environments and
for distinguishing marine carbonates from ^^C-deficient freshwater carbonates
was discussed in an earlier paper (Allen and Keith 1965) and will not be repeated
here, except for caveats regarding limitations on palaeosalinity estimates derived
from Measurements of isotopic compositions are reported in the standard

8-notation, i.e. as differences of (and ratios in parts per thousand

relative to those of the Chicago PDB standard, a Cretaceous belemnite:

gi3C — { sample ^ ^^C/^^C standard \ jqqq

^ ^^C/i^C,tandaTd '

[Palaeontology, Vol. 16, Part 3, 1973, pp. 607-621.]

608

PALAEONTOLOGY, VOLUME 16

Interpretation of past salinities can be based on isotopic comparisons of ‘ideal’
ancient carbonates with the ranges of modern carbonates (Keith et al. 1964).
Two categories appear meaningful; ‘marine’ > — 27oo) and ‘freshwater’

(S^^C < — 2%o)? with some overlap across the arbitrary boundary. An ‘ideal’
sample (not easily obtained or recognized in practice) would consist of unrecrystal-
lized carbonate formed in isotopic equilibrium with the environment to be deter-
mined and still retaining the original population of carbon (and oxygen) atoms.
Complications arise mainly because other kinds of carbonate may be associated,
including (a) clastic carbonates from limestones, {b) biogenic carbonates with vari-
able or non-equilibrium carbon isotopic compositions due to vital effects (e.g. of
corals, echinoids) or to localized concentrations of decomposing organic matter,
(c) carbonates formed in other environments and transported to the site of deposi-
tion, or intermixed following environmental change at the depositional site, {d)
carbonates formed by diagenetic recrystallization or cementation in pore-waters of
variable salinity.

Any of these complications can be troublesome in samples from marginal variable-
salinity environments (Keith and Parker 1965; Lloyd 1969). The present investiga-
tion relates particularly to (c) and {d). We emphasize, as did Tan and Hudson (1971),
the importance of using well-preserved skeletal carbonates wherever possible, rather
than whole-rock samples.

Extraneous CaCO^. In so far as the Weald Basin was bordered by Jurassic and Lower
Carboniferous limestones (Allen 1961, 1967a), complication (a) is always possible,
though no actual clasts of limestone have been confirmed in the Wealden of the
Weald. The misleading influence of biogenic carbonates (allochems) transported
from other environments in the same basin is illustrated by sample S8126 of Allen
and Keith (1965, fig. 1 and table 1). This sample contains, among the presumably
indigenous oysters, reworked bioclastic materials, both marine and non-marine
(echinoid plates and spines, Viviparus shells, ostracod carapaces). Any salinity
estimate based on the ratio of the whole rock (— 4-37oo) is clearly invalid.

Sediments deposited during marine transgressions across less saline environments
must generally be suspect for this reason.

The complicating effects of interstitial materials and recrystallization were in-
vestigated by analysing materials from the I.G.S. borehole No. 1 at Wadhurst Park,
Sussex (Anderson et al. 1967). Eight Neomiodon medius beds in the Wadhurst Clay,
with shells preserved wholly or partly as aragonite, were sampled from 14-20 m to
30-25 m depth. All the shell samples gave carbon isotopic compositions in the
marine range S^^C > — 27oo (Table 1). All the matrices, substantially calcite, gave
lower 8^^C values than the shells (text-fig. 1). varied in the same way. Thus the
matrix carbonates may have originated or recrystallized in less saline environments
than those of Neomiodon, or possibly came from organisms whose calcification
processes involved different degrees of isotopic fractionation (see below). Never-
theless the matrix samples, as well as the embedded shells, seem to be marine rather
than freshwater. There is only one borderline analysis (Bw 7173 matrix) with 8‘^C
less than — 27oo-

Indigenous CaCOy Even carbonates demonstrably formed in the same environment

TABLE 1 . Wadhurst Clay of Wadhurst Park : isotopic comparisons between aragonitic Neomiodon medius
shells and their more calcitic matrices (I.G.S. borehole No. 1, Grid reference TQ 6325 291 1).

Specimen Nos.

Depth

(m)

in

borehole

Components

Mineral

(X-ray)

Isotopic analysis

I.G.S.

Reading
Univ.j
Penn. St.
Univ.

8^^C(7oo)

8‘*0(7,

)o)

Bw 7113

S8129

68-426

14 20*

Shells

Matrix

Aragonite

Calcite

-|-3^95

-TT78

-T79

-3^26

Bw 7138

S8131

68-428

1704

Shells

Matrix

Aragonite
Calc, and arag.

+ 210{
-105

-f2^11t

-h2^09t

-2-35 {“
-3 30 ^

2^30f
2 39t

Bw 7154

S8132

68-429

1991

Shells

Matrix

Arag. » calc, (tr.)
Calcite

+ 2^78
-0^27

-2-12

-550

Bw7173

S8133

68-430

20^42

Shells

Matrix

Arag. = calc.
Calcite

+ 1-35
-219

-300

-5-31

Bw 7194

S8134

68-431

2T79

Shells

Matrix

Aragonite

Calcite

+ 1-61
-0 78 1

-0 65t
-090t

-2-24
-5 80|

5^50t
6 09t

Bw 7274

S8135

68-432

26-11

Shells

Aragonite

-h2-28

-3^53

Bw 7284

S8136

68-433

29-26

Shells

Matrix

Arag. = calc.
Calcite

+ 2^21
+ T46

-265

-5^98

Bw 7287

S8137

68-434

30 25t

Shells

Matrix

Calc, -h arag.
Calcite

+ 2^05
+0^31

-3-'&l

-+16

The total thickness of Wadhurst Clay is about 67 m.

t Duplicate analyses.

* Horizon c. 47 m below Lower Tunbridge Wells Sand.

I Horizon c. 4 m above Top Ashdown Pebble Bed.

-3

-2

+ 2

+ 3

20

CL

LlI

Q

30-

•b

1 1 1

•

-50

\

-60

\

\

\

*.■.■;■■■■ ■

__ __ A

A— —

-70

\

\

\ Neomiodon shell - CaC03
\

-80

matrix -CaCOj

\

\

\

-90

1

1

A

/

-3

-2

-1 0 ■H

S '^C(7oo)

+ Z +3 ■rC

A-Bw 7113

■Bw 7138

P

-Bw 7154 CO
•Bw 7173

CO
>

■Bw 7194

"D

1“

m

■Bw 7274

-Bw 7284
Bw 7287

TEXT-FIG. 1 . Wadhurst Clay of Wadhurst Park. Carbon isotopic ratios (S^^C) from
Table 1 of separated Neomiodon medius ( a ) and matrix CaCOj ( • ). I.G.S. bore-
hole No. 1.

610

PALAEONTOLOGY, VOLUME 16

are not necessarily straightforward to interpret isotopically. Some organisms,
though living under similar conditions of salinity, temperature, etc., show different
abilities to fractionate carbon and oxygen isotopes when depositing CaCOj (Keith
and Weber 1965). Decomposing organic matter adds another complication, generat-
ing CO2 with variable depending partly upon the origin of the organic matter.
A further source of variation, in some cases, is isotopic fractionation between
CO2 and CH4.

Nevertheless, we show below that the ^^C-palaeosalinities derived from Wealden
carbonates are significant when considered in terms of the broad categories of
‘marine’ and ‘freshwater’ as defined on p. 608. But the transitional environments
of the Wealden, with waters of highly variable ^®0-content and salinity, offer little
hope of yielding useful palaeotemperatures ; the temperature effect is small
compared with those of mixing and evaporation.

CaCO^ recrystallization and cementation. Diagenetic crystallization of aragonite
and calcite can modify pre-existing isotopic ratios (Gross 1964; Hodgson 1966;
Stahl and Jordan 1969; Tan and Hudson 1971, and their refs.). The necessary con-
ditions include interstitial waters with different compositions from those of the
depositional environment. Recrystallization may occur soon after deposition, for
example during evaporation from closed basins (Keith and Parker 1965), or on
tidal flats such as sabkhas, where diagenesis occurs in the presence of supersaline
porewaters. At the other extreme, transitional carbonate sediments may undergo
diagenesis in the presence of low-salinity groundwater. This was apparently impor-
tant in several of the samples described here.

Diagenesis affected the matrices of most of the samples from Wadhurst Park
No. 1. It certainly changed all three from borehole No. 2, 200 m away (Table 2).
The Neomiodon medius shells from this core are almost entirely calcite and there is
either a little or no matrix carbonate. They give lower values of than any in
borehole No. 1. This is a good example of analyses yielding no useful information

TABLE 2. Wadhurst Clay of Wadhurst Park ; isotopic ratios of dominantly calcitic Neomiodon medius
shells and matrices (I.G.S. borehole No. 2, Grid reference TQ 6308 2920 about 200 m from No. 1 ; horizon

of top about 18-64 m above top of No. 1).

Specimen Nos.

Depth

(m)

in

borehole

I.G.S.

Reading
Univ./
Penn. St.
Univ.

Bw 7353

S8138

68-435

1615

Bw 7377

S8139

68-436

16-36

Bw 7385

S8140

68-437

16-43*

Components

Mineral

(X-ray)

Isotopic analysis

S”C(%„)

8'^0(“/o„

)

Shells

Calc. > arag.

-5-17

-7-87

Shells

Matrix

Shells

Calcite
Calcite
Calc. > arag.

-2-29
-0-39
-3-37 ^

■ -3-42f
. -3-25t

-3-46

-4-68

-5-79 1 “

5-89t

■5-70t

Matrix

Calcite

-0-81 ^

' -0-91f
. -0-72t

-2-78 1“

•2-79t

-2-77t

* Horizon c. 30-6 m below Lower Tunbridge Wells Sand and c. 36-4 m above Top Ashdown Pebble Bed.
t Duplicate analyses.

ALLEN ET AL.: WEALDEN ENVIRONMENTS

611

about an original depositional environment. The shells are recrystallized, they give
lower ratios than their matrices (cf. borehole No. 1) and they straddle the
arbitrary division ( 2%o) between marine and freshwater shells.

Four analyses of Neomiodon-r ock previously published (Allen and Keith 1965,
table 1) probably reflect the combined influences of calcite cementation and the
diagenetic change of aragonite to calcite. The rocks are shelly calcareous sandstones
(Tilgate stone’), comprising quartz clasts and calcite valves cemented with clear
calcite (S8112, S8113, S8115, S8118). All give low (-8-3 to -A-Tjoo)- Three
other samples (S8109-S81 1 1) are quartz-free, virtually uncemented, and their
Neomiodon medius shells retain some aragonite. All yield substantially higher
values ( — 1-3 to +0-2°/oo)- Apparently some of the variation previously attri-
buted to a wide salinity tolerance in Neomiodon arises from the postdepositional
production of calcite through the action of groundwater.

In general, therefore, cementation or recrystallization reduced the and

180/160 ratios of the Wadhurst sediments. The same is true of the Weald Clay.
Prentice (1969, pp. 3-4) describes Weald Clay cycles and concludes that they record
upward-decreasing salinities which are parts of a general salinity series Cassiope—>
'Cyrena [presumably Filosina]^' Paludina [presumably K/vzpurw^] ^ ostracods ^
Equisetites. This series has been largely confirmed. During the course of our work
a small cycle at Ewhurst, high in the Weald Clay (Table 3), showed how effectively
recrystallization of shell aragonite could destroy original isotopic evidence.

Shells of Wealden age from Germany show the same kind of variation. Marine
Cucullaea, collected (with the ammonite Platylenticeras) from the basal Valanginian
of Sachsenhagen (1 m above the Wealden), are entirely calcite, though originally
aragonite (Hall and Kennedy 1967). Their S^^C- and S^^O-values are low ( — 2-3
and — 8-5%o)? i e- spuriously ‘freshwater’ (samples S8 128/68-444). On the other hand
Neomiodon (S8161) and Paraglauconia (S8162) from the underlying Wealden, though
similarly recrystallized to calcite, have carbon ratios (S^^C = —1-32 and — T667oo
respectively) in the arbitrary marine range.

HASTINGS BEDS

Salinity. Palaeontological evidence suggests that, of the two major Wealden sub-
divisions, the lower (or Hastings Beds) records a generally narrower range of salinities.
Prior to our isotopic work, no macrofossils of near-marine aspect were known, nor
any microfossils, though foraminifera had been alluded to in general terms (Ander-
son et al. 1967, p. 175). Salinity variations had, however, long been suspected and
sought (e.g. Allen 1962, p. 226). The basis was the observed antipathy, on single lami-
nation surfaces, between swarms of Neomiodon (Casey 1955/)) and the other fossil
assemblages. The latter, hinting at rather fresher water, include stoneworts (one
species of Circonitella: Watson 1969), liverworts (two species of Hepaticites: op. cit.),
probable liverworts preserved in situ (op. cit., fig. 13), horsetails in situ (two or three
species of Equisetites), and dominant Viviparus, Physa, Unio, or Cypridea spp.
(Anderson et al. 1967). Neomiodon was thought to have lived in more brackish con-
ditions. Its German contemporaries were already known to consort with unequivo-
cally brackish fossils (Allen 1967a, p. 60). The new isotopic evidence places Neomiodon

612

PALAEONTOLOGY, VOLUME 16

3

on

3

43

CJ

43

o

a

H

u

3

43

w

o

? ?!

Q

0-,

oj

43

'M S

“ I

cS >

u

CQ

w

w

e

£• ^

§ £.

!3.-0

s §
^ §

bo CD

— C

s s

-C X)

H >,
U-. a

O cb

X £

X

o

U

2

2

OQ

C<0

G C.<5

cn

0 —

CO

13 ?u 3

g = ^

0

0

0 0
^ 0

q

c

T3

(U

Lh

ti

X

^ 25

. 0

• .$e

1 u

(L>

U

c/5

•3 c

to 3
-2 ^
^ T3

X Q
.ex
0

Z ^

• 0

: X

1 *

lU

c/5

^ Si)

c

0

0> 'o

I>-| c/5

o> b
a2

u

cf2

ex

ex

p

G

o

>%

o

Cl ^

?l

00 ^

b ^

00 ^
o .£ o

.y 6

-J CL

o y

c

0 „

3 ^

1 §

00 o? '“'

^■5

’~X c
*£ «J

X c/5 O

“ S

c/5 U

CO

3 C

3

>)

»C5

3

CO

ALLEN ET AL.\ WEALDEN ENVIRONMENTS

613

medius squarely in our ‘marine’ category. This recalls Tan and Hudson’s work (1971)
showing that well-preserved aragonitic shells of Neomiodon from the Hebridean
Middle Jurassic have a ‘marine’ range of

The notion that salinity exerted a major control arose from the theory that the
Hastings clays were broadly transgressive and the sandstones regressive (Allen

1959) . Accepting that these movements were caused by the sea and physically con-
nected with it, the basal strata of each clay formation should record rising salinities
(Allen 1959, p. 342) and the top sands of each arenaceous formation falling salinities.
This could explain the marked antipathy between basal reedswamp and burrowed
beds (Allen 1962) and the absence from the Wealden of strictly freshwater ostracod
genera. If the lowest salinities were normally achieved during deposition of the
regressive sands then such ostracods are the least likely to have been preserved
(Anderson 1967, in lift.)-

Salinities from Wadhurst Clay. Previous isotopic analyses of aragonite-bearing
Neomiodon medius from the ? transgressive basal Wadhurst (S8109-S81 1 1, in Allen
and Keith 1965) strengthened the idea that, as the waters rose, their salinities rose
too (Si^C= - 1-3 to +0-27oo)-

Salinity fluctuations in the type-area of Wadhurst Park (I.G.S. boreholes Nos.
1-3) were independently investigated by Anderson, using ostracods (in Anderson
et al. 1967). Less saline and more saline episodes were recognized, assuming that
the former are recorded by CvpnV/ca-dominated faunas and the latter by ‘non-
Cypridea" faunas {Theriosynoecum, Mantelliana, Rhinocypris, Darwinulids, Dicro-
rygma). We attempted to test the distinction by analysing C. laevigata carapaces
(‘the nearest to a freshwater ostracod we have in the Purbeck-Wealden’) and T. alleni
(‘strongly brackish to marine’), kindly prepared and so interpreted by Professor
F. W. Anderson from the Westfield I.G.S. borehole (TQ 8204 1614, Shephard-Thorn
1971). C. laevigata gave a freshwater ratio (S^^C = — 4-47oo) but its carapaces en-
closed much secondary calcite. Those of T. alleni were too few for analysis ; associated
aragonitic fragments (molluscan?) gave a marine ratio (S^^C = +0-48%o)-

Our new analyses of well-preserved Neomiodon medius (Table 1) support the
contention that the Wadhurst waters became saline at times. Rejecting isotopic
ratios based on subordinate or no aragonite, we have eight horizons scattered over
16 m of the local 67-m succession. All their values range from + T4%o (shell
carbonate half-aragonite) to +4 0%o (carbonate wholly aragonite) and thus fall
into the ‘marine’ category.

Salinities from Gr instead Clay. Comparison of the base (? transgressive) of this forma-
tion with that of the Wadhurst Clay raises an interesting ecological question. Strik-
ingly similar in physical sedimentology, they differ in that the Grinstead base was
not colonized widely (if at all*) by aquatic horsetails (Allen 1962, p. 236-7). Prob-
ably it also contains a richer and more abundant ostracod fauna. Perhaps more
saline water explains both features.

* Rootlets (i.e. downward-branching tubules with carbonaceous linings) are found rarely and locally,
but not as commonly as previously supposed, when animal burrows were mistaken for them (Allen 1959,

1960) . The few true rootlets do not visibly originate at this level, but probably from rhizomes higher up,
e.g. the mid-Grinstead Equisetites soil bed (Gallois 1963) 6-7 m above.

614

PALAEONTOLOGY, VOLUME 16

TABLE 4. Palaeontological and isotopic analyses of ostracod carapaces (calcite) near base of Grinstead
Clay at Philpots Quarry, West Hoathly, Sussex (TQ 355 322).

Source of Palaeontological analysis

material

Microfauna Salinity

Height Lamina
above
TLTWPBj^

2-90 m

1-43 m
to

1-32 m

B

I A

101 m

Mantelliana phillipsiana domi-
nant

Cypridea recta tillsdenensis
and C. bispinosa suttin-
gensis predominantly
(S8145/7U151C)

M. phillipsiana, Cypridea
recta tillsdenensis, C. bi-
spinosa suttingensis
(S8144)

Mantelliana phillipsiana
mainly

(S8143/71-151S)
Theriosynoecum alleni domi-
nant
(S8146)

Relatively

high

Relatively

low

Intermediate

Relatively

high

Relatively

high

Isotopic analysis

S'®0 Environmental

(7oo) (7oo) category

Not analysed

-HO-58* -2-39* Marine?

-013t -2-46t

Not analysed

-I-0 05 —2-86 Marine?

Not analysed

Sample nos. given in microfauna column (Reading Univ./Penn. State Univ.)
* Treated 3 days in H2O2 to remove organic matter,
t Treated 3 days in sodium hypochlorite (5% solution).

4 Top Lower Tunbridge Wells Pebble Bed.

Our new isotopic results are compared with Professor Anderson’s palaeontological
analyses in Table 4. Laminae A, B, and C were half-millimetre partings of closely
packed ostracod carapaces, many articulated, separated by black clay (1-5 mm)
with sparse carapaces. The higher salinities predicted by the transgressive model
seem confirmed. However, the technique cannot apparently resolve all the
palaeontological diflferences recognized by Professor Anderson and for the isotopic
interpretation we assume that the ostracod calcite was primary. We have been unable
to confirm the last assumption. But it appears to be strengthened, as Dr. J. D. Hudson
points out, by our Hastings ostracods generally giving ratios that could be
depositional (see below), like the aragonitic but unlike the recrystallized shells.

Shells of 'Tornafe//a\ Neomiodon, Unio, Viviparus, etc., from the succeeding clays
prove to contain little or no aragonite. We therefore have no Grinstead samples of
primary molluscan carbonate, and isotopic analysis does not provide reliable esti-
mates of salinity. A new analysis of Unio subtruncalus shell (subsample 71-161 from
Philpots Quarry, West Hoathly, Sussex) gave no aragonite, = — 7 087oo and

- ~8-30%o-

Temperature. No reliable temperatures can be calculated because the oxygen iso-
topic compositions of the marginal waters are unknown and the temperature efifect
is small relative to those of mixing and evaporation (Keith and Parker 1965; Lloyd
1969). Nevertheless, the variation of aragonitic Neomiodon ( — 3-9 to - l-87oo)

ALLEN ET AL.\ WEALDEN ENVIRONMENTS

615

is generally within the range of modern marine mollusc shells from warm tempera-
ture waters (Keith et al. 1964). This agrees well with recent opinions concerning the
English Purbeck-Wealden ostracods (‘water temperatures . . . suggest a Mediter-
ranean type climate’: Professor F. W. Anderson 1970, in lift.), molluscs (‘warm tem-
perate’: late W. J. Arkell), land reptiles (‘subtropical’: Dr. W. E. Swinton 1970, in
Hit.), reptiles generally (‘warm temperate’ : Dr. K. A. Kermack 1970, in litt.), and flora
(‘warm temperate . . . alternations of wet and dry periods . . . drought over some
months . . . normal’: Professor T. M. Elarris 1970, in litt.). Bowen (1966), on the basis
of values from marine sediments, quoted temperatures of 17-4 °C to 22T °C for
the contemporary Neocomian seas of the Hautes Alpes, 800 km SSE. of the Weald.

WEALD CLAY

Salinity. More drastic variations in environment are indicated by the Weald Clay
fossils above the Horsham Stone. Palaeontologically, freshwater is suggested by
numerous beds with Chara, Equisetites, Cypridea, Viviparus, and Unio, often as
largely separate associations. Marine or near-marine conditions are recorded by at
least three thin bands variously containing Ostrea, Nemocardium, Mytilus, Gervillky
Corbula, Filosina, Melanopsis, Cassiope, Paraglauconia, echinoids, cirripedes, fora-
minifera, Ophiomorpha, etc. (see references on p. 607). Other horizons are domi-
nated by only one or two of these genera (Kennedy and MacDougall 1969) or by
non-Cypridea ostracods (Anderson 1963, 1968), suggesting less stable environments.
Of the ecological factors, fluctuating salinity may have been important, perhaps
limiting.

Our previous isotopic reconnaissance (Allen and Keith 1965) did little more than
support the broad picture. Theriosynoecum fittoni (identified subsequently by
Professor F. W. Anderson as dominant in sample S8127) may have lived in fresh-
water, as deduced by Kilenyi and N. W. Allen (1968, pp. 158, 162). If confirmed, this
might contrast with the Wadhurst species of the same genus, for which brackish to
marine conditions are possible (cf. Tan and Hudson 1971, table 3). Unfortunately
our carapaces of T. fittoni contained some secondary calcite; and the evidence con-
cerning T. alleni is indirect, being based on associated ‘marine’ shell fragments.

Near the arbitrary freshwater/marine boundary our original results (1965) were
suspect owing to the recrystallization of T//o5'mn-aragonite to calcite (S8120) and
to the choice of whole-rock analysis for the oyster bed (S8126). Subsequent deter-
minations of for the oyster-carbonate alone gave == +T167oo (Table 5).

Skeletal CaCOy Like their predecessors, the Weald Clay fossil carbonates were
commonly recrystallized after deposition (Table 5 and text-fig. 2).

For originally aragonitic shells, and assuming that recrystallization of aragonite
generally yielded calcite, the proportion of aragonite may be used to judge whether
the original isotopic ratios may have been retained. Unfortunately, we were unable
to establish any independent non-isotopic criteria for identifying the primary, un-
recrystallized calcite of ostracods, oysters, etc. Such criteria may eventually be
developed. Thus preliminary studies show differences in the cathodoluminescence
of organic and diagenetic carbonates, and most ostracod valves have a three-layered
structure (formed by the calcitization of a chitinous envelope) which is frequently

616

PALAEONTOLOGY, VOLUME 16

c

<u

"O

a

TD

C

2

c3

o

2 ^
2 'p

o> ^

jj 3
■S

c cx
C c«
o ^
,u! w)

^ a

O ^

Cj

cd

U x>

P3 X)

2 2

(U Xi

c/5 .5

2

a

a

U '

5

s V

<N ir>

I I

— r-

+ I

2 2
c3 c3

u u

P XJ ,

V-« .M

I 1 I

_u _o _o
U U U

s

J=

cd

U

t£;

^ dj

.Si a o

Q c: ^ Si

.C d) ^ ?3 j-

S 15, ci S:
55- 5s S; a
k; O Cj Q

I I

wm OO O

I I

u c3

< u u

o a
U U

CJ)

O

2

G

o

(D

03

2

X

"O

G

a

% §
§ O

^ §
^ CO

3

S w

c

o

*0

: <u

X)

H

B 6

CO Cd
- X

2 P
X

c3

u ^

c3

X

o

CO

c s:

m (N <N
iT5

X

VO

uo

ro ro

rf

^ J O
S C/) O

2 o 2 2

00 >. * OO 00

C/D !> w CO c/5

arag.

S8149 68-348B Capel, Surrey Cassiope br Arag. > calc. —1-20 —4-63 Marine

S8148 68-348A „ Cassiope br Calc. > arag. —2-85 —6-46

ALLEN ET AL.\ WEALDEN ENVIRONMENTS

617

destroyed by recrystallization. Our Weald Clay ostracods, moreover, show
ratios that could be depositional, as in the Hastings Beds (pp. 614-15).

As for the Hastings Beds, semiquantitative estimates of aragonite : calcite ratios
were made from X-ray diffraction patterns. Samples in which aragonite is dominant
are depicted as black rectangles in text-fig. 2. The two genera represented (Filosinal,
Cassiope) deserve particular attention as potential indicators of the environment
of deposition. Both appear, like Neomiodon, to have been ‘marine’.

Shells of the freshwater mollusc Viviparus, presumably originally aragonite, are
now calcite, and do not provide a basis for judging whether their present isotopic
composition is due mainly to conditions of deposition or diagenesis. As pointed
out above, calcite shells of Filosina (subsamples 71-152 and 71-155 belong to another
species) are sometimes deficient in and thereby isotopically indistinguishable
from recrystallized freshwater shells such as Viviparus (71-153).

Because of the non-preservation of aragonitic Viviparus shells (including those
of Allen and Keith 1965) it is still not possible either to support or refute

KEY TO TABLE 5

* Identified, counted, and interpreted by Professor F. W. Anderson (some carapaces 2- or 3-layered, many in-
filled micrite or drusy calcite).

t Provisionally identified by Dr. R. Casey (life assemblage of articulated shells, young and old),
ij; Early post-mortem cement?

/ = freshwater, br = brackish water, m = marine.

Bold type indicates samples in which the analysed carbonate was mostly aragonite (primary?).

Details of localities

colc^ I Seacliff,Sandown, Isle of Wight (SZ 620 853).

So1j7 )

cols^ i ESE. parish church, Ewhurst, Surrey (TQ 101 401).

So 154 )

58151 Vann Lane Brickworks, Hambledon, Surrey (SU 974 374).

58152 Bunce Common, Leigh, Surrey (TQ 205 466).

S8156/W1423-3 I.G.S. Warlingham boring No. 1, Surrey (TQ 349 571).

58150 Clock House Brickworks, Capel, Surrey (TQ 176 384).

58147 Graylands Brickworks, Warnham, Sussex (TQ 173 345).

S814Q T

58148 ) Clock House Brickworks, Capel, Surrey (TQ 176 384).

Stratigraphical horizons

S8157 } of Bristow 1889, 15, top line.

coi 1 1 ™ below I.G.S. Bed 8b, c. 90 m (300 ft) below top of Weald Clay (Thurrell et al. 1968).

So1j4 )

58151 I.G.S. Bed 7g?, c. 120 m (400 ft) below top of Weald Clay (Thurrell et al. 1968; Kennedy and MacDougall
1968).

58152 180 m (600 ft) below top of Weald Clay (Professor F. W. Anderson in litt.).

S8156/W1423-4 Subjacent to I.G.S. Bed 5, c. 115 m (377 ft) below top of Weald Clay (Worssam and Ivimey-
Cook 1972, pp. 22, 30, 64).

S8150 c. 10-7 m (35 ft) above Cassiope Bed (S8148-9), beneath Upper Rootlet Bed (Mr. J. D. S. MacDougall).

58147 c. 4-9 m ( 16 ft) above Gossops Green Pebble Bed, c. 91 m (300 ft) above Horsham Stone (c. 150 m (500 ft)
above Weald Clay base).

58149 I Just above Gossops Green Pebble Bed, c. 82 m (270 ft) above Horsham Stone (c. 146 m (480 ft) above

58148 / Weald Clay base (Mr. J. D. S. MacDougall in litt.)).

618

PALAEONTOLOGY, VOLUME 16

GENUS / SPECIES

Liostrea ^disforta
Ftlosina membronaceo
Filosino gregona
Neomiodon ? medius
Neormodon sp
Cassiope sp
Poroglauconio sp.

Montelhona phillipsiana momiy
Cyprideo bispinosd suttingensis / |
Cypridea recta tillsdenensis \ «
Theriosynoecum aUeru

5'X(%o)
-8 -6 -4

V

D D

I I

\k

Unio subtrur)catus
corbulid

Viviporus susseKensis
Cyprideo rolundala
Cypridea valderisis
Damonello pygmaeo

Cyprideo laevigata
Theriosynoecum fittoni
Equisetites "^burchardti
Equisetites lyelh

567o

21%

23%

D D

-U-

-V-

A

-23

-15

-10

-8

Molluscs I >50% 'original' aragonite g < 50% original' oragonite (1 bar = trace)

U >50%> skeletal (originol?) calcite

Ostrocods [J >507o skeletol (original?) calcite 0 < 50% skeletal calcite (>50% drusy spar)

Horsetoils [J calcite cement (early"?)

Q 0% 'original' oragonite (G=German)

[?] ossocioted orogonitic frogments
(moliuscon?)

TEXT-FIG. 2. Wealden as a whole. Provisional environmental classification of fossils based on of
separated shells and carapaces. (Estimated relative frequencies of species in the ostracod faunas were
supplied, with samples, by Professor F. W. Anderson. Two molluscan samples, labelled ‘G’, came from

N. Germany.)

suggestions that some viviparids tolerated brackish or marine conditions in the
Purbeck- Wealden.

Temperature. As concluded for the Hastings Beds, the variation of the aragonite
shells lies in the range of modern warm temperate molluscs and this is consistent
with recent opinion based on other palaeontological evidence.

CONCLUSIONS

(1) Whole-rock isotopic analyses are unsatisfactory for environmental studies of
marginal (transitional) sediments. Primary, skeletal, untransported carbonates only
should be used. This means in practice only unfragmented articulated aragonitic
material.

(2) One species each of Filosinal and Cassiope (from the Weald Clay) and of
Neomiodon (Hastings Beds) are confirmed as ‘marine’.

(3) Conversion to, and/or cementation with, calcite commonly reduces the S*^C

and values of Wealden molluscan shells. ‘Marine’ molluscs can therefore

ALLEN ET AL.: WEALDEN ENVIRONMENTS

619

appear as ‘freshwater’, isotopically indistinguishable from truly freshwater shells.
This was examined closely for Neomiodon medius, having been misinterpreted earlier
(Allen and Keith 1965) as resulting from the organisms’ wide salinity tolerance.

(4) Wealden skeletal carbonates giving high are therefore likely to be ‘marine’,
even if secondary (e.g. German Neomiodon and Paraglauconia).

(5) Skeletal carbonates yielding low cannot be attributed to any environment
unless shown to be primary. Freshwater molluscs now recrystallized to calcite
{Unio, Viviparus) are unrecognizable as such on an isotopic basis alone.

(6) Skeletal carbonate which was originally all, or nearly all, calcite (e.g. ostracod
carapaces, oyster shells) should be treated with caution. Recrystallization may or
may not have occurred. At present there are no certain criteria for recognizing primary
calcite or for distinguishing it from secondary calcite, though there are petrological
and isotopic grounds for optimism with the ostracods.

(7) Ignoring this, and assuming that any recrystallization would have reduced
their ^^C-ratios ((3) above), one species of the ostracod Mantelliana and two species
of Cypridea might be marine. Other species of Cypridea may be freshwater, but this
cannot be confirmed isotopically until their calcite is proved to be primary.

(8) Areal distributions of horsetail-reedswamp growing in similar substrates and
water-depths were possibly controlled by salinity. But the source of the ^^C-deficient
calcite cementing the plants in their growing positions is not known, nor when it
was precipitated. Where plants are absent, the ^^C-rich ostracod carapaces appear.

(9) New evidence supports the transgressive models for the major Hastings clay
formations (Allen 1967^, fig. 1) in that the Wadhurst and Grinstead Clays are there
seen as deposited in waters more liable to saline influxes than the intervening sands.
Later, during Weald Clay times, the salty invasions became more frequent and
extensive, so that not only many muds but also some sands were laid down in con-
ditions more saline than any of the Hastings Beds.

(10) Unequivocal palaeotemperatures cannot be deduced from Wealden carbon-
ates. Nevertheless the aragonitic shells yield ^®0-‘palaeotemperatures’ broadly
consistent with warm temperate-subtropical conditions, as deduced from the
palaeobotanical and palaeozoological evidence.

Acknowledgements. We thank the following for specimens and advice: Professor F. W. Anderson, Dr. R.
Casey, f.r.s., Mr. R. V. Melville, Dr. E. R. Shephard-Thorn, and Dr. B. C. Worssam of the Institute of
Geological Sciences; Dr. N. W. Allen, Dr. J. M. Hancock, Dr. J. D. Hudson, Dr. W. J. Kennedy, Mr.
J. MacDougall, and Dr. D. H. Rayner. We are grateful to Professor F. W. Anderson, Dr. K. A. Kermack,
Professor T. M. Harris, f.r.s., and Dr. W. E. Swinton for permission to quote their current views on the
Purbeck-Wealden climate.

REFERENCES

ALLEN, p. 1959. The Wealden environment: Anglo-Paris basin. Phil. Trans, roy. Soc., B, 242, 283-346.
1960. Geology of the Central Weald: a study of the Hastings Beds. Geol. Assoc. Centenary Guide,

No. 24.

1961. Strand-line pebbles in the mid-Hastings Beds and the geology of the London uplands. Car-
boniferous pebbles. Proc. Geol. Assoc. 72, 271-285.

1962. The Hastings deltas: recent progress and Easter field meeting report. Proc. Geol. Assoc. 73,

219-243.

620 PALAEONTOLOGY, VOLUME 16

ALLEN, p. 1967a. Origin of the Hastings facies in northwestern Europe. Proc. Geol. Assoc. 78,
27-106.

19676. Strand-line pebbles in the mid-Hastings Beds and the geology of the London uplands. Old

Red Sandstone, New Red Sandstone and other pebbles. Proc. Geol. Assoc. 78, 241-276.

and KEITH, M. L. 1965. Carbon isotope ratios and palaeosalinities of Purbeck-Wealden carbonates.

Nature, 1278-1280.

ANDERSON, F. W. 1963. In WORSSAM, B. c. 1963, 16-19.

1966o. In SHEPHARD-THORN, E. R. et ol. 1966, 82-88.

19666. New genera of Purbeck and Wealden Ostracoda. Bull. Brit. Mus. (Nat. Hist.) Geol. 11 (9),

435-446.

1967. Ostracods from the Weald Clay of England. Bull. geol. Surv. Gt. Brit., No. 27, 237-269.

1968. In THURRELL, R. G. et al. 1968, 27-30.

BAZLEY, R. A. B. and SHEPHARD-THORN, E. R. 1967. The Sedimentary and faunal sequence of the Wad-

hurst Clay (Wealden) in boreholes at Wadhurst Park, Sussex. Bull. geol. Surv. Gt. Brit., No. 27, 171-235.

ARKELL, w. J. 1947. The geology of the country around Weymouth, Swanage, Corfe, and Lulworth. Mem.
geol. Surv. Gt. Brit.

BATTEN, D. J. 1969. Some British Wealden megaspores and their facies distribution. Palaeontology, 12,
62-67.

BOWEN, R. 1966. Palaeotemperature Analysis. 265 pp. Amsterdam.

BRISTOW, H. w. 1889. The geology of the Isle of Wight (2nd edition, revised and enlarged by reid, c. and
STRAHAN, A.). Mem. geol. Surv. Gt. Brit.

CASEY, R. 1955a. The pelecypod family Corbiculidae in the Mesozoic of Europe and the Near East. J. Wash.
Acad. Sci. 45, 366-372.

19556. The Neomiodontidae, a new Family of the Arcticacea (Pelecypoda). Proc. mal. Soc. Lond.

31, 208-222.

DINES, H. G., BUCHAN, s., HOLMES, s. c. A. and BRISTOW, c. R. 1969. Geology of the country around Seven-
oaks and Tonbridge. Mem. geol. Surv. Gt. Brit.

GALLOis, R. w. 1963. In Sum. Progr. geol. Surv. Gt. Brit, for 1962, 39.

GROSS, M. G. 1964. Variations in and C^^/C^^ ratios of diagenetically altered limestones in the

Bermuda Island's. J. Geol. 72, 170-194.

HALL, A. and KENNEDY, w. J. 1967. Aragonite in fossils. Proc. roy. Soc., B, 168, 377-412.

HODGSON, w. A. 1966. Carbon and oxygen isotope ratios in diagenetic carbonates from marine sediments.
Geochim. Cosmochim. Acta, 30, 1223-1233.

KEITH, M. L., ANDERSON, G. M. and EiCHLER, R. 1964. Carbon and oxygen isotopic composition of mollusk
shells from marine and fresh water environments. Geochim. Cosmochim. Acta, 28, 1757-1786.

and PARKER, R. H. 1965. Local variation in ^^C and content of mollusk shells and the relatively

minor temperature effect in marginal marine environments. Marine Geol. 3, 1 15-129.

and WEBER, J. N. 1965. Systematic relationships between carbon and oxygen isotopes in carbonates

deposited by modern corals and algae. Science, 150, 498-501.

KENNEDY, w. J. and MACDOUGALL, J. D. s. 1969. Crustacean burrows in the Weald Clay (Lower Cretaceous)
of south-eastern England and their environmental significance. Palaeontology, 12, 459-471.

KiLENYi, T. I. and ALLEN, N. w. 1968. Marine brackish bands and their microfauna from the lower part
of the Weald Clay of Surrey and Sussex. Palaeontology, 11, 141-162.

LLOYD, R. M. 1969. A palaeoecological interpretation of the Caloosahatchee Formation, using stable iso-
tope methods. J. Geol. 77, 1-25.

MACDOUGALL, J. D. s. and PRENTICE, J. E. 1964. Sedimentary environments of the Weald Clay of south-
western England. In van straaten, l. m. j. u. (ed.). Developments in Sedimentology, 1 : Deltaic and
shallow marine deposits, 257-263. Amsterdam.

PRENTICE, J. E. 1969. Sediments: past, present and future. Inaugural Lecture, University of London, King's
College, 1-8.

SHEPHARD-THORN, E. R., SMART, J. G. o., BISSON, G. and EDMONDS, E. A. 1966. Geology of the country around
Tenterden. Mem. geol. Surv. Gt. Brit.

1971. In I.G.S. Annual Report for 1969, 20.

SMART, J. G. o., BISSON, G. and WORSSAM, B. c. 1966. Geology of the country around Canterbury and Folke-
stone. Mem. geol. Surv. Gt. Brit.

ALLEN ET AL.\ WEALDEN ENVIRONMENTS

621

STAHL, w. and Jordan, r. 1969. General considerations on isotopic palaeotemperature determinations
and analyses on Jurassic ammonites. Earth planet. Sci. Lett. 6, 173-178.

TAN, F. c. and HUDSON, J. D. 1971. Isotopic composition of carbonates in a marginal marine formation.
Nature phys. Sci. 232, 87-88.

THURRELL, R. G., woRSSAM, B. c. and EDMONDS, E. A. 1968. Gcology of the country around Haslemere.
Mem. geol. Surv. Gt. Brit.

WATSON, j. 1969. A revision of the English Wealden flora, I. Charales-Ginkgoales. Bull. Br. Mus. nat. Hist.
(Geol.). 17, No. 5.

WORSSAM, B. c. 1963. Geology of the country around Maidstone. Mem. geol. Surv. Gt. Brit.

1965. In Summ. Progr. geol. Surv. Gt. Brit, for 1964, 46-47.

and iviMEY-cooK, h. c. 1972. The stratigraphy of the Geological Survey borehole at Warlingham,

Surrey. Bull. geol. Surv. Gt. Brit., No. 36, 1-146.

and THURRELL, R. G. 1967. Field meeting to an area north of Horsham, Sussex. Proc. Geol. Assoc.

67, 263-272.

P. ALLEN

Department of Geology, University of Reading
Whiteknights, Reading RG6 2AB, U.K.

M. L. KEITH, F. C. TAN,* P. DEINES

Department of Geochemistry and Mineralogy
Pennsylvania State University
University Park, Pennsylvania 16802, U.S.A.

* Present address:

Atlantic Oceanographic Laboratory
Bedford Institute of Oceanography, Dartmouth
Final typescript received 11 October 1972 Nova Scotia, Canada

N

BUOYANCY CONTROL AND SIPHUNCLE
FUNCTION IN AMMONOIDS

by H. MUTVEi and r. a. reyment

Abstract. The question of buoyancy control of ammonoids in relation to the function of the siphuncle is analysed
in the light of flotational experiments on exact models of moderately evolute and highly evolute shell types, and the
structure of the siphuncle. It is demonstrated that, if the mode of life of the ammonoid animal were analogous to
that of living Nautilus, the relatively more buoyant shell of most ammonoids would have needed considerably more
liquid in its chambers than Nautilus, with many of the chambers completely filled. The structure of the siphuncle,
and its location in the last chambers of the majority of coiled ammonoids, is such that it may have been non-functional
in these chambers so that the animal did not vary the quantity of liquid in them. We believe that most ammonoids
were fairly efficient at moving themselves vertically but less efficient as swimmers.

Speculation about the function of the chambered cephalopod shell has pro-
vided one of the more fruitful sources of research topics in palaeontology for more
than a hundred years. Starting with Moseley’s (1838) mathematical treatise, recently
updated by Raup and Chamberlain (1967), numerous papers on the subject have
appeared. A review of earlier work is given by Westermann (1971). However, as
far as we are aware, the line of reasoning we use, and the data upon which it is founded,
are new.

Apart from attempts at finding a purely mathematical solution for the problem
(an attractive one because the shell approximates a logarithmic spiral), by deter-
mining the location of the centre of gravity and the volume of idealized and, of
necessity, unornamented shells, there have been attempts to derive conclusions by
a combination of deduction and direct observation on particular shell types. The
most noteworthy of these trials was made by Trueman (1941). The conclusions
drawn in this study have had a remarkably strong and persistent influence on workers
in the field, despite the fact that some of his suppositions, such as the asserted posi-
tive correlation between the dimensions of the body chamber and those of the
chambered portion of the test, are no longer possible to maintain (Reyment 1973).

Reyment (1958) made an experimental study of the post-mortal fate of chambered
cephalopod shells. No attempts at reconstructing the mode of life of fossil cephalo-
pods were made in this work, and all experiments were carried out on models of
straight and coiled nautiloids, as well as the shells of two living species of Nautilus.
In so far as the results of this work could be applied to buoyancy control inter-
pretations of fossil chambered cephalopods, they did not suggest anything other
than that the empty Nautilus shell can be brought into hydrostatic equilibrium with
seawater by a slight increase in the load of the shell. The work of Bidder (1962) and
Denton and Gilpin-Brown (1966) demonstrated that weight adjustments in the
shell of Nautilus, for compensating for shifts in the hydrostatic conditions, are made
by the secretion or transfer of relatively small quantities of liquid via the siphuncle
into the final chambers. The last chamber is normally full or almost full of this liquid
and earlier chambers contain progressively less and less of it. The majority of the
chambers are gas-filled and empty of liquid.

[Palaeontology, Vol. 16, Part 3, 1973, pp. 623-636.]

624

PALAEONTOLOGY, VOLUME 16

Until we compared notes on our recent research on the ammonoid shell, there
did not seem to be any reason to doubt that the ammonite animal functioned in
the same way as the living Nautilus, both with respect to the operation of the siphun-
cular mechanism and to the approximate number of chambers containing cameral
liquid. The results of our studies, Mutvei on the structure and interpretation of
the ammonoid siphuncle, and Reyment on the post-mortal properties of ammonoid
shells based on experiments on models, have shown that the currently accepted
interpretation of the hydrostatic behaviour of the ammonoid shell cannot be sustained
without modification and clarification.

Acknowledgements. The research accounted for in this paper was supported by Grant 2320-45 of the
Swedish Natural Science Research Council. The ammonoid models were made by Mr. Eric Stahl and
the illustrations were prepared by Mr. C. G. Andersson and Mrs. Dagmar Engstrom, Uppsala, and Mr.
B. Bliicher, Stockholm.

BUOYANCY OE EMPTY AMMONOID SHELLS

There are two approaches open for the study of the buoyancy of empty ammonoid
shells. One is by means of accurately constructed models of shells. The other is by
calculations of the buoyancy properties of various kinds of shells, the basic assump-
tion being that the growth of the shells has taken place in accordance with the
logarithmic spiral.

The second approach has an appeal in this day of the computer, but it has the
obvious limitation that it can only give an indication of the floating orientation of
a particular shell type. At the present stage of its development, it cannot be used to
find the proportion of the empty shell that floats above water, nor is it suited for
investigations of the critical sinking loads of particular shell types.

Study by accurately constructed models of key morphological types is the only
really satisfactory means of testing hypotheses on the relative buoyancy properties
of chambered shells and their application to palaeoecological reconstructions.
However, such models are expensive to make, requiring the assistance of a highly
skilled technician with an ability to sculpture, and suitable specimens upon which
to base the models are hard to come by. A detailed account of the way in which our
models were built is being presented elsewhere (Reyment 1973). They were made
from plastic by the method of vacuum-moulding so as to conform exactly to the
structure of the actual ammonoid shells upon which they were based. The correct
specific weight (Reyment 1958) was obtained by electroplating the model, internally
and externally, with a suitable metal.

The shell categories here studied by this method are of the ceratitic type, being
based on Ceratites nodosus (von Buch), Discoceratites intermedius (Philippi), and
Ceratites {Acanthoceratites) spinosus Philippi, selected from the extensive collec-
tions in the Paleontologiska Institution. All three possess the same basic shell
structure, namely, high whorls, moderately evolute coiling, and moderately strong
ventrolateral tuberculation : they differ from each other only in the degree of inflation
of the shell. Thus, the Discoceratites is sub-oxynote and the Acanthoceratites is
sub-cadicone.

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS

625

TEXT-FIG. 1. Floating position of Nautilus text-fig. 2. Floating position for model

pompilius. Fresh shell obtained from the of Discoceratites. Scale in cm.

Solomon Islands. Scale in cm.

The important class of highly evolute shells was studied with models based on
dactylioceratids. No particular species was used as basis for construction.

Flotational experiments. The weight increase required to cause the models of ceratitids
to just attain the point of sinking was determined. This may be done in two ways:
either by filling larger chambers with water, or by merely weighting the shell with
lead shot in the body chamber. The former method is biologically more realistic,
while the latter is more expedient and does not damage the models; it has the draw-
back that it is accompanied by an unrealistic displacement of the centre of gravity.
Although the difference in loads yielded by the two procedures is slight (they cannot
be the same because of the load to air volume ratios), the chamber-filling alternative
was selected for the present work.

Firstly we draw attention to text-fig. 1, which illustrates the floating position of
a fresh Nautilus pompilius from the Solomon Islands. This shell, which floats rela-
tively low in the water, required no more than an increase of 13% in weight in order
to sink.

Text-fig. 2 shows the floating position adopted by a model of Discoceratites. It
floats very much higher in the water than the Nautilus, 27% of the shell being above
water as compared with 10%. Approximate determinations of the buoyancy of the
empty shells show that the ceratites require weight increases of from 35% to 45%
in order to sink. They are, therefore, when empty about three times more buoyant
than the Nautilus. Text-fig. 3 compares, schematically, the load-to-sinking-point
relationships for Discoceratites and Acanthoceratites; this is linear as to be expected.
As is also to be expected, the more depressed and spacious shell requires a greater
load to make it sink than does the compressed shell.

Why these substantial differences between the nautiloid and the ammonoids? There

626

PALAEONTOLOGY, VOLUME 16

SHELL PROPORTION ABOVE WATER

TEXT-FIG. 3. Load-to-sinking curve for the
models of Discoceratites and Acanthoceratites.

are several reasons. Firstly, N. pompilius is highly involute and only the chambered
part of the last whorl contributes to the uplift (Reyment 1958). The more evolute
ceratites have a significantly greater chambered surface area available for uplift
involving part of the inner whorls. Secondly, the ceratites have a greater relative
cameral volume, owing to their higher whorl sections. Thirdly, ammonoids tend to
contain relatively less shell substance for a given conch volume than nautiloids,
owing to their mostly thinner walls : this is a well-known fact, most recently studied
by Westermann (1971). Fourthly, the ammonites studied in the present connection
have a smaller apical angle than the nautiloids, the body chamber does not ‘flare’
pronouncedly in the manner characteristic of N. pompilius and N. scrobiculatus,
for example (cf. Reyment 1958, pi. 2). In consequence, the dead weight of the body
chamber is less in the ammonoids studied than in the Nautilus species.

The differences between the buoyancies of the Nautilus shell and the ceratite
models are so great that they exceed any reasonable limits of experimental error.
It was ascertained experimentally that even with a construction error of 25% in the
shell weight, the buoyancy differences are such as would require special explanation
and interpretation.

Highly evolute shells. Two models of shells of dactylioceratid type were made. There
were two categories considered in this part of the work: those with a body chamber
occupying seven-eighths of a whorl, and those with a body chamber occupying
one-and-a-third whorls.

The floating position taken up by the highly evolute shells was shown to be highly

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS

627

TEXT-FIG. 4. Floating position for model of dactylio-
ceroid shell with a body chamber length of one-and-
one-third whorls. Scale in cm.

TEXT-FIG. 5. Floating position for model of
dactylioceroid shell with a body chamber
length of seven-eighths of a whorl. Scale in cm.

dependent on the length of the body chamber. The model shown in text-fig. 4 has
the greater body chamber. It floats as low in the water as the Nautilus depicted in
text-fig. 1. The liquid in the chambers of the living animal would not need to have
occupied more than a few of the chambers. The model shown in text-fig. 5 has the
smaller body chamber. In order for the ammonite to have been oriented vertically
during life, numerous chambers would have had to have been permanently flooded.

IMPLICATIONS OF THE BUOYANCY EXCESS OF AMMONOID SHELLS

The relative buoyancy of the ammonoid shell becomes greater with increasingly
evolute shells; this may in part be offset by an increase in the length of the body
chamber. This is a logical outcome of the increase in ‘effective volume’ of the shell
in relation to its weight. Thus, the more completely evolute the shell is, the greater
is the uplift on it, and the lesser is the downward force due to the shell weight. If
the ammonoid animal had the same mode of life as the living Nautilus, certain con-
clusions concerning the function of the hydrostatic mechanism of its shell are in-
escapable.

Empty ammonoid shells more evolute than Nautilus, with the same length of body
chamber, must have floated considerably higher in the water. We note in passing
that there is a differentiation within the species of living Nautilus', the slightly more
evolute N. scrobiculatus floats higher out of the water than does N. pompilius, although
there is a slight difference in the apical angles of the whorls of these shells (Reyment
1958).

628

PALAEONTOLOGY, VOLUME 16

It is unfortunate that the surviving genus of chambered cephalopods comprises
species whose shell morphologies are not typical of most members of the class, and
in particular the ammonoid branch. In fact the atypical nature of the Nautilus shell,
with its highly involute shape and large body chamber diameter, may be responsible
for obscuring the probable function of the ammonoid shell for so long.

For shells of the ceratitid type, with high, subrectangular whorls and moderately
evolute shape, more than half of the chambers of the last whorl, and probably all
the lower chambers of the second last whorl, must have been entirely filled with
cameral liquid if the shell functioned hydrostatically in the same manner as poised
shells of living Nautilus. This has certain, albeit minor, consequences for the floating
position adopted by the animal in life. Buoyancy adjustments must have been made
in the chambers located in the uppermost third of the vertically poised shell, and at
least in the last two to three whorls, depending on the nature of the coiling, and the
amount of cameral liquid.

THE STRUCTURE AND INFERRED FUNCTION OF THE SIPHUNCLE

The structure of the wall of the siphonal tube. The position of the siphonal tube in
most ammonoids tends to be constant. Except for the ontogenetically oldest part of
the shell, it is situated close to the ventral (anatomically posterior, outer) side of
the whorls. Its diameter is usually small. Characteristic for the majority of Mesozoic
ammonoids, as well as several Palaeozoic ammonoids, is that the septal necks change
their direction during the ontogenetic growth of the shell. Thus, in the early stages
they are directed towards the shell apex; in the later stages they point towards the
body chamber. On the basis of their direction, the septal necks are said to be retro-
siphonate and prosiphonate, respectively.

The direction of the septal necks is intimately related to the structure and origin
of the connecting rings. In Nautilus, which has retrosiphonate septal necks, each
connecting ring is composed of an outer spherulitic-prismatic layer and an inner
conchiolin layer. The spherulitic-prismatic layer is a direct continuation of that
layer in the outer part of the septal neck, whereas the conchiolin layer originates
from the nacreous layer of the septal neck (Mutvei 1964u, 1972). On the other
hand, the ammonoids with prosiphonate septal necks have connecting rings which
seem to be composed solely of a conchiolin layer (Mutvei 1967 ; Erben and Reid
1971). The calcareous prisms and the carbonate fluorapatite in the connecting rings,
reported by Birkelund and Hansen (1968) and Andalib (1972), respectively, are
probably of secondary origin. As pointed out in a previous paper (Mutvei 1967),
the conchiolin layer of the connecting rings does not originate from the nacreous
layer of the septal neck, but constitutes a separate shell unit which is secreted after
the adjacent septal necks have completed their growth. The latter condition is
obvious when we consider the growth of the septal neck. It must have projected into
a circular invagination of the body proper (text-figs. 6a, b). The epithelium which
lined the outer (peripheral) face of this invagination must have secreted the septal
neck, whereas in Nautilus the septal neck is secreted by the epithelium on the proxi-
mal portion of the siphonal cord (Mutvei 1964u). Under these circumstances, the
conchiolin layer of the connecting ring in prosiphonate ammonoids cannot be a

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS

629

B

TEXT-FIG. 6. A. Reconstruction to show the relationship between
the soft body and the shell in a prosiphonate ammonoid. b. Detail
of fig. A in a higher magnification to show the ontogenetically
youngest, prosiphonate septal neck immediately after its forma-
tion. c. Similar reconstruction as in fig. b, but at a somewhat later
growth stage, showing the formation of the primary conchiolin
membrane of the siphonal tube. In all figures the soft body is
dotted and the cameral liquid stippled, e. inv., circular invagina-
tion of the soft body; p.m., primary conchiolin membrane of the
siphonal tube.

continuation of the nacreous layer of the septal neck, as it is in Nautilus (cf. Mutvei
1964fl, 1972). There is, however, a thin conchiolin membrane, here termed the
‘primary membrane’, which covers the outer face of the septal neck and continues
as a thin tube to the succeeding septum, where it is fused to the conchiolin layer
covering the dorsal (adapical) face of that septum (Mutvei 1967, and unpublished

630

PALAEONTOLOGY, VOLUME 16

scanning electron microscope observations). The primary membrane was probably
successively secreted by the epithelium which lined the bottom of the circular in-
vagination of the soft body (p.m., text-fig. 6c). The wall of the siphonal tube in the
newly formed chamber was consequently very thin in that it consisted only of the
septal neck and the primary membrane (text-fig. 7a). This is in sharp contrast with
the condition found in Nautilus, where the distal portion of the septal neck and the
contiguous connecting ring reach their maximum thickness when only about one-
third of the total thickness of the last septum has been attained (Denton and Gilpin-
Brown 1966; Mutvei 1972).

TEXT-FIG. 7. A. Reconstruction of an early growth stage in a pro-
siphonate ammonoid when the wall of the siphonal tube was
composed of the septal necks and primary conchiolin membrane
only. B. Fully developed wall of the siphonal tube.

Inferred function. The above-mentioned differences in the growth of the wall of the
siphonal tube between the retrosiphonate Nautilus and the prosiphonate ammonoids
have an important functional significance. As demonstrated by Denton and Gilpin-
Brown (1966), the new chamber is always completely filled by liquid. This liquid
can be pumped osmotically into the siphonal cord through the permeable connecting
ring. In Nautilus, with its heavy shell, the removal of the liquid from the new chamber

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS 631

is accomplished at the fastest possible rate, in order to maintain the buoyancy of
the animal, because, as the soft body and the shell continue to grow, they increase
in weight. However, this cannot take place before the septal neck, and particularly
the contiguous, permeable connecting ring have reached their full thickness and
strength to withstand the hydrostatic pressure of the sea. Therefore, the wall of the
siphonal tube is required to grow faster than the last septum. On the other hand, as
just shown, the shell of most ammonoids was considerably more buoyant than that
of Nautilus, and must consequently have contained much more cameral liquid
during life in order to have been in hydrostatic equilibrium with seawater. As a
result, there was not the same pressing need for the removal of liquid from the shell
chambers and the growth of the connecting rings was retarded (Mutvei 1967).
Evidence for the latter condition is that the connecting ring in each chamber was
deposited on the inner faces of the two successive septal necks, and consequently
secreted by the epithelium of the siphonal cord after the formation of these septal
necks and the primary membrane was completed (text-fig. 7b). Both ends of the
connecting ring are fused to the adjacent septal necks by a calcareous structure, the
auxiliary ridge (text-fig. 7b). The auxiliary ridge is also present in Nautilus, but
here it only effects the fusion of the distal end of the succeeding connecting ring to
the preceding septal neck (Mutvei 1972). As in Nautilus, the emptying of cameral
liquid in the ammonoids could have taken place only after the connecting rings had
reached their full strength. At which stage this has occurred is still obscure. How-
ever, the connecting ring of varying numbers of ontogenetically youngest chambers
(up to one whorl) are often absent (Trueman 1920; Westermann 1971), which might
indicate that there they were not fully developed and were therefore easily destroyed
during diagenesis.

Denton and Gilpin-Brown (1966, 1971) have shown that Sepia and Spirula are
capable of regulating the volumes of the liquid in their chambers. This is probably
also true for Nautilus. For such regulation of the cameral liquid in the ammonoids,
a ventral position of the siphonal tube would have been the most advantageous
arrangement, and this is supported by the experimental work.

FUNCTIONAL ANATOMY

A comparison between certain shell characters of Nautilus and ammonoids allows
the following conclusions on the functional anatomy of the latter.

The body chamber of many ammonoids is much longer than that of Nautilus.
The soft body in these ammonoids must therefore also have been correspondingly
longer, and in some forms worm-like (text-figs. 8a and b).

The short-bodied Nautilus is well adapted for swimming by jet-propulsion. The
mantle fold in Nautilus is attached to the shell aperture by the periostracum (Mutvei
1964fl). This is a general feature for all molluscs which secrete an external shell.
Unlike the dibranchiate cephalopods with internal shells, the mantle fold of Nautilus
therefore contains only a thin muscular layer, the contractions of which are in-
sufficient to drive out the water from the mantle cavity. The ammonoids certainly
had the same relationship between the mantle fold and the shell aperture as Nautilus.

The swimming mechanism in Nautilus has not yet been fully explained. For an

632

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 8. A. Diagrammatic representation of the anatomy in
the recent Nautilus, b. Reconstruction of the anatomy in an
ammonoid. c.c., cephalic cartilage; ct., ctenidia; /., funnel;
m.c., mantle cavity; r.m., paired retractor muscles; r.m.ct., an
unpaired muscle, probably representing the retractor muscle of
ctenidia.

understanding of this mechanism, the following anatomical features must be taken
into consideration. The roof above the spacious mantle cavity is formed by a pair
of powerful retractor muscles, which originate from the lateral, inner faces of the
shell wall {r.m., text-fig. 8a). These muscles extend to the head, where they are
rigidly inserted into the cephalic cartilage (c.c., text-fig. 8a). The function of these
muscles is not only to attach the soft body to the shell, and withdraw it into the
body chamber, as with the columellar muscles in the gastropods. They are also an im-
portant part of the swimming equipment of the animal (Griffin 1900; Mutvei 1964/)).

In order to test the latter assumption, experiments with anaesthetized animals

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS 633

of the dibranchiates, Sepia and Loligo, were carried out by H. M. in the summer
of 1965 at the Marine Biological Station (ARAGO) at Banyuls-sur-Mer, France.
Like other dibranchiate cephalopods, Sepia and Loligo have a highly muscular
mantle, the contractions of which are sufficient to produce a powerful water jet.
In order to study the action of the mantle musculature, and the retractor muscles
of the head and funnel, the mantle of the anaesthetized animals was cut apart so
that the mantle cavity was exposed. On touching the hypobranchial ganglia with
a needle, rhythmic, simultaneous contractions of all these muscles take place. In
all likelihood, similar simultaneous contractions of the muscles in question occur
when the animals swim rapidly, as when escaping from an enemy. These experiments
make it reasonable to conclude that when Nautilus swims rapidly, the retractor
muscles do not remain inactive, but by their contractions create the main force for jet-
propulsion. This conclusion is in agreement with the above-mentioned ‘roof-position’
of these muscles above the mantle cavity. During the muscular contractions, the head
of the animal is probably slightly withdrawn into the shell, and the roof of the mantle
cavity lowered. This would cause a considerable decrease in the volume of the mantle
cavity, as a result of which the water is forced out through the funnel (cf. Griffin 1900;
Mutvei 19646). For slow swimming movements, and for respiration, the water is
expelled from the mantle cavity by contractions of the funnel (Bidder 1962).

The myo-adhesive scars for the attachment of the paired retractor muscles to the
shell wall have been described in several Mesozoic ammonoids by Crick (1898),
Jones (1961), and Jordan (1968). These scars have a constant position, in that they
are always situated on the dorsal (anatomically anterior) face of the body chamber,
irrespective of the shape of the shell. Contrary to this, the scars of the retractor
muscles in the fossil ‘nautiloids’ have a different number and position in different
groups (Mutvei 1957, 19646). As in Nautilus, the paired retractor muscles of the
ammonoids quite probably extended to the head, where they were inserted into the
cephalic cartilage (r.m., c.c., text-fig. 8b; see also Mutvei 19646, fig. 8). Owing to
their dorsal position, and to the curvature of the soft body, these muscles must have
been situated in the dorsalmost portion of the body over most of their extension,
close to the dorsal (anatomically anterior) wall of the body chamber (text-fig. 8b).
One may therefore conclude that their topographic relationship to the main mantle
cavity was different from that in Nautilus, and this being so, they could not have
formed a roof above this cavity (text-fig. 8b; see also Mutvei 19646, fig. 8e).

To judge from the probable worm-like body shape of many ammonite species, the
main mantle cavity in the ammonoids must also have been long and comparatively
narrow {m.c., text-fig. 8b), and thus different from the short and broad mantle
cavity in Nautilus {m.c., text-fig. 8a). The number of the ctenidia is still unknown
in the ammonoids, but if they were present, their length would probably have been
positively correlated with the length of the mantle cavity. Thus, instead of the
rather short ctenidia of Nautilus {ct., text-fig. 8a), the ctenidia in the ammonoids
may have been long {ct., text-fig. 8b).

Taking into account the shape of the mantle cavity, and the topographic relation-
ship between this cavity and the paired retractor muscles, one may conclude that
most ammonoids would have been incapable of efficient swimming by jet-propulsion.
Also, the presence of a funnel, necessary for jet-propulsion and steering, is doubtful

634

PALAEONTOLOGY, VOLUME 16

for certain ammonoids, which in the adult develop an unpaired, ventral, keel-like
projection of the apertural margin. On the other hand, the fossil ‘nautiloids’ are
always provided with an opening for the funnel. The ammonoids were naturally
capable of vertical movement in the sea by means of regulation of the volume of
the liquid in the shell chambers.

In addition to the paired retractor muscles, the Mesozoic ammonoids also had
a small, unpaired muscle in the ventral (anatomically posterior) portion of the body
(Jones 1961; Jordan 1968). The myo-adhesive scar for the latter muscle has not
yet been found in the Palaeozoic ammonoids (Crick 1898; unpublished observa-
tions of H. M.). The extension and function of this unpaired muscle are unknown,
but judging from its general position, it may have been homologous with the retractor
muscles of the ctenidia in dibranchiate cephalopods {r.m.ct., text-fig. 8b).

POST-MORTAL FATE OF AMMONOID SHELLS

It is now well known that even recently dead Nautilus shells do not contain any
cameral liquid (Bidder 1962; Denton and Gilpin-Brown 1966). How the post-
mortal assimilation and loss of this liquid takes place is still a mystery.

After the death of the animal, the gases deriving from the processes of decomposi-
tion of the carcass soon expel water from the body chamber and inflate the decaying
soft parts; the dead animal is driven to the surface. The time taken for this is rela-
tively short, usually some few hours. It is therefore clear that at this stage, in the
immediate post-mortal phase, the liquid in the chambers remains unaltered in
volume. Within a few days at the most, the body parts company with the shell and
each passes to its separate fate.

Significant conclusions arising here are that : (1) The ammonoid shell with cameral
liquid must have been slightly lighter than water (by analogy with living shell-
bearing cephalopods); (2) The density of the body of the ammonoid animal must
have been higher than that of water (by analogy with living cephalopods); (3) Irre-
spective of its structure, the shell and carcass would normally have been forced to
the surface shortly after death.

Whether or not the ammonoid lost its cameral liquid during the initial post-
mortal phase is a matter for conjecture. There is a reasonable likelihood that at least
part of the liquid may have been dissipated during this phase, if the observations
made on Nautilus have any generality.

There is ample evidence for the vertical embedding of fossil cephalopod shells in
sediments (Reyment 1970), and there is little doubt that many shells stranded in
semifluid sediment or ooze in their vertical floating positions. In view of the com-
parative lightness of many ammonoid shells, their relatively high buoyancy, and
the necessary requirement of their being lighter than water during life, it is difficult
to envisage a situation in which the shell would sink post-mortally without dis-
playing evidence for implosion.

CONCLUDING REMARKS

The structure of the ammonoid siphuncle suggests that it functioned in the same
manner as that of living Nautilus, albeit with some modification. The development

MUTVEI AND REYMENT: BUOYANCY OF AMMONOIDS

635

from retrosiphonate to prosiphonate septal necks tends to be associated with in-
creasing complication of the septa. Secondary simplification of the connecting rings
and their retarded growth appears to have been a part of this change. This structural
shift could have hardly come about if the ammonoid shell had not evolved in the direc-
tion of a buoyancy excess, offset by an increase in the volume of permanent cameral
fluid in the living animal. The buoyancy regulation of ammonoid shells probably
took place only in those chambers in which the siphuncle had a ventral location, and
the cameral liquid, therefore, was in continual contact with the siphuncle. This could
be the reason why the majority of the ammonoids had a ventrally located siphuncular
tube and why the porous calcareous layers of the siphuncles of many nautiloids, which
in living Nautilus function in the manner of a ‘wick’, are absent in ammonoids.

The situation just described is applicable to a large number of ammonoid types.
However, extremely evolute shells sometimes have body chambers in excess of one
whorl length (Reyment 1973). The effect of this is to offset the uplift of the chambered
whorls. It is, we believe, certain that this type of shell must have contained consider-
ably less cameral liquid than other ammonoids, and could therefore have had a
specialized mode of life.

Some of our work may cast light on the reason why ammonoids tended to develop
complicated cameral sutures with the shell wall. It is no new idea that sutural com-
plications evolved as a means of withstanding water pressure while retaining an
optimal thinness of the shell wall. The present study has indicated the strong possi-
bility of the ammonoids having been poorly suited for swimming and more adapted
for vertical movement, perhaps connected with a mode of life related to following
the diurnal migration of plankton upon which they may have fed. Small size of the
prey is indicated by the shape of the jaws (Kaiser and Lehmann 1971). It might
therefore be possible that the primary mode of life of the ammonoids involved the
need for continual adjustments to a pressure gradient. There is some indirect evi-
dence in support of this interpretation. We know that some ammonites definitely
inhabited a shallow-water environment throughout their lives and that they could
not have been confronted with the need to adjust to important differences in water
pressure. Is this relaxation in the suggested primary mode of life of ammonoids
reflected in the suture line?

There are several groups of Cretaceous ammonites that inhabited, permanently,
shallow seas. The best known of these is the group of the vascoceratids, the main
development of which took place in north-western Africa and the Iberian Peninsula
during the Lower Turonian. The vascoceratids have often been said to possess a
‘degenerated acanthoceratid suture’. The suture of almost all vascoceratids is indeed
greatly simplified in comparison with that of its antecedents and it is not inconceiv-
able that this could have resulted as a genetical response to the pressure-independent
mode of life followed by these ammonites in the shallow, epeiric, trans-Saharan sea
of the Lower Turonian. This vast inland sea began to develop during the late Ceno-
manian. At that time it contained the genus Neolobites, characterized by a ‘pseudo-
goniatitic’ suture. It reached its maximum extension during the Lower Turonian,
during which time it stretched from North Africa, across the Saharan region, to
Nigeria. The major part of this transcontinental ocean was extremely shallow: its
average depth can hardly have exceeded 10 metres, as indicated by the spread of

636

PALAEONTOLOGY, VOLUME 16

the sediments in relation to the Upper Cretaceous topography. The shells of most
vascoceratid species are more stoutly built than is usual in ammonoids and this
would suggest that they were adapted to withstand mechanical damage, resulting
from wave action in a shallow environment, rather than the force of water pressure.

REFERENCES

ANDALiB, F. 1972. Mineralogy and preservation of siphuncles in Jurassic cephalopods. N. Jb. Geol. Paldont.,
Abh. 140, 33-48.

BIDDER, A. M. 1962. Use of the tentacles, swimming and buoyancy control in the pearly Nautilus. Nature,
196, 451-454.

BIRKELUND, T. and HANSEN, H. J. 1968. Early shell growth and structures of the septa and the siphuncular
tube in some Maastrichtian ammonites. Medd. Dansk Geol. Foren. 18, 71-78.

CRICK, G. c. 1898. On the muscular attachment of the animal to its shell in some fossil Cephalopoda (Am-
monoidea). Trans. Linnean Soc. 7, 71 113.

DENTON, E. J. and GILPIN-BROWN, J. B. 1966. On the buoyancy of the pearly Nautilus. J. mar. biol. Ass.
U.K. 46, 723-759.

1971. Further observations on the buoyancy of Spirula. Ibid. 51, 362-373.

ERBEN, H. K. and REID, E. H. 1971. Ultrastructurc of shell, origin of conellae and siphuncular membranes
in an ammonite. Biomineralisation, 3, 22-31.

GRIFFIN, L. E. 1900. The anatomy of Nautilus pompilius. Mem. nat. Acad. Sci. 8, 103-197.

JONES, D. L. 1961. Muscle attachment impressions in a Cretaceous ammonite. J. Paleont. 35, 502-504.
JORDAN, R. 1968. Zur Anatomic mesozoischer Ammoniten nach den Strukturelementen der Gehause-
Innenwand. Beih. geol. Jb. 77, 1-64.

KAISER, p. and LEHMANN, u. 1971. Vergleichcnde Studien zur Evolution des Kieferapparates rezenter und
fossiler Cephalopoden, Paldont. Z. 45, 18-32.

MOSELEY, H. 1838. On the geometrical forms of turbinated and discoid shells. Phil. Trans. Roy. Soc. for
1838, 351-370.

MUTVEi, H. 1957. On the relations of the principal muscles to the shell in Nautilus and in some fossil nauti-
loids. Ark. Miner. Geol. 2,219-254.

1964a. On the shells of Nautilus and Spirula with notes on the shell secretion in non-cephalopod

molluscs. Ark. Zool. 16, 221-278.

19646. Remarks on the anatomy of recent and fossil Cephalopoda. Stockh. Contr. Geol. 11, 79-112.

1967. On the microscopic shell structure in some Jurassic ammonoids. N. Jb. Geol. Paldont., Abh.

129, 157-166.

1972. Ultrastructural studies on cephalopod shells. Part I, The septa and siphonal tube in Nautilus.

Bull. geol. Instn. Univ. Uppsala, N.s. 3, 237-261.

RAUP, D. M. and CHAMBERLAIN, J. A. 1967. Equations for volume and center of gravity in ammonoid shells.
J. Paleont. 41, 566-574.

REYMENT, R. A. 1958. Factors in the distribution of fossil cephalopods. Stockh. Contr. Geol. 1, 91-184.

1970. Factors in the distribution of fossil cephalopods, II. Vertically imbedded cephalopod shells.

Palaeoec. Palaeogeogr. Palaeoclim. 7, 103-111.

1973. Factors in the distribution of fossil cephalopods: Part 3. Bull. geol. Instn Univ. Uppsala, N.s. 4,

7-41.

TRUEMAN, A. E. 1920. The ammonite siphuncle. Geol. Mag. 57, 26-32.

1941. The ammonite body chamber with special reference to the buoyancy and mode of life of the

living ammonite. Q. Jl geol. Soc. Lond., 96, 339-383.

WESTERMANN, G. E. G. 1971. Form, structure and function of shell and siphuncle in coiled Mesozoic am-
monoids. Life Sci. Contr. R. Ontario Mus., 78, 1-39.

H. MUTVEI R. A. REYMENT
Paleontologiska Institutionen
Uppsala Universitet
Box 558

S-75122 Uppsala

Typescript received 10 July 1972 Sweden

AN UNUSUAL AGGLUTINATING
FORAMINIFER FROM THE
UPPER CRETACEOUS OF ENGLAND

by C. G. ADAMS, R. H. KNIGHT, and R. L. HODGKINSON

Abstract. Labyrinthidoma dumptonensis gen. et sp. nov., belonging to the Superfamily Lituolacea, is described from
the Chalk of southern England. It has a distinctive mode of growth, chambers with labyrinthic interiors, and cannot
be fitted into any of the existing families of the Lituolacea as presently defined.

The species described in this paper was first found by one of us (R. H. K.) in the
Chalk of Dumpton Bay, Broadstairs, Kent. An intensive search subsequently led
to the discovery of numerous additional specimens within the Micraster coranguinum
Zone in the coastal sections between the Western Undercliff, Ramsgate, and Kings-
gate Bay, Broadstairs (text-fig. 1). For details of the local stratigraphy see Peake
(1967a, b).

The morphological characters of this species do not conform with those of any
previously described genus, and the generic characters are such that the species
cannot be fitted readily into the subfamilies of the Lituolidae as presently defined.
However, since this merely reflects the imperfection of the existing classification,
which is based on genera and species described mainly on their external form, a
new family or subfamily is not erected here.

It may, perhaps, be thought curious that this relatively large foraminifer should
have passed unnoticed in Europe where so much attention has been paid to faunas
of Senonian age. However, it has been found in old, undescribed collections in the
British Museum (Natural History) under the name Lituola nautiloidea (Lamarck),
a species to which it bears a superficial resemblance, and we have no doubt that it
will also be found in similar collections elsewhere.

All type, described and figured specimens (except PI. 78, fig. 8) are deposited in
the collections of the British Museum (Natural History).

Acknowledgements. The authors wish to thank Dr. F. T. Banner (Swansea University) and Mr. D. J.
Carter (Imperial College) for examining and eommenting helpfully on some of the material. Thanks are
also due to Dr. H. Malz, Senckenberg Museum, Frankfurt am Main, for facilitating the loan of compara-
tive material. One of us (R. H. K.) is grateful for permission to undertake part of this work at Thanet
Technical College.

Superfamily lituolacea de Blainville 1825
Genus labyrinthidoma nov.

Type species. Labyrinthidoma dumptonensis nov.

Derivation of name. From the Greek, meaning a labyrinthic house.

Diagnosis. Test free, agglutinating. Initially coiled streptospirally in the megalospheric
form, later becoming uncoiled. Microspheric form similar, but with a trochospiral

[Palaeontology, Vol. 16, Part 3, 1973, pp. 637-643, pi. 78.]

O

638

PALAEONTOLOGY, VOLUME 16

and/or biserial stage prior to becoming streptospiral. Most chambers labyrinthic;
wall canaliculate but non-labyrinthic, not composed of a distinct epidermis and
hypodermis. Aperture cribrate.

Remarks. The labyrinthic interior, ‘non-labyrinthic walls’ (see below), strepto-
spiral coil, and non-adherent habit distinguish this genus from all other described
lituolaceans.

Labyrinthidoma dumptonensis sp. nov.

Plate 78, figs. 1-12; text-figs. 2 and 3

Material. More than 300 specimens, of which three are known to be microspheric.

Diagno.sis. As for the genus.

Holotype. P. 48623 (PI. 78, fig. 3); a megalospheric form.

ADAMS ET AL.\ AGGLUTINATING FORAMINIFER

639

Horizon. Senonian (Upper Coniacian or Lower Santonian: see Barr 1966). Upper Micraster coranguinum
Zone, Bedwell’s Columnar Band (see Peake 1967a).

Locality. North of Dumpton Gap, Dumpton Bay, Isle of Thanet, Kent (exact position unknown).

Description.

Megalospheric form. Test finely agglutinating, with a prominent involute streptospiral initial coil com-
prising numerous chambers, and a rectilinear portion with up to fifteen chambers; all chambers much
wider than high, chambers of uncoiled portion usually subcircular in cross-section. Sutures slightly de-
pressed. Chamber walls thick, composed of agglutinated chalk grains, shell debris and other microfossils.

TEXT-FIG. 2. Typical external
appearance of Labyrinthidoma
dumptonensis gen. et sp. nov.

and characterized by the presence of numerous randomly arranged cytoplasmic canals. These do not
open to the exterior except in slightly abraded specimens. Proloculus followed by up to eight simple cham-
bers; all later chambers have labyrinthic interiors. In the rectilinear part of the test, short, stout, irregular
vertical partitions project inwards from the wall. The lumen of each chamber is partly occluded with
pillars and sinuous partitions of irregular width (PI. 78, fig. 8). Both pillars and partitions are often per-
forated by coarse pores. Coil streptospiral, very difficult to see clearly owing to the labyrinthic nature of
the chambers. Aperture cribrate with numerous circular or elongate pores, usually on a slightly convex
apertural face.

Dimensions of holotype.

Length 4 0 mm.

Width of rectilinear portion 2-2 x 2-8 mm.

Width of coiled portion 2-7 x 2 0 mm.

Height of chambers in rectilinear portion up to 0-50 mm.

640

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 3. Semi-diagrammatic representations of the
initial stages of three microspheric individuals, (a) Speci-
men with a clear trochospiral or biserial stage; (b and c)
Two dissected specimens showing some of the early cham-
bers. The uniserial appearance is believed to be an accident
of dissection.

Microspheric form. Externally similar to the megalospheric form. Proloculus followed by a high trocho-
spiral coil or biserial stage (up to 0-53 mm in length) comprising at least 1 1 chambers. This is followed by
a streptospiral coil in which the chambers quickly become labyrinthic. The last few chambers are recti-
linear as in the megalospheric generation.

The earliest stage of growth is difficult to describe accurately from the three specimens presently avail-
able. The best individual (PI. 78, figs. 4 and 5; text -fig. 3u) starts with a high trochospiral or biserial stage
while the other two appear to be uniserial— a condition probably resulting from damage during dissection.

Variation.

Proloculus. Sectioned and dissected specimens show that the megalospheric form has a large subglobular
proloculus. Although accurate measurements have not been possible, the internal diameter appears to
be over 0-4 mm in the five specimens in which it is visible.

EXPLANATION OF PLATE 78

Figs. 1-12. Labyrinthidoma dumptonensis sp. nov. 1, Dissection of megalospheric form showing strepto-
spiral coil and several uncoiled chambers, x 12. P.48627. Bedwell’s Columnar Band, Western Under-
cliff Promenade, Ramsgate, Kent. 2, External view of large specimen, x 10. P.48628. South of Joss
Bay, Broadstairs, Kent. See also fig. 12. 3, Holotype, x 10. P.48623. The partially dissected initial coil
shows a cavity which may mark the position of the proloculus. Cf. fig. 1 . North of Dumpton Gap,
Kent. 4, Thin section of microspheric form showing three phases of growth beginning with a high
trochoid spire orbiserial stage, xll-5. P.48638. South of Dumpton Gap, Kent. 5, Enlargement of
initial stage of fig. 4, x33. 6, Branched specimen with partly abraded surface, x9-5. P.48630. In flint
pebble from beach at Pegwell Bay, Kent. 7, Thin section through uncoiled chambers showing internal
partitions and wall structure, x 12. P.48639. 8, Transverse cut through an uncoiled chamber showing
vertical internal pillars and short radial plates, x 10. Specimen destroyed during serial sectioning. 9, In-
ternal cast of the uncoiled portion of a test after removal of the wall with dilute acid, x 10-5. P. 48625.
10, Typical apertural view, x 12. P.48626. 1 1, Part of a septal face showing canals and spongy texture,
X 1 30. P.48947. 12, Apertural face of specimen shown in fig. 2, x 10. P.48628. Note incipient branching.

All in situ specimens from the Upper Chalk, Micraster coranguinum Zone, Kent.

PLATE 78

ADAMS, KNIGHT, and HODGKINSON, Labyrinthidoma

642

PALAEONTOLOGY, VOLUME 16

Spire. The initial coil is always tight in the megalospheric form and usually comprises 2-3 streptospiral
whorls. However, the plane of coiling appeared to change abruptly by 90° in one specimen, while in another
a reversal of coiling was observed. The high trochospiral or biserial stage of the microspheric form has
been seen in three individuals and is not the result of fortuitous sections through specimens which happen
to have incorporated the shell of another species in the wall. Bartenstein (1952) illustrated a specimen of
Lituola irregularis irregularis (Roemer) from the Campanian of Hannover which seemed to have grown
round a Heterohelix globulosa seen in the centre of the spire. He noted that this was quite a common
phenomenon, but did not indicate whether the individuals concerned were microspheric or megalospheric.
Kaever (1971) described specimens which undoubtedly include a biserial stage in microspheric forms from
the Santonian of Dortmund-Eving. The present specimens resemble those described by Kaever in their
mode of growth, but not in their internal structure.

Uncoiled portion. This varies considerably in length and may comprise up to eighteen chambers. Some
broken specimens, believed to belong to this species but lacking an initial coil, possess up to 26 rectilinear
chambers. A few of the larger specimens either branch or show a tendency to branch (PI. 78, figs. 6 and 12).
Branched individuals have the same wall structure and labyrinthic interior as unbranched specimens.

Apertures. Up to 31 (normally 16-18) circular, subcircular or elongate pores have been observed in some
of the larger specimens. They are usually irregularly disposed over the apertural face (PI. 78, fig. 10) which
is often slightly convex.

Wall. Finely granular, incorporating shell fragments, spicules, small foraminifera (miliolids, bolivinids,
and globigerinids), and chalk grains. Abraded or broken surfaces reveal large numbers of fine cytoplasmic
canals from 10-40 /j.m in diameter. They are randomly oriented and are particularly well seen on broken
septal faces. The wall is non-labyrinthic in the sense of Loeblich and Tappan (1964, C61), since it does
not contain interlaced dendritic channels perpendicular to the surface. It is, nevertheless, spongy, albeit
on a finer scale than in species with typical labyrinthic walls.

Labyrinthic interior. The chamber roofs are supported peripherally by short, stout, radial plates, and
centrally by vertical, thick, smoothly finished, sinuous, irregular partitions and pillars. These partially
occlude the lumen of each chamber, but leave intercommunicating spaces of irregular size and shape
into which the apertures open.

Size. Length 1-9 10 0 mm (mean 4-7 mm).

Length (uncoiled portion) up to 7-2 mm (mean 2-5 mm).

Width (coil) T5-4-6 mm (mean 2-8 mm).

Width (uncoiled portion) up to 3-6 mm.

Number of chambers visible in coil 8-22 (mean 15).

Number of chambers visible in uncoiled portion up to 18 (mean 7).

Measurements made on forty well-preserved specimens. Broken individuals suggest that some may have
been larger than these measurements indicate.

Affinities.

Labyrintliidoma clearly belongs to the Superfamily Lituolacea, but does not fit
readily into any of the families or subfamilies as presently defined.

Pokorny (1958) appears to be the only person to have figured specimens which
may be conspecific with those described here. He referred his material to Coscino-
phragma cribrosa (Reuss). However, although there is some doubt about the accuracy
of Reuss’s original description, Coscinophragma is believed to be an adherent genus
lacking an initial coil and possessing labyrinthic walls (Loeblich and Tappan 1964).
It gives its name to the subfamily Coscinophragmatinae (family Lituolidae). The
types of C. cribrosa are from the Upper Cretaceous of Czechoslovakia.

Secondary transverse septa occur in some members of the family Lituolidae, e.g.
Labyrinthina Weynschenk, while at least one genus, Navarella Ciry and Rat, has a

ADAMS ET AL.\ AGGLUTINATING FORAMINIFER

643

fully streptospiral initial coil. On the other hand, no known member of this family
has a biserial or trochospiral initial stage.

The family Ataxophragmiidae includes Coprolithina Marie, a genus with radial
partitions and a trochospiral coil but lacking internal pillars. Only in the Pavoni-
tinidae is the complexity of the internal structure similar to that of Labyrinthidoma,
but the mode of growth of the present specimens, particularly in the initial and adult
stages, is rather different from anything known in this family.

The fact that the microspheric form of Labyrinthidoma combines characters best
seen in three different families of the Lituolacea reflects our ignorance of the mode
of growth of this generation throughout the superfamily and indicates that the
classification is defective. For this reason no attempt is made to assign Labyrinthidoma
to a family.

The existence of a biserial stage in Lituola irregularis prompted Kaever (1971)
to speculate that it might represent a planktonic phase in the life cycle. While this
is possible, it is more likely to indicate that the ancestral form is to be found amongst
the numerous small biserial and triserial agglutinating benthic species known from
the Mesozoic.

REFERENCES

BARR, F. T. 1966. The foraminiferal genus Bolivinoides from the Upper Cretaceous of the British Isles.
Palaeontology, 9, 220-243, pis. 34-38.

BARTENSTEiN, H. 1952. Taxonomische Bemerkungen zu den Ammobaculites, Haplophragmium, Lituola
und venvandten Gattungen. (For.). Senckenbergiana, 33, 313-342, pis. 1-7.

KAEVER, M. 1971. Recherches sur le Foraminifere Lituola irregularis (Roemer). C.r. somm. Seance. Soc.
geol. Fr., 35-37.

LOEBLiCH, A. R. and TAPP AN, H. 1964. Sarcodina chiefly Thecamoebians’ and Foraminiferida. In moore,
R. c. (ed.), Treatise on Invertebrate Paleontology, Part C, Protista 2. Kansas.

PEAKE, N. B. 1967a. The Coastal Chalk of North-east Thanet. Itinerary 2, Geol. /I55. Guides No. 30B; The
London Region (South of the Thames), 14-19.

1967Z). North Kent Coast— Pegwell Bay. 2. The Chalk in Pegwell Bay. Itinerary 8, Ibid. 30-31.

POKORNY, V. 1958. Grundzuge der zoologischen Mikropaldontologie, Bd. 1, v-xii-|-582 pp., figs. 1-549.
VEB Deutscher Verlag der Wissenschaften, Berlin.

C. G. ADAMS and r. l. hodgkinson
Department of Palaeontology
British Museum (Natural History)
London SW7 5BD

R. H. KNIGHT

34 Foads Lane
Cliffsend

Ramsgate, Kent CT12 5JP

Typescript received 14 July 1972

^■Y'

•Vt>".

-v’'- i

ui

.• >

SCANNING ELECTRON MICROSCOPY OE
LATEX CASTS OF FOSSIL PLANT
IMPRESSIONS

by W. G. CHALONER and M. M. GAY

Abstract. Rubber latex casts of fossil lycopod stem impressions in fine-grained matrices may be subjected to scanning
electron microscopy to reveal details of the original epidermal structure. This technique offers the potential of
obtaining microscopic detail from plant impression fossils even if the cuticle is not preserved.

Impressions of plant surfaces on a rock matrix are generally regarded as a rather
poor and uninformative type of fossil. Where some of the coalified plant material
has survived, from which a cuticle may be prepared by maceration (a ‘compression
fossil’), then this may be subjected to microscopic examination, and the value of
the fossil to a palaeobotanist is proportionately greater. In reviews of methods of
investigating fossil plants, it is generally suggested that impression fossils will reveal
only the outline of the plant material, and perhaps in the case of leaf impressions,
the venation pattern. This paper is an account of a method of further investigating
such impression fossils by preparing a rubber latex cast of the surface, and photo-
graphing the microtopography of the cast by scanning electron microscopy (SEM).
Latex replicas have been in use in palaeontology for some years, for preparing casts
and moulds of fossils (Rigby and Clark 1965). The application of SEM to such
replicas appears to have considerable potential for revealing epidermal features—
cell outlines, position and orientation of stomata, hairs, etc., in a class of plant
fossils which is not generally rated as susceptible to microscopic examination. The
use of SEM in palaeobotany has recently been thoroughly reviewed by Taylor (1968),
Muir (1970fl, b), and Snigirevskaya (1971). These and other authors have empha-
sized the value of SEM in studies of spores (Leffingwell and Hodgkin 1971; Reyre
1971), of wood (Alvin and Muir 1969), and of cuticles (Boulter 1971), but its applica-
tion to plant impression fossils does not seem to have been exploited.

MATERIAL AND METHOD

A typical plant ‘impression fossil’ shows on the rock surface a mould of the outer
surface of the organ, with a microtopography which is ‘negative’ with respect to
the original surface, so that stomatal cavities appear as small protrusions, and so on.
A rubber latex cast of such an impression or mould gives a replica of the original
plant surface. The quality of such a plant impression fossil appears to depend
largely on :

1. The extent to which the original plant tissue surface showed a topography
reflecting underlying epidermal or subepidermal features. In some cases, as cell
contents collapsed post-mortem, the outer surface of the cell walls may even show
more of the underlying cell arrangement than was the case in life.

2. The rapidity with which the plant material became incorporated and the accruing

[Palaeontology, Vol. 16, Part 3, 1973, pp. 645-649, pi. 79.]

646

PALAEONTOLOGY, VOLUME 16

sediment formed a mould in juxtaposition to it, before microbial activity or diagenesis
caused collapse and loss of structure of the outer surface of the plant tissue.

3. The particle size of the matrix (whether clastic, or more or less syngenetic in
character); the smaller the effective particle size, obviously the greater the fidelity
of the mould to the minutiae of the original microtopography. In the latex replicas
of plant impression fossils in fine-grained matrices that we have examined, the SEM
even at magnifications of up to 10 000 times has revealed surprisingly little detail
of the particulate nature of the matrix, and does not resolve any texture induced by
the character of the latex itself. The latex, once dried at room temperature for 24
hours, withstands both the exposure to vacuum involved in specimen coating and the
electron beam itself. Shrinkage of the latex cast appears to be insignificant ; specimens
up to 6 months old showed a linear shrinkage of less than 3%. No special study of the
possibilities of differential contraction was attempted, but clearly if size or shape dif-
ferences of this order were consequential, this aspect would need further consideration.

In a preliminary investigation, we have found that Carboniferous argillaceous
sediments may retain a high degree of epidermal detail of lycopod stems, capable
of being picked up on a latex replica. The most satisfactory results have been obtained
from specimens from which the coaly plant material has either been burnt off or
removed from the mineral matrix by weathering. Either of these processes reveals
a surface of matrix with the highest possible fidelity to the original plant surface
microtopography. A less satisfactory result is obtained where the impression has
been exposed by a fracture plane running more or less along the interface between
matrix and coaly material.

The advantages of the method of SEM examination of latex replicas are:

1. The method reveals epidermal characters on specimens which, lacking a cuticle,
would not previously have been rated capable of yielding such detail.

2. In addition, this procedure may reveal cellular character on surfaces which
have never had a cuticle (e.g. lycopod leaf abscission scars) — see PI. 79, fig. 2.

EXPLANATION OF PLATE 79

Replicas in latex of leaf cushions of Lepidodendron (fig. 6), showing epidermal detail under the scanning

electron microscope (figs. 1-5).

Figs. 1-4, 6. Lepidodendron subdichotomum Sterzel, sensu Thomas, from old tip heap, Radstock Colliery

(Nat. Grid. ref. 696 554), British Museum (Natural History) V 67053. Specimen probably from Radstock

Group (Westphalian D) or possibly from the underlying Farrington Group.

Fig. 1. Leaf cushion with scar, x28.

Fig. 2. Detail of leaf scar surface, with vascular scar, and on either side the two parichnos, x68.

Fig. 3. Detail of ligule pit, the fissure abutting obliquely on upper edge of leaf scar. (Hole to right of ligule
pit is artifact caused by an air bubble in the matrix), x 205.

Fig. 4. Stomata on lower field of the leaf cushion : note clarity with which surrounding epidermal cell walls
appear on the latex surface, and the stomatal apertures within the two stomatal depressions, X 900.

Fig. 5. Single stoma from the leaf cushion just above the leaf scar on a specimen of Lepidodendron vel-
theimii Sternberg (I.G.S., Kidston Collection No. 5115) from the Edge Coal Group, Stirling, Scotland,
x600.

Fig. 6. A photograph, with oblique illumination, of a white latex rubber cast (‘positive’) prepared from
the same specimen of Lepidodendron subdichotomum as figs. 1-4, X 10.

Figs. 1, 4, and 5 were taken on a Cambridge S600; Figs. 2 and 3, on a Cambridge Stereoscan.

PLATE 79

CHALONER and GAY, latex casts of fossil plants

648

PALAEONTOLOGY, VOLUME 16

3. The fossil is left completely intact, which is an obvious advantage in a figured
or type specimen.

4. A latex cast may be prepared from a relatively large specimen (e.g. in a museum),
which could never itself be subjected directly to scanning microscopy by any other
means. The resulting cast may readily be removed, or sent through the post, without
needing to move the original specimen.

5. An incidental advantage is the homogeneity of the latex cast, in terms of its
secondary electron emission (forming the SEM image). This contrasts with a clastic
matrix which, if of petrologically heterogeneous nature, will give a varied secondary
emission under the SEM unrelated to the microtopography.

TECHNICAL DETAILS

Casting. We have prepared casts from a number of Devonian, Carboniferous, and Permian plants. The
surface of Lepidodendrid stems shows a surprising amount of detail; this is evidently because partial
collapse of the original epidermal cells of the leaf cushion of these plants produces a surface topography
reflecting their underlying structure. General information on the preparation of rubber latex casts is
given in Rigby and Clark (1965). We have found that the product marketed as ‘Revultex’ (Bellman, Ivey,
and Carter Ltd., 385b Grand Drive, London, S.W. 20) gives satisfactory results. A first coat is applied
using the liquid diluted with an equal volume of water, and is worked into the surface of the fossil with
a fine brush to ensure contact with depressions in the surface. Subsequent layers should be applied only
after each coat has dried completely. The setting of the latex may be accelerated by placing a table lamp
immediately above the latex-covered fossil. For a large surface (greater than 10 x 10 cm) the latex may be
strengthened with gauze bandage or other textile applied to the wet surface of the third or subsequent
layer of latex. In some cases (especially with museum specimens) it was found that the first cast carried
a good deal of air-bome dust and other extraneous matter, and a second or even third cast gave a cleaner
subject for SEM study. However, it was also found that once dry, a latex cast could be washed free of much
of this debris with soap and water without loss of cellular detail.

Coating. We found that latex with white pigment (supplied by the manufacturer) gave particularly good
results; the white surface is more satisfactory for examination by light microscopy prior to SEM study.
Suitable pieces of the latex replica were cut out and mounted on stubs with ‘Durofix’ cement, painted
around the margin with ‘silver dag’, and coated using carbon-paladium rods in an arc in an Edwards
‘Speedivac’ 12E6 high vacuum coating unit. Multi-directional coating was achieved by rotating the stubs
on an improvised disc mounted eccentrically on the rotating spindle of the unit, set at 6 cm from the arcing
source. The stubs were given two coatings in this way, being rotated through about 45° individually (with
respect to the rotating disc) between each coating operation. Fifteen 1 -second bursts of arcing were found
to give an adequate coating under these conditions.

RESULTS

We have used this method successfully on a number of Palaeozoic lycopod stem
impressions {Lepidodendron, Sigillaria, Lepidodendropsis, and Lycopodiopsis). The
type of detail revealed is shown for two species of Lepidodendron in PI. 79, figs. 1-6.
The depressed stomata characteristic of the Lepidodendrids (Thomas 1966, 1970)
show as a particularly clear feature of microtopography on these specimens—
probably rather more so than would be the case with other plant groups. A specimen
of Lepidodendron subdichotomum from the Westphalian C or D of Radstock forms
the basis of figs. 1-4 and fig. 6. This specimen had evidently been burnt off, probably
by combustion of the tip heap on which it was collected. A white latex cast of part
of the stem surface showing leaf scars and cushions is shown in fig. 6, taken with

CHALONER AND GAY: MICROSCOPY OF PLANTS

649

oblique illumination. Figs. 1-3 show the leaf cushion, the leaf scar, and part of the
adjoining cushion surface under SEM at successively higher magnification: these
show detail of the cellular pattern on the abscission surface (fig. 2) in addition to
the parichnos and vascular cicatricule; the ligule pit is seen as a small fissure abutting
on the upper edge of the leaf scar (fig. 3). Stomata show clearly among the epidermal
cells on the leaf cushion surface : the orientation of their long axes is evident, and in
some cases the stomatal aperture appears in the cast (fig. 4). A single stoma from
another species, L. veltheimii is shown in fig. 5 for comparison. (This specimen, in
the Kidston Collection, No. 5115, is cited in Crookall 1964, p. 302). While this
technique has proved successful with a limited number of Palaeozoic plant fossils,
it could equally be applied to many types of animal fossils where microtopography
is likely to be preserved, such as the chitinous covering of arthropods and the cal-
careous shells and tests of other invertebrates. No doubt other means of preparing
replicas of the surface may give equally good results, but latex has the ability to
accommodate to large re-entrant features in the topography, while picking up with
equally high precision the microscopic details of the surface.

Acknowledgements. We gratefully acknowledge the receipt of a grant made to one of us (W. G. C.) by
the Royal Society towards the purchase of a scanning electron microscope, S600, and to Professor C. H.
Carlisle and Mr. N. Moore of Birkbeck College for help with microscopy on the ‘Stereoscan’ instrument
in their charge.

REFERENCES

ALVIN, K. L. and MUIR, M. D. 1969. Scanning Electron Microscopy— a new method of studying lignite.
Rev. Palaeobotan. Palynol. 9, 115-118.

BOULTER, M. c. 1971. Fine details of some fossil and recent conifer leaf cuticles. In heywood, v. h. (ed.).
Scanning Electron Microscopy, Systematic and Evolutionary Applications. Systematics /I55. Spec.
Vol. No. 4, 211-235.

CROOKALL, R. 1964. Fossil plants of the Carboniferous rocks of Great Britain. Mem. Geol. Surv. Gt. Brit.
4(3), 217-354.

LEFFiNGWELL, H. A. and HODGKIN, N. 1971. Techniques for preparing fossil palynomorphs for study with
the scanning and transmission electron microscopes. Rev. Palaeobotan. Palynol. 11, 177-199.

MUIR, M. 1970fl. A new approach to the study of fossil wood. Proc. 3rd Annual Electron Micr. Symp. 129-
136.

19706. Scanning electron microscopy in palynology. Rev. Palaeobotan. Palynol. 10, 85-97.

REYRE, Y. 1971. Interpretation Botanique des Pollens Inapertures du Mesozoique Saharien. Essai de
Classification d’apres I’Observation en Microscopic Electronique a Balayage. In heywood, v. h. (ed.).
Scanning Electron Microscopy, Systematic and Evolutionary Applications. Systematics .4^5. Spec.
Vol. No. 4, 145-154.

RIGBY, j. K. and CLARK, D. L. 1965. Casting and Moulding. In kummel, b. and raup, d. (eds.). Handbook
of Paleontological Techniques, 389-413. San Francisco and London, Freeman & Co.

SNiGiREVSKAYA, N. s. 1971. Application of the Scanning Electron Microscope in Botany. Botanicheski
Zhurnal, 56, 549-558. (In Russian.)

TAYLOR, T. N. 1968. Application of the scanning electron microscope in paleobotany. Trans. Amer. Microsc.
Soc. 87, 510-515.

THOMAS, B. A. 1966. The cuticle of the Lepidodendroid stem. New Phyt. 65, 296-303.

1970. Epidermal studies in the interpretation of Lepidodendron species. Palaeontology, 13, 145-173.

W. G. CHALONER M. M. GAY

Botany Department Department of Botany and Microbiology
Birkbeck College University College

Malet Street Gower Street

Typescript received 7 August 1972 London, W.C. 1 London, W.C. 1

k

kr.k

't.r ' ' %.

■

. :'i

V-

•f '

THE PALAEONTOLOGICAL ASSOCIATION

The Association was founded in 1957 to further the study of palaeontology. It holds meetings
and demonstrations as well as publishing Palaeontology and Special Papers in Palaeontology.
Membership is open to individuals, institutions, libraries, etc. on payment of the appropriate
annual subscription ;

Institutional membership .... £10 00 (U.S. $26.00)

Ordinary membership .... £5 00 (U.S. $13.00)

Student membership .... £3-00 (U.S. $8.00)

There is no admission fee. Institute membership is only available by direct application, not
through agents. Student members are persons receiving full-time instruction at educational
institutions recognized by the Council; on first applying for membership, they should obtain
an application form from the Membership Treasurer. All subscriptions are due each January,
and should be sent to the Membership Treasurer, Dr. E. P. F. Rose, Department of Geology,
Bedford College, Regent’s Park, London, NWl 4NS, England.

PALAEONTOLOGY

All members who join for 1973 will receive Volume 16, Parts 1-4.

All back numbers are still in print and may be ordered from B. H. Blackwell, Broad Street,
Oxford, 0X1 3BQ, England, at £5 per part (post free). A complete set, Volumes 1-15, consists
of 59 parts and costs £295.

SPECIAL PAPERS IN PALAEONTOLOGY

The subscription rate is £8 (U.S. $22.00) for Institutional Members and £4 (U.S. $11.00) for
Ordinary and Student Members. Subscriptions should be placed through the Membership
Treasurer, Dr. E. P. F. Rose, Department of Geology, Bedford College, Regent’s Park, London,
NWl 4NS, England. Members only may obtain individual Special Papers from Dr. Rose at
reduced rates. Non-members may obtain them at the stated prices from B. H. Blackwell, Broad
Street, Oxford, 0X1 3BQ, England.

COUNCIL 1973-1974

President: Professor M. R. House, The University, Kingston upon Hull, Yorkshire, HU6 7RX
Vice-Presidents: Mr. N. F, Hughes, Department of Geology, Sedgwick Museum, Cambridge, CB2 3EQ
Dr. Isles Strachan, Department of Geology, The University, Birmingham, B15 2TT
Treasurer: Dr. J. M. Hancock, Department of Geology, King’s College, London, WC2R 2LS
Membership Treasurer : Dr. E. P. F. Rose, Department of Geology, Bedford College, Regent’s Park,

London, NWl 4NS

Secretary : Dr. W. D. I. Rolfe, Hunterian Museum, The University, Glasgow, G12 8QQ
Assistant Secretary : Dr. C. T. ScRUTTON, Department of Geology, The University, Newcastle upon Tyne,

NEl 7RU

Editors

Dr. R. Goldrino, Department of Geology, The University, Reading, RG6 2AB
Dr. J. D. Hudson, Department of Geology, The University, Leicester, LEI 7RH
Dr. D. J. Gobbett, Department of Geology, Sedgwick Museum, Cambridge, CB2 3EQ
Dr. L, R. M. Cocks, Department of Palaeontology, British Museum (Natural History), London, SW7 5BD

Other members of Council

Dr. M. G. Bassett, Cardiff
Dr. D. D. Bayliss, Llandudno
Dr. E. N. K. Clarkson, Edinburgh
Prof. D. L. Dineley, Bristol
Dr. Julia A. E. B. Hubbard, London
Dr. C. P. Hughes, Cambridge*

Dr. J. K. Ingham, Glasgow

Mr. M. Mitchell, Leeds
Dr. Marjorie D. Muir, London
Dr. J. W. Murray, Bristol*

Dr. B. Owens, Leeds

Dr. P. F. Rawson, London

Prof. D. Skevington, Galway

* Co-opted as Editors

Overseas Representatives

Australia : Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane
Canada: Dr. B. S. Norford, Institute of Sedimentary and Petroleum Geology, 3303-33rd Street NW., Calgary,
Alberta

India: Professor M. R. Sahni, 98 The Mall, Lucknow (U.P.), India

New Zealand: Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 30368, Lower Hutt
West Indies and Central America: Mr. John B. Saunders, Geological Laboratory, Texaco Trinidad, Inc., Pointe-a-
Pierre, Trinidad. West Indies

Western U.S. A. : Professor J. Wyatt Durham. Department of Paleontology, University of California, Berkeley 4.
California

Eastern U.S. A.: Professor J. W. Wells, Department of Geology, Cornell University, Ithaca, New York

Palaeontology

VOLUME 16- PARTS

CONTENTS

The eyes of Asaphus raniceps Dalmati
(Trilobita)

E. N. K. CLARKSON 425

Remopleurides and other Upper Ordovician
trilobites from New South Wales

B. D. WEBBY 445

Lower Carboniferous conodont faunas from the
Eastern Mendips, England

M. BUTLER 477

The structural evolution of the bivalve shell

J. D. TAYLOR 519

The problematical Precambrian fossil Chuaria

T. D. FORD and w. J. breed 535

Visean trilobites from Holwell, Somerset

G. HAHN and R. HAHN 551

Symbiotic relationships between ectoprocts and
gastropods, and ectoprocts and hermit crabs in
the French Jurassic

T. J. PALMER and C. D. HANCOCK 563

Palynologic correlation of the Dorset ‘Wealden’

N. F. HUGHES and C. A. CROXTON 567

Isotopic ratios and Wealden environments
P. ALLEN, M. L. KEITH, F. C. TAN, and
P. DEINES 607

Buoyancy control and siphuncle function in
ammonoids

H. MUTVEi and R. a. reyment 623

An unusual agglutinating foraminifer from the
Upper Cretaceous of England

C. G. ADAMS, R. H. KNIGHT, and

R. L. HODGKINSON 637

Scanning electron microscopy of latex casts of
fossil plant impressions

W. G. CHALONER and M. M. GAY 645

Printed in Great Britain at the University Press, Oxford
by Vivian Ridler, Printer to the University

Palaeontology

VOLUME 16 ■ PART 4 NOVEMBER 1973

Published by

The Palaeontological Association • London

Price C5

THE PALAEONTOLOGICAL ASSOCIATION

The Association publishes Palaeontology and Special Papers in Palaeontology. Details of member-
ship and subscription rates may be found inside the back cover.

PALAEONTOLOGY

The journal Palaeontology is devoted to the publication of papers on all aspects of palaeontology.
Review articles are particularly welcome, and short papers can often be published rapidly. A high
standard of illustration is a feature of the journal. Four parts are published each year and are sent
free to all members of the Association.

SPECIAL PAPERS IN PALAEONTOLOGY

This is a series of substantial separate works. Members may subscribe to the Series or obtain
individual copies at reduced rates. The following Special Papers are available:

1. (for 1967): Miospores in the Coal Seams of the Carboniferous of Great Britain, by a. h. v. smith and

M. A. BUTTERWORTH. 324 pp., 72 text-figs., 27 plates. Price £8 (U.S. $22.00), post free.

2. (for 1968): Evolution of the Shell Structure of Articulate Brachiopods, by a. williams. 55 pp., 27 text-
figs., 24 plates. Price £5 (U.S. $13.00).

3. (for 1968): Upper Maestrichtian Radiolaria of California, by Helen p. foreman. 82 pp,, 8 plates.
Price £3 (U.S. $8.00).-

4. (for 1969): Lower Turonian Ammonites from Israel, by r. freund and m. raab. 83 pp., 15 text-figs.,
10 plates. Price £3 (U.S. $8.00).

5. (for 1969): Chitinozoa from the Ordovician Viola and Fernvale Limestones of the Arbuckle Moun-
tains, Oklahoma, by w. a. m. jenkins. 44 pp., 10 text-figs., 9 plates. Price £2 (U.S. $5.00).

6. (for 1969): Ammonoidea from the Mata Series (Santonian-Maastrichtian) of New Zealand, by
R. A. HENDERSON. 82 pp., 13 text-figs., 15 plates. Price £3 (U.S. $8.00).

7. (for 1970): Shell Structure of the Craniacea and other Calcareous Inarticulate Brachiopoda, by
A. WILLIAMS and A. D. WRIGHT. 51 pp., 17 text-figs., 15 plates. Price £1-50 (U.S. $4.00).

8. (for 1970); Cenomanian Ammonites from Southern England, by w. j. Kennedy. 272 pp., 5 tables,
64 plates. Price £8 (U.S. $22.00).

9. (for 1971): Fish from the Freshwater Lower Cretaceous of Victoria, Australia, with Comments on
the Palaeo-environment, by m. waldman. 130 pp., 37 text-figs., \% plates. Price £5 (U.S. $13.00).

10. (for 1971): Upper Cretaceous Ostracoda from the Carnarvon Basin, Western Australia, by r. h. bate.
148 pp., 43 text-figs., 27 plates. Price £5 (U.S. $13.00).

1 1. (for 1972): Stromatolites and the Biostratigraphy of the Australian Precambrian and Cambrian, by
m. r. WALTER. 268 pp., 55 text-figs., 34 plates. Price £10 (U.S. $26.00).

12. (for 1973); Organisms and Continents through Time. A Symposium Volume of 23 papers edited by

N. F. HUGHES. 340 pp., 132 text-figs. Price £10 (U.S. $26.00) (published with the Systematics Asso-
ciation).

SUBMISSION OF PAPERS

Typescripts on all aspects of palaeontology and stratigraphical palaeontology are invited. They
should conform in style to those already published in this journal, and should be sent to The
Secretary, P.A. Publications Committee, Department of Geology, Sedgwick Museum, Downing
Street, Cambridge, CB2 3EQ, England, who will supply detailed instructions for authors on
request (these are published in Palaeontology, 15, pp. 676-681).

© The Palaeontological Association, 1973

Cover illustration: Ancyrodelta element (Conodont), Cashaqua Shale, Upper Devonian,

New York State, x70.

A NEW SILURIAN ECHINOID GENUS FROM

SCOTLAND

by PORTER M. KIER

Abstract. A new genus and species of echinoid, Aptilechinus caledonensis, is described from the Silurian (Llandovery)
of Scotland. This species, the sixth known from the Silurian, is a flexible echinoid belonging to the family Lepido-
centridae. It is unusual in having its largest spines attached to the ambulacra.

A NEW Silurian echinoid is described from the Pentland Hills of Scotland. This
discovery is of great importance because of the rarity of echinoids of this age, and the
sweeping evolutionary changes which were occurring in echinoids during this time.
Only five Silurian echinoids are known : Echinocystites pomum Thomson, Myriastiches
gigas Sollas, and Palaeodiscus ferox Salter from Great Britain, Gotlandechinus
balticus Regnell from Sweden, and Koninckocidaris sUurica Jackson from the United
States. The addition of the sixth adds considerably to our understanding of the evolu-
tion of these primitive species.

Although the specimens are preserved as moulds and are flattened, many minute
details of their tests are visible. The most unusual morphological feature, one that
would have given it a most ‘peculiar’ appearance in life, as reconstructed on text-
fig. 4, is the presence of the largest spines on the ambulacra, and their absence on
most of the interambulacral plates. One of these spines occurs beside each porepair;
with presumably a function of protecting the tubefeet from predators. A few of the
more adoral of these spines could have been used for locomotion, but probably the
echinoid ‘walked’ mainly on the numerous peristomial spines.

Structurally, Aptilechinus possesses many of the characters which would be expected
in a Silurian echinoid according to the evolutionary trends described by Kier (1965)
for these flexible echinoids. The ambulacra are not expanded adorally, and the pore-
pairs are near the perradial suture as typical in Silurian flexible echinoids. The radial
water vessel is covered, as in many early Paleozoic echinoids, but it lacks the groove
on the exterior along the perradial suture found in most of the other Silurian echinoids.
The regularity of the shape of its plates coincides with that found in some other
Silurian echinoids such as Koninckocidaris and Myriastiches, and its small, high test
with few plates is typically primitive.

There appears to be in this genus only a single genital plate lacking genital pores.
Although the apical area is not well preserved in any of the specimens, enough of the
area is visible on several specimens to convince me that only one genital plate was
present. Thus the absence of the other four genital plates and the lack of genital pores
in the single madreporitic-genital plate is to be expected in echinoids from the
Ordovician and Silurian. The Ordovician genera Aulechinus, Ectinechinus, and
Eothuria have only one genital plate with no genital pore as does the Silurian Echino-
cystites. The apical system is not known in the Silurian Myriastiches, and the apical
system is not clear in Palaeodiscus. According to Jackson (1912, p. 286, pi. 20, fig. 5)

[Palaeontology, Vol. 16, Part 4, 1973, pp. 651-663, pis. 80-83.]

652

PALAEONTOLOGY, VOLUME 16

there is more than one genital plate in the Silurian Koninckocidaris silurica Jaekson,
but this figure shows no genital pores. However, this speeies is based on only a frag-
ment showing the interior, and until this specimen has been found and re-examined,
the presence and character of these genital plates is not certain. The earliest echinoid
which definitely has more than one genital plate is the Devonian Lepidechinoides
whitnalli in which Cooper (1931, p. 138) found a genital plate at the head of each
interambulacrum, and each genital plate had numerous genital pores. This is also
the earliest known Paleozoic echinoid that has genital pores. Therefore, it is apparent
that the development of genital pores and more than one genital plate in the Paleozoic
echinoids occurred at some time in the Silurian or Devonian.

Stratigraphical and geographical occurrence of echinoids. These specimens were collected by Mr. David
Hardie from Silurian rocks of the Pentland Hills, Midlothian, Scotland and purchased by the Royal
Scottish Museum in Edinburgh in 1897. They are labelled ‘Starfish Bed’, Gutterbom Bum, Pentland Hills.
Dr. Robin Cocks of the British Museum (Natural History) has examined the material and believes that the
locality data are correct, and states that these beds belong to the crenulata Zone, which is latest Llandovery
(Cocks, Holland, Rickards, and Strachan, 1971, fig. 9). The beds and their fauna have been described by
Lamont (1947, pp. 193-208, 289-303), who later (1952, p. 27) proposed that they be referred to a new
division of the Silurian System called ‘Pentlandian’. He also considered them to be Gala-Tarannon
(equivalent to latest Llandovery) in age.

Order echinocystitoida Jackson

Family lepidocentridae Loven
APTiLECHiNUS gen. nov.

The test is flexible, composed of regularly shaped plates with the interambulacral
plates imbricating adapically and laterally, and the ambulacral plates imbricating
adorally and under the interambulacra. The apical system has one genital plate lack-
ing genital pores and there are five ocular plates. The ambulacra are composed of
two columns of plates in each area, and do not broaden adorally. The porepairs are
situated near the perradial suture in well-developed peripodia. On the interior, each
ambulacral plate has an elevated process forming an arched covering for the radial
water vessel. The interambulacra are composed of four columns of regularly shaped
plates. One large spine is attached to each ambulacral plate and to a few of the most
adorally situated interambulacral plates. Small spines occur on the ambulacral and
interambulacral plates, and pedicellariae were probably present. The surface of the
plates are deeply pitted. The peristome is covered with many low, strongly imbricate

EXPLANATION OF PLATE 80

Ligs. 1-3. Aplilechinus caledonensis sp. nov. 1, Adoral view of latex pull of RSM 1897.32.537A showing
portion of two interambulacra and three ambulacra. Note small spines on some of the interambulacral
plates, and larger spines on ambulacral plates, x 5. A drawing of the plate arrangement of this specimen
is on text-fig. 3. 2, Side view of latex pull of RSM 1897.32.537B (holotype) showing presence of four
columns of plates in each interambulacrum with the median column overlapping laterally the others,
and the long ambulacral spines, x 5. A drawing of the plate arrangement of this specimen is on text-
fig. 2. 3, Interambulacral plates of latex pull in fig. 2 showing the pitted surface of the plates, and the
single protuberance or peg on each plate which probably served for attachment of a spine or pedicellaria,
X 15.

PLATE 80

KIER, Silurian echinoid

654

PALAEONTOLOGY, VOLUME 16

ambulacral plates and a few interambulacral plates. Jaws, braces, and longitudinally
ridged grooved teeth have been found.

Type species. Aptilechinus caledonensis sp. nov.

Comparison with other genera. This genus clearly belongs in the order Echinocystitoida
because of its strongly imbricate plates with the ambulacral plates bevelling under the
interambulacra and adorally over each other, and the interambulacral plates imbrica-
ting adapically. It is referred to the family Lepidocentridae because it has only two
columns in each ambulacrum. Aptilechinus is easily distinguished from the Silurian
genera of this family, Palaeodiscus, Myriastiches, and Koninckocidaris. Myriastiches
has many more plates, lacks larger spines on the ambulacra, and has a much less
developed covering over its radial water vessel. Palaeodiscus is easily distinguished
by the external groove in its ambulacra, lack of covering over the radial water vessel,
and wider ambulacra with lower plates, lacking larger spines. Too little is known of the
Silurian species of Koninckocidaris to compare it with Aptilechinus. The holotype and
only known specimen of K. silurica Jackson is a fragment showing only part of the
interior of the test, and it is not possible from Jackson’s figures to discern the character
of the interior of the ambulacral plate. Unfortunately, this holotype is not at the
University of Rochester, as reported by Jackson, and is presumably lost.

Aptilechinus differs from the Ordovician genera Aulechinus, Ectinechinus, and
Eothuria in having more regularly arranged interambulacral plates, its porepairs
more distant from the median suture, its pores completely divided from each other,
and well-developed spines. It differs from the Devonian Lepidechinoides in having
its radial water vessel covered, its porepairs much nearer the perradial suture, larger
ambulacral spines, and smaller interambulacral spines. The Devonian Albertechinus
and Lepidocentrus have primary tubercles on the interambulacra, and Porechinus is
easily distinguished by its ambulacral plates which have the inner pore of each
pair open.

Aptilechinus caledonensis sp. nov.

Material. The specimens are preserved as impressions in a silty shale whose grains are small enough to
preserve minute details of the test. All the specimens are flattened, with the plates somewhat shifted but not
disassociated. There are impressions of twenty-one specimens, six of which have been collected with their
counterparts. Both the interior and exterior of portions of the echinoids are visible, but none of the calcite
of the tests is preserved.

Presumably, the echinoids were covered and killed by the sediment in which they now occur. Although

EXPLANATION OF PLATE 81

Figs. 1-3. Aptilechinus caledonensis s,p. nov. 1, Adapical view of latex pull of RSM 1 897. 32. 538B showing
the small apical system. Several of the ocular plates are visible; the single large plate is probably the
madreporite, x 5. 2, A brace on specimen in fig. 1, x 10. 3, View of portion of ambulacrum of specimen
in fig. 1 showing position of porepairs near perradial suture, and greater height adradially of each plate
when not overlapped by adjacent plates. Note long striated spines with deep concavity at their bases.
These spines were evidently attached to the single nodes occurring on each ambulacral plate visible slightly
adradial and adapical to the outside pore of each porepair, x 10. 4, Side view of latex pull of RSM
1897.32.551 showing two interambulacra and three ambulacra, x5.

PLATE 81

KIER, Silurian echinoid

656

PALAEONTOLOGY, VOLUME 16

the spines have shifted slightly, they are still very near to where they were attached to the individual echinoid.
Most of the ambulacral spines extend vertically from the ambulacra, showing that the echinoids were not
disturbed after death by predators or currents.

Size. It is difficult to estimate the original size of the specimens because many are only partially preserved,
and all are flattened. A rough estimate was possible on twelve of the twenty-one specimens, with the smallest
estimated to have been 14 mm high, the largest 30 mm and the average 20 mm. These estimates are probably
accurate to within 20%.

Shape. Three specimens are flattened sideways with little oblique distortion, therefore preserving what
appears to be the original profile (PI. 80, fig. 2; PI. 82, figs. 1, 2). The test is higher than wide with a width
approximately 80% of the height. A reconstruction of the original shape of the test is given on text-fig. 4.

Apical system. In none of the specimens is the apical system well preserved, but on
three specimens portions of it are present. The system was very small, with a diameter
equal to approximately 25% of the diameter of the test. There were five ocular plates
but probably only one genital plate. Four ocular plates are preserved on specimen
RSM 1897.32.5388 (PI. 81, fig. 1). These plates are wider than high, approximately
IT mm wide, and have a smoothly curved dorsal margin, and a straight ventral
margin. The dorsal part of the plate is thin with coarse-lattice structure, whereas the
ventral part is thickened, prominently elevated, and lacks this coarse-lattice structure.
A single indentation on the ventral portion of each plate may be the ocular pore.
There is a single larger plate, 1-5 mm wide, on this specimen which is probably a
genital plate and may be the madreporite. The indentations on the surface of this
plate are smaller than the coarse-lattice structure on the interambulacral plates and
are probably madreporic pores. No genital pores are visible. No similar plates are
preserved on this specimen and probably this species, like the Silurian Echinocystites
pomum Thomson, had only a single genital plate. Many small narrow angular plates
occur in the centre of the apical system, and presumably they are the periproctal
plates.

Ambulacra. The greatest width of each ambulacrum in specimens in which the plates
are in position is 37-45% of the width of the interambulacrum. Each ambulacrum is
approximately the same width above and below the midzone until near the apical
system (PI. 83, fig. 2) and peristome where they narrow gradually. Each has two
columns of low plates (PI. 82, fig. 3; text-fig. 2) which strongly imbricate adorally
over each other, but are overlapped by the interambulacra. The height of each plate
is approximately 40% of the width when the plates in proper position; the transverse

EXPLANATION OF PLATE 82

Figs. 1-5. Aptilechinus caledonensis sp. nov. 1, 2, Side views of latex pulls of RSM 1897.32.540B and
1897.32.540A in which most of exterior of the specimen is preserved, x 5. 3, Portion of ambulacrum of
latex pull of RSM 1897.32.537A showing the well-developed peripodia, position of porepairs near
perradial suture, imbrication of plates, and node for attachment of large spine, x 15. 4, View of interior
of portion of test of latex pull of RSM 1897.32.552 showing the lack of pits on the inner surfaces of
the plates and the interior structure of the ambulacrum, X 5. A more enlarged view of this ambulacrum
is on fig. 5. 5, Interior of ambulacrum of specimen in fig. 4 showing arched covering over radial water
vessel, X 1 5. A reconstruction of this structure is on text-fig. 1 .

PLATE 82

KIER, Silurian echinoid

658

PALAEONTOLOGY, VOLUME 16

sutures are parallel. However, on specimens in which some of the ambulacral plates
are isolated by post-mortem distortion and not overlapped by the adjacent inter-
ambulacra (PI. 81, fig. 3), it can be seen that the ambulacral plates expand greatly in
height adradially. When in proper position approximately 40% of the surface area
of each ambulacral plate is covered by the overlapping interambulacral plate and
adapical ambulacral plate. Due to the nature of the preservation, it is not possible to
count the number of ambulacral plates in a full column on most of the specimens,
but on two specimens a fairly accurate estimate can be made. One specimen,
approximately 20 mm high has at least fifty plates in an ambulacrum (excluding
the peristomial ambulacral plates), and another with a height estimated at 24 mm
has sixty plates in an ambulacrum, plus another sixteen to twenty extending on
to the peristome. The peristomial ambulacral plates are lower (PI. 83, fig. 3) than
the other ambulacral plates, more imbricate, and have peripodia approximately one-
half as large. Approximately three ambulacral plates occur opposite each adjacent
interambulacral plate. The porepairs are situated near the perradial suture (PI. 82,
fig. 3) approximately one-third the distance from the perradial to the adradial suture.
The pores of a pair are oblique with the outer pore more adapical. This outer pore is

narrower than its partner on some of the plates, but
the same width on others, but this narrowing may be
due to post-mortem distortion. The porepairs occur
in well-developed peripodia (PI. 82, fig. 3) with a high
ridge separating the pores, and with a rim surrounding
the porepair. This rim is absent on the adradial side of
each outer pore but is well developed curving around
the inner pore.

Interior. The interior of each ambulacral plate has
a prominent elevated process forming an arched cover-
ing extending along the perradial suture, and no
doubt serving as a passageway for the radial water
vessel (PI. 82, fig. 5; text-fig. 1). A ridge extends trans-
versely on each plate from this passageway and forms
a high rim along the adapical edge of each porepair.
This ridge extends approximately one-half the width
of each plate. A deep depression occurs between this ridge and the adjacent porepair
with a small gap extending from the inner pore and leading into the covered passage-
way presumably for the entrance of the side branch joining the tubefoot to the radial
water vessel.

Interambulacra. Four columns of thin plates are present in each interambulacrum.
They strongly imbricate adapically and laterally and are more or less regularly shaped.
The median column (text-figs. 2, 3) is composed of narrower plates than in the other
columns with a width equal to approximately only one-half the height. These plates
are hexagonal with their greatest width adapical, and they imbricate laterally over
the columns on their right and left. This median column varies in position on the
same specimen and between different specimens. On some specimens (PI. 81, fig. 1)
all the median columns (as viewed from above) have two columns on the left of the

TEXT-FIG. 1. Aptilechinus caledoneti-
sis sp. nov. A reconstruction of the
interior of part of an ambulacrum
showing the arched covering which
extends along the perradial suture,
and served as a passageway for the
radial water vessel.

KIER: NEW SILURIAN ECHINOID

659

TEXT-FIG. 2. Aptilechinus caledonensis sp. nov. Plate arrangement of a latex pull of the holotype, RSM
1897.32.537B, showing four columns in each interambulacrum, two in each ambulacrum. The dotted
circles on the ambulacral plates and a few of the adoral interambulacral plates mark the location of nodes
where probably were attached the larger spines, x 5. A photograph of this specimen is on PI. 80, figs. 2, 3.

TEXT-FIG. 3. Aptilechinus caledonensis sp. nov. Plate arrangement of a latex pull of RSM 1897.32.537A
showing adoral plate arrangement. Note the fragmental remains of the large spines which were attached
to the ambulacral plates and a few of the adoral interambulacral plates. The dotted circles mark the loca-
tion of nodes where these spines were probably attached, x 5. A photograph of this specimen is on PI. 80,
fig. 1.

median column and one on the right, whereas in other specimens (PI. 80, fig. 1) one
interambulacrum has two columns on the right but another interambulacrum has
one on the right.

The plates of the columns which border the ambulacra are less angular in outline
and the edge of the plate which overlaps the ambulacra and adapical interambulacral
plate is curved. Adorally, the first interambulacral plate is a single plate of the median
column (PI. 80, fig. 1 ; text-fig. 3), followed by two in the second row, three in the
third, and four in the fourth. Approximately thirty-five plates are present in each
interambulacrum with approximately ten plates in each column bordering the
ambulacra. It is not clear how many interambulacral plates occur on the peristome.

Spines and tuberculation. One of the most unusual features of this echinoid is the
presence of large spines on the ambulacra and their absence on the interambulacra,
except on a few adoral plates. Two sizes of spines occur on the ambulacra. The largest
are up to 3-4 mm long, gently tapering to a sharp point which is not preserved on most
of the specimens. They are expanded near their bases where a deep concavity (PI. 81,
fig. 3) is present presumably for the insertion of muscle or ligament. These spines are
longitudinally striated with approximately fifteen striations on each spine. One of
the spines is attached to each ambulacral plate as indicated by the fact that the number
of spines found on any specimen approximates but never exceeds the number of
ambulacral plates. These spines were evidently attached to a single node that occurs
on each plate slightly adradial and adapical to the outside pore of the porepair (PI. 81,
fig. 3; PI. 82, fig. 3; text-figs. 2, 3). These nodes lack the coarse meshwork found on

660

PALAEONTOLOGY, VOLUME 16

most of the surfaces of the ambulacral plates, and are simple protuberances lacking
mamelons. The smaller spines are approximately one-third the size of the larger, are
also striated and expanded at their bases, but the presence or absence of the basal
concavity cannot be determined. The number of these smaller spines is not clear as
most of them were removed during post-mortem sorting. However, they appear to
have been attached to a single smaller node that occurs near the median suture
adapical to the inner pore of each porepair (PI. 82, fig. 3).

No large spines were attached to the interambulacra except adorally where a single
large node similar to those found on the ambulacra is present on the first three plates
of some specimens (PI. 80, fig. 1 ; text-fig. 3) indicating that a large spine was attached
there. Small spines approximately 0-9 mm long appear to have been attached to most
of the interambulacral plates. They are slender, tapering to a sharp point, striated,
and have expanded bases. I have not been able to find any nodes or tubercles for the
attachment for these spines.

Many of the interambulacral plates bear a single protuberance with sharp sides
and a flat upper surface. These pegs occur on the outer adapical corner (PI. 80, fig. 3;
text-fig. 2) of each plate except on the narrow median column where they occur on
the middle of the adapical surface. They are approximately 0-2 mm in diameter.
Similar but smaller pegs occur on the ambulacral plates (PI. 82, fig. 3) with two to
three on each plate, one between the porepairs and the adradial suture and two be-
tween the porepair and the median suture. Perhaps pedicellariae were attached to
these pegs: several structures are present which appear to have been pedicellariae
but they are too poorly preserved to be certain.

The peristomial region was covered with many small spines (text-fig. 3) which
were attached to the narrow ambulacral plates which formed a very flexible surface
extending between the end of the interambulacra and the mouth opening. These
spines were attached to five or more nodes on each ambulacral plate.

Lantern. The lantern is not well preserved on any specimen, but enough of parts of it
are visible on several specimens to show that the pyramids, braces, and teeth were
well-developed. Although not enough of a pyramid is preserved on any specimen to
be able to determine the depth of the foramen magnum, the height of the pyramid is
estimated to have been approximately 2-7 mm. The pyramid appears to be small
relative to a tooth which is at least 3 mm long and 0-7 mm wide. The tooth is grooved,
and is typical of those found in Paleozoic echinoids. It has four to five longitudinal
ridges (PI. 83, fig. 3). The brace is likewise similar to those found in other Paleozoic
echinoids, and is approximately 2-5 mm long, 0-9 mm wide, and expanded at its ends
(PI. 81, fig. 2).

EXPLANATION OF PLATE 83

Figs. 13. Aptilechinus caledonensis sp. nov. 1, Side view of latex pull of RSM 1897.32.543 showing most
of length of two ambulacra with their long spines and part of the interior surface of the other side of
the test showing the lack of pits on the interior surface of the plates, X 5. 2, Adapical region of latex
pull of RSM 1 897.32.541 B, x5. 3, Oral region of latex pull of RSM 1897.32.538A showing portions
of pyramids, part of a striated tooth, and several peristomial ambulacral plates, x 10.

PLATE 83

KIER, Silurian echinoid

EXPLANATION OF TEXT-FIG. 4. A reconstruction of Aptilechinus caledonensis sp. nov. as it may have appeared
in life.

KIER: NEW SILURIAN ECHINOID

663

Surface ornamentation. All of the exterior surface of the interambulacral plates and
most of the exterior of the ambulacral plates is deeply pitted (PI. 80, fig. 3). The
largest of these pits are 0-8 mm in diameter, and they are irregularly arranged. At first
appearance they appear to be the meshwork structure of the plates themselves but on
some specimens parts of the interior of the plates are visible, and it can be seen that
the true meshwork is much finer and is regularly structured in a latticework typical
of echinoderms (PI. 83, fig. 1). This pitted surface is absent on the ambulacral plates
on the peripodia and on the nodes. Evidently, these pits served no exterior function
for they are present on that part of a plate that is covered in life by adjacent plates
(PI. 81, fig. 3).

Type-specimens. Royal Scottish Museum, Edinburgh, Scotland. Holotype RSM 1897.32.537b, figured
paratypes RSM 1897.32.537a, 1897.32.538a, b, 1897.32.540a, b, 1897.32.541b, 1897.32.543, 1897.32.551,
1897.32.552.

Acknowledgments. 1 thank Euan Clarkson who told me of these specimens, and Charles Waterston, the
Keeper of the Royal Scottish Museum, who kindly lent them for study. Both Ivor Henrichsen and
William James Baird of the Royal Scottish Museum provided me with data on the specimens, and Robin
Cocks examined the material and gave me his opinion on their occurrence and age. Larry B. Isham, scientific
illustrator, made the excellent reconstruction of the echinoids on text-figs. 1 and 4, and Thomas F. Phelan,
museum specialist, did the photography, made the latex pulls and gave me his valuable opinions on the
morphology. J. Wyatt Durham and David L. Pawson critically read the manuscript. Robert W. Lamond
prepared and studied this material and recognized that it represented a new genus, but was unable to
complete his study because of subsequent commitments.

REFERENCES

COCKS, L. R. M., HOLLAND, c. H., RICKARDS, R. B. and STRACHAN, I. 1971. A Correlation of Silurian Rocks in
the British Isles. J. Geol. Soc. Lond. 127, 103-136.

COOPER, G. A. 1931. Lepidechinoides Olsson, a genus of Devonian echinoids. J. Paleont. 5, 127-142,
pis. 18-19.

JACKSON, R. T. 1912. Phylogeny of the Echini, with a revision of Paleozoic species. Mem. Boston Soc. nat.
Hist. 7, 443 pp., 76 pis.

KIER, p. M. 1965. Evolutionary trends in Paleozoic echinoids. J. Paleont. 39, 436-465, pis. 55-60.
LAMONT, A. 1947. Gala-Tarannon Beds in the Pentland Hills, near Edinburgh. Geol. Mag. 84, 193-208,
289-303.

1952. Ecology and correlation of the Pentlandian — a new division of the Silurian system in Scotland.

Rep. 18th Geol. Congr., London, 10, 27-32.

PORTER M. KIER

Department of Paleobiology
Smithsonian Institution
Washington, D.C. 20560
U.S.A.

Typescript received 27 October 1972

I

■ '■ ' .■ ;Cl ^ 1

■'.•■■ ' '' ' yf'r 'i?

J "■>!•■. fi ■•■ ' -f '■ !. ", , '^.'V .;rw

. '; - . ^ .'.^ ' ■ ’. ' •'. "'V •P’*' • ' ;

..-I.' ' .r u;.-". . ■• , . .: f'^'. 4.' '•'■'* .

W‘" ' '

' ; .. ■ . ■ ,,■ '" :' ,:

■ ■ ' ■ '■ ■'- •■ • ""'....a;.'.,; "!*'■'

j^, ■'". ■ .'*" " ' ■ ' ' V ;;,/ ■ i»V'!?ta!.,.6'j»»ai ’'aGfTv'S'.'

I" r ■ ■■ ■' ; '.:v ' *•’•<& •.«'_ >.^ ;■ • O'

pi™} •••>'.;••' «•”■ .'■ . ■■; i ; ‘ !.' '•'■ '■ i's •■vi.iL ■}•' 'ii; ^ ■ ■•"^ iVs" ■.,

■■' ..aa;'’* :.' ^' '. ' A 'fi.. . . ■; ; ; ' .'■" /•■ /; ..V.w

^ . -■■ •.'(■, •• ^ 'w;.j ' I

. ■'. ■ ;■ ' ' , . . ." ..'A *; ‘ /;...■ . ■'vV\, , ;4.'' 'A' '’'p.;f.iVr

• ' ’'T-''.. '.' '" ‘T"'.). , ./v-,.. ■.’■.>] ' l.'.jA-''.' ''.. .i.yT "l.Yi."' . ‘ 1‘ , " -

•^T? . . • , • V . . • . . . V.<^^ '

: • }•;. ■''•jy;. , ' • ': , ,1. ' » .

■j:;

■' '.■ ,f^' ^'''.

.*- IS ,<..’ '.'-'iir'U.v
'

.;■'* ."x „V ::■■ " ■ .,• '"■,

', , . rid

" ' v.‘

r '

»• .' ,.'., , ■„(!;». 4.

'S' A . ' ' ; >#;ii

' • ' , , . i ' A i *. i, . "^r - 4,‘ V ’ ■

VA;.‘», .■/ '1

■ '' . if*'!

DINOFLAGELLATE CYSTS AND ACRITARCHS
FROM THE BEARPAW FORMATION (UPPER
CAMPANIAN) OF SOUTHERN ALBERTA,

CANADA

by REX HARLAND

Abstract. Fifty-three species of dinoflagellate cysts and six species of acritarchs are reported from the upper
Campanian Bearpaw Formation of southern Alberta. These include Diconodinium firmum sp. nov., Lejeunia parva
sp. nov., L. ampla sp. nov., Spinidinium clavum sp. nov. and Hystrichosphaeridium dowlingii sp. nov. The Bearpaw
Formation is divided into three informal assemblage zones. A parameter called the gonyaulacacean ratio is used
as a possible guide to the salinity of the Bearpaw sea or to the proximity of the coastline. Three periods of open
marine conditions are postulated. These open marine conditions are intimately connected with the assemblage zones
erected for the formation. Comparisons are made with Campanian assemblages of other areas.

The Campanian Bearpaw Formation is unique in being firmly placed in the geological
time-scale by virtue of its ammonite faunas and its radiometric dating. Since there is
a lack of knowledge of Campanian organic-walled microplankton, the Bearpaw
was an ideal body of rock on which to study these microfossils. Such a reliably dated
assemblage should be valuable in any further work on Campanian or suspected
Campanian strata from other areas. The Bearpaw Formation is best known from
southern Alberta, so this area was chosen for the study.

TEXT-FIG. 1. Sketch map of southern Alberta
showing the approximate positions of the
sample localities and the extent of the Cypress
Hills (above 2500').

[Palaeontology, Vol. 16, Part 4, 197.3, pp. 665-706, pis. 84-88.]
B

666

PALAEONTOLOGY, VOLUME 16

STRATIGRAPHY OF THE BEARPAW FORMATION

The Bearpaw Formation was named by J. B. Hatcher and T. W. Stanton (1903)
from localities in the vicinity of the Bearpaw Mountains, north-central Montana.
The first biostratigraphic zonation was that of Russell and Landes (1940), and
Loranger and Gleddie (1953) were the first to attempt a micropalaeontological zona-
tion. Potassium-argon radiometric dates are available for the Bearpaw Formation

R-23. R22. R21.

0 4 8

TEXT-FIG. 2. Detailed map of the Lethbridge area with the
approximate outcrop pattern of the Bearpaw Eormation
and the positions of the sample localities. (Geology after
GSC Calgary Sheet, 1928.)

0 8 16

I t H

TEXT-FIG. 3. Detailed map of the Cypress Hills
area with the approximate outcrop pattern of the
Bearpaw Formation and the positions of the
sample localities. (Geology after GSC Calgary
Sheet, 1928.) The cross-hatched area indicates
the Cypress Hills Provincial Park.

HARLAND: CAMPANIAN MICROPLANKTON

667

at 75 ±4 million years from a thin bentonite 65 feet above the base of the formation at
Lethbridge (Folinsbee et al. 1960, 1961).

The Bearpaw Formation is for the most part a sub-horizontal unit except where
it has been affected by the Sweetgrass Arch uplift. A maximum thickness of 1 170 feet
has been recorded in the Cypress Hills (Lines 1947, in lift. 1963), but at Lethbridge
it is only 726 feet thick (Link and Childerhose 1931). The lithology of the formation
is predominantly one of shales with intercalated sandstones, minor bentonites,
carbonate bands, and ironstone nodule horizons. Detailed lithological descriptions

TEXT-FIG. 4. Composite stratigraphical section text-fig. 5. Composite stratigraphical section for the Bear-

for the Bearpaw Formation of the Lethbridge paw Formation of the Cypress Hills area showing sample

area showing sample distribution and coding. distribution and coding. (After Lines 1963.)

(After Link and Childerhose 1931.)

668

PALAEONTOLOGY, VOLUME 16

are given by Williams and Dyer (1930); Link and Childerhose (1931); Russell and
Landes (1940); Furnival (1950); and Caldwell (1968). The base of the formation in
southern Alberta is generally regarded as being isochronous (Lines 1947, in lift.
1963) but it becomes diachronous to the east (Caldwell 1968). The upper boundary is
markedly diachronous (Russell 1950).

The age of the formation in southern Alberta is upper Campanian, for the most
part lying within the zones of Baculites compressus sensu stricto, B. cuneatus, B.
reesidei, B.jenseni and B. eliasi (Caldwell 1968). The fauna consists mainly of molluscs
and foraminiferans, but full descriptions are given in Warren (in Fraser et al. 1935);
Dowling (1917); Williams and Dyer (1930); Russell and Landes (1940); Warren
(1931, ,1934, 1937); Douglas (1942); and Caldwell (1968).

The deposition of the Bearpaw Formation was accomplished during the last major
marine transgression in western Canadian geological history (Warren and Stelck
1958). A connection with the Arctic is indicated by the work of Martin (1960) in
addition to a link with the Gulf of Mexico (Reeside 1957).

Bearpaw Formation of southern Alberta. The Bearpaw Formation of the Lethbridge
area has been described by Link and Childerhose (1931) and is exposed in the valley
of the St. Mary River (fig. 2), especially in the cut bank sides of meanders. The
regional dip of the formation is less than 10 degrees in a westerly direction. Locally,
however, there are normal faults and open folds. The Cypress Hills, a plateau rem-
nant, is capped by the Cypress Hills Formation of Oligocene age and is ringed by
successively older strata including the Bearpaw Formation. The latter is essentially
flat-lying and no complete section of the formation is exposed here. Lines (1947, in
litt., 1963) gives the most complete description (see fig. 5). The regional dip is less than
10 degrees in an easterly direction, little faulting has occurred, but there is much
slumping and deep weathering.

Treatment. The distribution and type of sampling of the two sections is illustrated in figs. 4 and 5. The
samples are stored at the Research Council of Alberta, at Edmonton. A standard palynological preparation
technique was used but beyond the hydrofluoric acid stage all the samples were handled using the filtration
system of Neves and Dale (1963). A Leitz Laborlux microscope 595949 was used with the slide label to the
right of the observer. A reference co-ordinate for the upper left comer of the slide is given on the slide
labels, following Pierce (1959). A set of slides is in the Department of Geology, University of Alberta.
Specimens were photographed under Leitz Ortholux microscope 594209, equipped with an Orthomat
camera attachment. Adox KB 14 film was used. All holotypes and figured specimens are in the Palynological
Collections of the Research Council of Alberta at Edmonton.

SYSTEMATIC DESCRIPTIONS

The abbreviations O.D. and S.D. are used to indicate original and subsequent
designation. In the dimension sections the figure in parenthesis is the arithmetic mean
of the measured morphological parameters. The geological ranges given for the
species and genera are after Sarjeant (1967) unless otherwise stated, and the affinities
are after Wall and Dale (1968).

HARLAND: CAMPANIAN MICROPLANKTON

669

Division pyrrhophyta Pascher
Class DiNOPHYCEAE Paschcr
Order peridiniales (Schutt) Lindemann

Cyst-Family gonyaulacystaceae Sarjeant and Downie 1966
Genus apteodinium Eisenack 1958

Type species. Apteodinium granulatum Eisenack 1958; O.D.

Apteodinium sp. A

Plate 84, fig. 2

The specimens compare quite favourably with /i . maculatum Eisenack and Cookson
(1960), except in the possession of a large prominent apical horn and in the style of
‘ornamentation’; and with A. tamboviensis Vozzhennikova (1967), except in lacking
a sulcus. Its affinities are gonyaulacacean with the apteodinioid lineage.

Figured material. Loc. 11,7 (2) at 104 0-40-6. The last two figures refer to the sample
number at the locality and the slide on which the specimen is to be found.

Dimensions. Range: Length 98 0 (112-5) 127-0 breadth 98-0 (99-0) 100-0 p.. Two
specimens observed.

Cyst-Family Uncertain

Genus diconodinium Eisenack and Cookson 1960

Type species. Diconodinium multispinum (Deflandre and Cookson) Eisenack and Cookson (1960); O.D.

Wall and Dale (1968) regard the genus as having a precingular archeopyle and
affinities with the Gonyaulacaceae, whereas Davey (19696) prefers to include it with
the Cyst-Family Deflandreaceae (of peridiniacean affinities). A precingular archeopyle
is demonstrated here.

Diconodinium fir mum sp. nov.

Plate 84, figs. 8, 9, 15, text-fig. 6

Derivation of name. Latin firmum — meaning solid, with reference to the structure of
the distal extremity of the apical horn.

Diagnosis. Proximate cyst, commonly fusiform in shape, consisting of autophragm
or two wall layers very closely adpressed. Test finely granulate. Epitract extends into
an apical horn having a solid distal tip which may be oblate, acuminate, or bifurcate.
The antapical horn always acuminate, and hollow throughout. The cingulum con-
spicuous and takes the form of a slight laevorotatory helicoid. Tabulation not
usually seen. Archeopyle precingular of the P type (Evitt 1 967), and rounded polygonal
in shape. (PI. 84, figs. 5, 6.)

Description. The granules on the walls are of variable size, usually fairly fine, but
always conspicuous. The solid apical tip appears to have a definite structure (PI. 84,
fig. 15). It was often seen to be banded but its exact nature must await further study.

670

PALAEONTOLOGY, VOLUME 16

The cingulum is displaced by less than one-quarter of its width. Very little variation
was seen in this cyst species except as documented in dimensions and in the nature of
the apical tip.

TEXT-FIG. 6. Diconodinium firmum sp, nov.
Semidiagrammatic sketch of the holotype.
xc. 1000.

Figured material. Holotype: Loc. 3, 7 (2) at 98 0-32-9; Bearpaw Formation, Cam-
panian, southern Alberta. Loc. 2, 39 (1) at 108-(C30-3.

Dimensions. Holotype: Length 42 0 breadth 29 0 gi. Range: Length 36 0 (42-7)
50 0 ix; breadth 18 0 (29-3) 32 0 fx. Fifty specimens measured, out of a studied popula-
tion of sixty-four.

Remarks. This species is characterized by its shape and the solid structure at the distal
extremity of the apical horn. It appears close to Diconodinium rhombiformis
Vozzhennikova 1967 but is different in its lack of a distinct tabulation and the lack
of small gonal processes in the cingular region. Although the archeopyle was rarely
observed the conspicuous nature of plate 3" also suggests that the archeopyle is
formed by the loss of this plate. The species has gonyaulacacean affinities with
apteodinioid lineage.

EXPLANATION OF PLATE 84
All figures at a magnification of x 600 unless otherwise stated.

Fig. 1. Lejeunia ampla sp. nov., holotype, dorsal view. Fig. 2. Apteodinium sp. A, oblique dorsal view,
showing the archeopyle and surface ornamentation. Fig. 3. Lejeunia parva sp. nov., detail of specimen
with the tabulation, xc. 1800. Fig. 4. Lejeunia tricuspis (Wetzel) comb, nov., dorsal view. Fig. 5.
Spinidinium clavum sp. nov., dorsal view. Fig. 6. Spinidinium clavum sp. nov., holotype, dorsal view.
Fig. 7, Lejeunia ampla sp. nov., holotype, detail of apex with the ?apical archeopyle. xc. 1800. Fig. 8.
Diconodinium firmum sp. nov., holotype, dorsal view. Fig. 9. Diconodinium firmum sp. nov., dorsal
view showing the archeopyle. Fig. 10. Spinidinium clavum sp. nov., a specimen with two well-developed
antapical horns. Fig. 1 1. Diconodinium arcticum Manum and Cookson, dorsal view. Fig. 12. Lejeunia
parva sp. nov., lateral view showing a lack of an antapical horn. Fig. 13. Lejeunia parva sp. nov., dorsal
view of specimen showing tabulation. Fig. 14. Lejeunia parva sp. nov., holotype, ventral view. Fig. 15.
Diconodinium firmum sp. nov., holotype, detail of the apex. xc. 1800.

PLATE 84

HARLAND, Campanian microfossils

672

PALAEONTOLOGY, VOLUME 16

Diconodinium arcticum Manum and Cookson 1964

Plate 84, fig. 1 1

1964 Diconodinium arcticum Manum and Cookson; 18-19, pi. 6, figs. 1-4.

The granulation on the surface of the test varied from coarse to fine in the specimens
observed. It was also noted that they are smaller than those described by Manum and
Cookson (1964). They cannot be compared to D. glabrum Eisenack and Cookson
(1960) because they lack a clearly defined sulcus and differ in the nature of the
‘ornamentation’. D. arcticum has an Upper Cretaceous range. Its affinities are
gonyaulacacean with the apteodinioid lineage.

Figured material. Loc. 3, 1 (2) at 103-6-37T.

Dimensions. Range: Length 300 (41T) 57 0 breadth 200 (26-3) 37 0 fx. Fifty
specimens measured, out of a studied population of seventy-three.

Genus lejeunia Gerlach 1961
Type species. Lejeunia hyalina Gerlach 1961; O.D.

Lejeunia parva sp. nov.

Plate 84, figs. 3, 12-14, text-fig. 7

Derivation of name. Latin parva — meaning small, with reference to the size of the
cyst.

Diagnosis. Proximate cyst, elongate to rhomboidal, made up of autophragm. Test
granulate. Epitract elongated into an apical horn which is distally oblate or indented.

TEXT-FIG. 7. Lejeunia parva sp. nov. Semidiagrammatic
sketch of the holotype. xc. 1200.

Hypotract carries two antapical horns, one of which larger than the other; both
distally acuminate. Cingulum, delimited by raised sutures, is planar to a slightly
laevo rotatory helicoid. Tabulation ?4', la, ?7", 4-?"', 2'"' . Archeopyle not observed.

Figured material. Holotype: Loc. 3, 3 (2) at 99 0-3 1-9; Bearpaw Formation, Cam-
panian, southern Alberta. Loc. 3, 3 (1) at 97 0-49-2. Loc. 3, 1 (1) at 103 0-36-7.

HARLAND: CAMPANIAN MICROPLANKTON

673

Dimensions. Holotype: Length 45 0 /x; breadth 30 0 jx. Range: Length 34 0 (44-4)
57 0 ix; breadth 25 0 (35-5) 41 0 /n. Twenty specimens measured, the number of
specimens studied.

Description. Apical horn has a thickened structure which may or may not be solid,
but invariably shows a suture apparently bisecting the horn. The antapical horns
may be well developed or one may appear as a swelling. The tabulation variously
developed and where present is delimited by raised sutures. One specimen seen with
a complete tabulation. The cingulum, usually conspicuous and approximately three
microns wide, may contain granular elements that appear aligned parallel to the
longitudinal axis of the cyst. Plate 3" of the tabulation large and conspicuous. Range
of variation within this species is not great. The specimen with the tabulation not
typical for this species and so not chosen as the holotype.

Remarks. This species is similar to that of L. tenella Morgenroth 1966. His specimens,
however, lack all trace of tabulation and are larger and more rhomboidal in shape.
It is also similar to Palaeoperidinium cretaceum Pocock but differs in being smaller
and in not having an endoblast. It has possible peridiniacean affinities.

Lejeunia tricuspis (Wetzel) comb. nov.

Plate 84, fig. 4

1933 Peridinium tricuspis Wetzel; 166, pi. 2, fig. 14.

The Bearpaw specimens compare well with those of Wetzel (1933). The author
considers that Lejeunia kozlowskii Gorka 1963 is a junior synonym. Gorka (1963)
figures both L. kozlowskii and L. cf. tricuspis. It appears that any difference between
the two can be accommodated by specific variation. L. tricuspis has a geological
range of Santonian-Maestrichtian, and its affinities may be peridiniacean.

Figured material. Loc. 13, 13 (1) at 99 0-29-9.

Dimensions. Range: Length 80 0 (104-3) 135-0 jj,; breadth 52-0 (73-8) 95-0 p.. Sixteen
specimens measured, the number of specimens studied.

Lejeunia amp la sp. nov.

Plate 84, figs. 1, 7, text-fig. 8

Derivation of name. Latin am/t/a— meaning large, with reference to the overall size
of this species.

Diagnosis. Proximate cyst, rhomboidal in shape, probably made up of autophragm
only. Test granulate, scabrate and/or reticulate. No tabulation visible. Cingulum
usually conspicuous, planar, approximately 5 microns wide, delimited by raised
sutures. Single apical horn, distally oblate ; two antapical horns one of which generally
larger than the other. Archeopyle apical.

Figured material. Holotype: Loc. 3, 13 (1) at 95-8-37-0; Bearpaw Formation, Cam-
panian, southern Alberta.

674

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 8. Lejeimia ampla sp. nov. Semidiagrammatic
sketch of the holotype. X c. 300.

Dimensions. Holotype: Length 149 0 ju; breadth 108 0 p.. Range: Length 70 0 (116-3)
154-0 p\ breadth 68-0 (97-0) 108-0 p. Fifty specimens measured out of a studied
population of eighty-nine.

Description. It is possible that a second wall layer, closely adpressed, is present. Only
one or two specimens show an apical archeopyle. Norris (pers. comm.) has observed
transapical archeopyles in specimens with the same gross morphology. The Bearpaw
specimens appear to have had one or two antapical plates making up the operculum.
The antapical horns distally evexate and one was seen to carry small spinules.

Remarks. This species differs from L. tricuspis in lacking the vertical striations, the
acuminate antapical horns, and the coarse ‘ornamentation’. Doubt is expressed in
the generic assignment because of uncertainties in the nature of the archeopyle in
this species and in the genus. The archeopyle as seen in the holotype is perfectly clear
and does not appear to be of accidental origin. Evitt {pers. comm.) has commented on
the resemblance of the small folds, often observed on the test of these cysts, to growth
lines such as those exhibited by Palaeoperidinium pyrophorum (Ehrenberg). Its
affinities are unknown.

Genus spinidinium Cookson and Eisenack 1962/)

Type species. Spinidinium styloniferum Cookson and Eisenack 1962ft; O.D.

Spinidinium is characterized by the possession of ‘spines’. It is, however, morpho-
logically similar to Deflandrea in that the genus is cavate and possesses an intercalary
archeopyle. Wilson (1967) has placed certain spiny specimens in the genus Deflandrea
and certainly a review of the situation is indicated as Deflandrea is presently defined
to include only forms with smooth or granulate tests.

Spinidinium clavum sp. nov.

Plate 84, figs. 5, 6, 10, text-fig. 9

Derivation of name. Latin clavum — meaning spike, with reference to the development
of short, acuminate processes along the sutural crests.

Diagnosis. Cavate cyst, fusiform in shape, made up of two wall layers closely adpressed
except at the apex and antapex where pericoels may be evident. Test usually smooth
with the presence of occasional discrete granules. Epitract slightly more conical than

HARLAND: CAMPANIAN M I C ROPL AN KTON

675

the hypotract. Prominent apical horn, tapering with a bifid tip; antapical horns
acuminate. Sutural ridges, up to 5 ju. tall, carry short oblate and acuminate processes.
Certain plate areas of the tabulation may be delimited due to sutural development.
A tabulation 74', la, 71", ?4c, 5-6'", 72"" indicated. Cingulum planar, sulcus con-
spicuous extending on to both the epitract and hypotract. Archeopyle indeterminate,
but it is almost certain that loss of the conspicuous intercalary plate forms the
archeopyle.

Description. Cavate cyst often appears proximate because of
poor pericoel development. The apex appears to be made up
of four apical plates which are separated by large sutural
ridges, characteristic of this species. These ridges add to the
prominence of the apex. The crests of the ridges carry small
oblate and acuminate processes. The precingular plate series
appears to consist of five to seven plates, of which plate 4"
is conspicuous and polygonal in shape; directly above this
plate there is a single rectangular anterior intercalary plate.

The cingulum is often the focus for folding and crumpling of
the cyst. The post-cingular plate series appears to comprise
five or six plates but only certain indeterminate plate boundaries
were seen. The hypotract is more rounded than the epitract.

The sutural crests are better developed on the epitract than on
the hypotract. The range of variation of this species is seen
in Plate 84. The most variable feature is the size of the endo-
blast in relation to periblast.

Figured material. Holotype: Loc. 11, 1 (1) at 95 0-40T ; Bearpaw Formation, Cam-
panian, southern Alberta. Loc. 5, 1 (2) at 94-7-38 0. Loc. 5, 1 (1) at 104 0-32 0.

Dimensions. Holotype: Length 5 10 /x; breadth 29 0 fx. Range: Length 40 0 (45-3)
60 0 fx; breadth 20 0 (26 0) 35 0 fx. Seventeen specimens observed.

Remarks. This species is characterized by the nature of the large sutural ridges. It is
similar to Palaeoperidinium caulleryi Deflandre 1934 which Deflandre (1966) con-
siders to be a member of the genus Diconodinium. It was not, however, formally com-
bined (re Article 33 of I.C.B.N.). It has peridiniacean affinities with the deffandreoid
lineage.

Cyst-Family microdiniaceae Eisenack emend. Sarjeant and Downie 1966
Genus microdinium Cookson and Eisenack emend. Sarjeant 1966

Type species. Microdinium ornatum Cookson and Eisenack 1960; O.D.

Microdinium cf. irregular e Clarke and Verdier 1967

Plate 85, figs. 15, 16

1967 1 Microdinium irregulare Clarke and Verdier; 65-66, pi. 7, figs. 5-8, text-fig. 27.

Description. Proximate cyst, spheroidal to ovoidal in shape, composed of periphragm
and endophragm. The periphragm makes up the large and conspicuous sutural crests.

TEXT-FIG. 9. Spinidinium
clavum sp. nov. Semi-
diagrammatic sketch of
the holotype. xc. 1000.

676

PALAEONTOLOGY, VOLUME 16

Cyst microgranulate to granulate except for the crests which are smooth. Tabulation
present. Cingulum 3-4 microns wide, planar or weakly helicoidal in nature. Epitract
small in comparison to the hypotract, the tabulation difficult to decipher because
of the granules and crests but plates 6"', Ip, V"' were observed. Epitractal details are
especially difficult to see. The archeopyle apical of A type, and probably formed by
the loss of three or four apical plates (Evitt, pers. comm.).

Figured material. Loc. 11, 1 (2) at 100 0-43-3. Loc. 10, 4 (2) at 94 0-32-6.

Dimensions. Range: Length 24 0 (34-5) 39 0 ju,; breadth 26 0 (32-6) 38-0 w Thirty-six
specimens observed.

Remarks. These specimens compare well with those described by Clarke and Verdier
(1967). This species has a geological range Cenomanian-Santonian. It has gonyaula-
cacean affinities with the lithodinioid lineage.

Cyst-Eamily Uncertain

Genus dinogymnium Evitt, Clarke and Verdier 1967

Type species. Dinogymnium acuminatum Evitt, Clarke and Verdier 1967; O.D.

Dinogymnium cf. albertii Clarke and Verdier 1967

Plate 85, fig. 18

1967 IDinogymnium albertii Clarke and Verdier; 33, pi. 17, figs. 3, 4, text-fig. 13.

Description. Proximate cyst, subspheroidal to ovoidal in shape, the epitract more
conical than the hypotract, made up of two wall layers closely adpressed. Test carries
a number of longitudinal grooves, that are commoner on the epitract than on the
hypotract. In addition, the test perforated by many punctae (wall canals). Tabulation
not present except for a very conspicuous deep cingulum which is 4 microns wide.

EXPLANATION OF PLATE 85

All figures at a magnification of x 600 unless otherwise stated.

Fig. 1. Hystrichosphaeridium dowlingii sp. nov., ventral view, holotype, showing the archeopyle and the
nature of the processes. Fig. 2. Dinogymnium longicornis (Vozzhennikova) comb, nov., detail of the
apex. xc. 1800. Fig. 3. Dinogymnium longicornis (Vozzhennikova) comb, nov., ventral view of a
specimen with a low cingulum index. Fig. 4. Dinogymnium longicornis (Vozzhennikova) comb, nov.,
lateral view of a specimen with a high cingulum index. Fig. 5. Canningia senonica Clarke and Verdier,
dorsal view. Fig. 6. Hystrichosphaeridium cf arborispinum Davey and Williams, lateral view. Fig. 7.
VJvatodinium cf nasutum Vozzhennikova, dorsal view. Fig. 8. Canningia senonica Clarke and Verdier,
lateral view. Fig. 9. Hystrichosphaeridium salpingophorum (Deflandre) emend. Davey and Williams,
lateral view. Fig. 10. Hystrichosphaeridium tubiferum var. brevispinum Davey and Williams, lateral view.
Fig. 11. Hystrichosphaeridium cf arborispinum Davey and Williams, ventral view. Fig. 12. Hystricho-
sphaeridium salpingophorum (Deflandre) emend. Davey and Williams, lateral view. Fig. 13. ICoronifera
oceanica Cookson and Eisenack, lateral view. Fig. 14. 'IMembranosphaera cf maastrichtica Samoylovich
ex. Norris and Sarjeant emend. Drugg, lateral view. Fig. 15. Microdinium cf irregulare Clarke and
Verdier, lateral view. Fig. 16. Microdinium cf irregulare Clarke and Verdier, lateral view showing well-
developed smooth sutural crests. Fig. 17. sp. A, lateral view. Fig. 18. Dinogymnium

cf albertii Clarke and Verdier, lateral view.

PLATE 85

HARLAND, Campanian microfossils

678

PALAEONTOLOGY, VOLUME 16

and takes the form of a laevorotatory helicoid. The test is often crumpled along this
feature. Displacement approximately equal to half the width of the cingulum. Sulcus
also present but only on the hypotract. Archeopyle apical, regarded as miscellaneous
by Evitt (1967), formed by loss of a ? single plate at the very tip of the epitract.

Figured specimen. Loc. 9, 37 (1) at 105-7-42 0.

Dimensions. Range: Length 40 0 (41-3) 42 0 jx; breadth 20 0 (22 0) 25 0 fx, cingulum
index 50 0 (52-3) 55 0. Three specimens observed.

Remarks. This species is compared to D. albertii Clarke and Verdier 1967 by virtue of
the presence of punctae and in its general form, but it is smaller in size. It was pre-
viously recorded from the Santonian by Clarke and Verdier 1967. In general morpho-
logy the genus closely resembles the modern genus Gymnodinium, with which it has
often been confused. The affinities of this cyst are as yet unknown.

Dinogymnium longicornis (Vozzhennikova) nov. comb.

Plate 85, figs. 2-4

1967 Gymnodinium longicornis Vozzhennikova; 46, pi. 1, fig. 8, pi. 3, fig. 6, pi. 4, figs. 6a, 6b, 1.

Description. Proximate cyst, ovoidal to markedly elongate, made up of two closely
adpressed wall layers. Test carries some longitudinal grooves and a fine micro-
punctation which is better developed in some specimens than others. The epitract
conical to very elongate, drawn out into a long apical horn. The hypotract always
hemispheroidal. A conspicuous cingulum always present, approximately 2 microns
wide, in the form of a laevorotatory helicoid which is displaced by up to twice its
width. A sulcus is present on the hypotract only. Faint sutural ridges are sometimes
present, delimiting a possible tabulation; a possible reflected plate Ip was observed
in one specimen. Archeopyle apical, typical for the genus.

Figured material. Loc. 9, 20 (3) at 102 0-29-3. Loc. 9, 24 (3) at 107-9-36-3.

Dimensions. Range: Length 39 0 (49-4) 59 0 ju,; breadth 19 0 (27-3) 33 0 fx, cingulum
index 58 0 (65-75) 75-0. Eleven specimens studied.

Remarks. This species is very similar to that figured by Vozzhennikova (1967); there
can be no doubt that the species belongs to the genus Dinogymnium as in Vozzhenni-
kova’s figures the archeopyle is perfectly evident. The specimens from the Bearpaw
Formation seem more variable than those of Vozzhennikova (1967) and are generally
smaller. The Russian specimens are Senonian in age and were recovered from western
Siberia. Its affinities are unknown.

Cyst-Family fromeaceae Sarjeant and Downie 1966
Genus membranosphaera Samoylovich ex. Norris and Sarjeant emend.

Drugg 1967

Type species. Membranosphaera maasirichtica Samoylovich ex. Norris and Sarjeant, 1965; S.D.

HARLAND: CAMPANIAN MICROPLANKTON

679

1 Membranosphaera cf. maastrichtica Samoylovich ex. Norris and Sarjeant emend.

Dmgg 1967

Plate 85, fig. 14

1961 1 Membranosphaera maastrichtica Samoylovich (in Samoylovich et al. 252, pi. 83, figs. 1, 2.

1965 ‘I Membranosphaera maastrichtica Samoylovich ex. Norris and Sarjeant; 40.

1967 1 Membranosphaera maastrichtica Samoylovich ex. Norris and Sarjeant emend. Drugg;
29-30, pi. 5, figs. 12, 13.

Description. Proximate cyst, spheroidal to ovoidal in shape, composed of endophragm
and periphragm. The endophragm makes up small cylindrical capitate processes that
appear to support an outer membraneous periphragm. A faint trace of the cingulum
observed; it appears planar, approximately three microns in width and is delimited
by areas devoid of endophragmal processes. No other tabulation discernible. The
archeopyle apical, possibly of a type formed by loss of a single apical plate; the sulcal
notch was observed.

Figured material. Loc. 10, 5 (3) at 103 0-34-4.

Dimensions. Range: Length 28 0 (30 0) 32 0 pt; breadth 27 0 (27-5) 28 0 pt. Two
specimens observed.

Remarks. The Bearpaw specimens differ from those of Drugg (1967) in possessing
a faint cingulum and a smaller apical archeopyle. This species had a previous geo-
logical range of Maestrichtian-Danian (Drugg 1967). Its affinities are unknown.

Cyst-Family canningiaceae Sarjeant and Downie 1966
Genus canningia Cookson and Eisenack 1960

Type species. Canningia reticulata Cookson and Eisenack 1960; O.D.

Canningia senonica Clarke and Verdier 1967

Plate 85, figs. 5, 8

1967 Canningia senonica Clarke and Verdier; 20, 21, pi. 1, figs. 12-14, text-fig. 7.

It was noticed in the specimens attributed to this species that there was consider-
able variation with regard to process development and the apparent reticulation.
This species was previously reported from the Senonian of the Isle of Wight, England,
by Clarke and Verdier (1967). It has gonyaulacacean affinities with the lithodinioid
lineage.

Figured material. Loc. 11, 7 (3) at 103-0-30 0. Loc. 9, 39 (1) at 109 0-28-2.

Dimensions. Range: Length 36 0 (48 0) 67 0 /a.; breadth 40 0 (51-5) 63 0 ju, the pro-
cesses range in length from 2 0-8 0 pi. Four specimens observed.

Cyst-Family pyxidiellaceae Sarjeant and Downie 1966
Genus uvatodinium Vozzhennikova 1963

Type species. Uvatodinium nasutum Vozzhennikova 1963; O.D.

680

PALAEONTOLOGY, VOLUME 16

lUvatodinium cf. nasutum Vozzhennikova 1963

Plate 85, fig. 7

1963 lUvatodinium nasutum Nozzh&nmV.ova.', 182, figs. 13a, 13Z?.

Description. Proximate cyst, subspheroidal in shape, composed of periphragm and
endophragm closely adpressed; the former makes up the ‘ornamentation’. Test
granulate. Epitract bears an apical horn, which is distally evexate, and a distinct
‘shoulder’. The hypotract carries a slight antapical boss. The cingulum conspicuous,
delimited by raised sutures, takes the form of a laevorotatory helicoid and displaced
by half the width of the cingulum which is four to five microns wide. A sulcus present,
delimited by raised sutures, and confined to the hypotract. No other tabulation
present. Archeopyle intercalary of the I type.

Figured material. Loc. 11, 7 (1) at 99 0-29-6.

Dimensions. Range: Length 61 0 (68 0) 75 0 ij.; breadth 540 (57-6) 60 0 jx. Three
specimens observed.

Remarks. These specimens compare quite well with those illustrated by Vozzhenni-
kova (1967) but lack the coarse reticulation; although in pi. 8, fig. 4 of Vozzhennikova
(1967) the reticulation is not at all obvious. This species has previously been recorded
from the Palaeocene (Vozzhennikova 1967). Its affinities are unknown.

Cyst-Family hystrichosphaeridiaceae Evitt emend. Sarjeant and Downie 1966
Genus hystrichosphaeridium Deflandre emend. Davey and Williams 19666

Type species. Hystrichosphaeridium tubiferum (Ehrenberg) Deflandre 1937; O.D.

Hystrichosphaeridium cf. arborispinum Davey and Williams 19666

Plate 85, figs. 6, 1 1

19666 '] Hystrichosphaeridium arborispinum Davey and Williams; 61, pi. 9, figs. 5, 10.

Description. Chorale cyst, subspheroidal to ovoidal in shape, made up of both
periphragm and endophragm, the former making up the processes. Cyst smooth to
microgranulate. The processes intratabular and reflect the tabulation 6", 6c, 5'",
they are hollow, slender to latispinous, erect, cylindrical, and distally flared.
The distal extremities recurved, digitate to serrate. In many specimens the processes
appear fibrous. Additional cylindrical recurved and digitate sutural processes present
in many specimens. The archeopyle apical of the A type.

Figured material. Loc. 11, 1 (3) at 93 0-30-5. Loc. 9, 39 (2) at 107 0-33-8.

Dimensions. Range: Long axis of cyst 29 0 (40 0) 52 0 /u; short axis of cyst 30 0
(38-35) 55-0 /X, processes range in length from 1 1 to 25 p.. Twenty specimens observed.

Remarks. This species is fairly common in the Bearpaw assemblages and may be
recognized by the character of its processes. It had a geological range Lower
Barremian -Middle Barremian. Its affinities are gonyaulacacean in the hyslricho-
sphaeridioid lineage.

HARLAND: CAMPANIAN MICROPLANKTON

681

Hystrichosphaeridium dowlingii sp. nov.

Plate 85, fig. I, text-fig. 10

Derivation of name. Named in honour of D. B. Dowling, one of the first geologists
to work in southern Alberta.

Diagnosis. Chorate cyst, spheroidal in shape, made up of periphragm and endo-
phragm. Test microgranulate. Tabulation 6", 6c, 6"', l-2p, I"". Processes hollow
slender to latispinous, erect to curved, cylindrical, distally flared and fenestrate.
Archeopyle of the A type.

TEXT-FIG. 10. Hystrichosphaeridiimi dowlingii nov.

Semidiagrammatic sketch of the holotype. x c. 1300.

Figured material. Holotype: Loc. 9, 39 (2), at 106 0-33-9; Bearpaw Formation,
Campanian, southern Alberta.

Dimensions. Holotype: Long axis of cyst 44 0 \ short axis of cyst 38 0 ju., processes
range in length from 15 to 24 p,. Range: Long axis of cyst 25 0 (34-4) 46 0 p; short
axis of cyst 29 0 (34 0) 54 0 p, processes range in length from 10 to 28 p. Twelve
specimens measured, the number of specimens studied.

Description. The periphragm makes up the processes which do not appear to connect
to the interior of the cyst. Test ‘ornament’ does not extend on to the process shafts.
The sulcal area devoid of all processes except for a group of two sulcal processes.
A sulcal notch may also be seen in some specimens.

Remarks. This species is easily distinguishable from H. cf. arborispinum by the nature
of the processes, i.e. fenestrate nature and lack of ‘ornamentation’ on the processes
shafts. The test ‘ornamentation’ is also much coarser than that of H. cf. arborispinum.
Its affinities are gonyaulacacean with the hystrichosphaeridioid lineage.

c

682

PALAEONTOLOGY, VOLUME 16

Hystrichosphaeridium tubiferum var. brevispinum Davey and Williams 1966^

Plate 85, fig. 10

\966b Hystrichosphaeridium tubiferum var. brevispinum Davey and Williams; 58, pi. 10, fig. 10.

This variety compares well with those specimens described by Davey and Williams
(1966Z?). The Bearpaw specimens were, however, smaller in size with a slightly more
variable process habit. The variety is characterized by process length, and had a range
of Eocene (Davey and Williams 1966/)). Its affinities are gonyaulacacean with the
hystrichosphaeridioid lineage.

Figured material. Loc. 9, 39 (1) at 10T0-3T3.

Dimensions. Range: Long axis of cyst 25-0 (31-2) 38 0 /u,; short axis of cyst 25-0 (30-6)
38 0 /X, processes range in length 6 to \5 p.. Twenty specimens observed.

Hystrichosphaeridium salpingophorum (Deflandre) Davey and Williams 1966b

Plate 85, figs. 9, 12

1935 Hystrichosphaera salpingophora Deflandre; 232, pi. 9, fig. 1.

\966b Hystrichosphaeridium salpingophorum (Deflandre) emend. Davey and Williams; 61, 62,
pi. 10, fig. 6.

This species compares well with those specimens figured by Davey and Williams
(1966/)). There is an obvious morphological overlap with H. tubiferum var. brevi-
spinum but usually the two species can be recognized. The species has a geological
range of Upper Jurassic-Lower Eocene. Its affinities are gonyaulacacean with the
hystrichosphaeridioid lineage.

Figured material. Loc. 9, 39 (2) at 99-0-36 0. Loc. 13, 5 (3) at 92-0-40-6.

Dimensions. Range; Length of long axis 31 0 (38-8) 44 0 p; length of short axis 29-0
(35-75) 48 0 p, processes range in length from 10 to 18 p. Nine specimens observed.

Genus cleistosphaeridium Davey, Downie, Sarjeant and Williams 1966

Type species. Cleistosphaeridium diversispinosum Davey, Downie, Sarjeant and Williams 1966; O.D.

EXPLANATION OF PLATE 86

All figures at a magnification of x 600 unless otherwise stated.

Fig. 1. Cleistosphaeridium diversispinosum Davey et al., lateral view. Fig. 2. Forma A, lateral view, show-
ing nature of the archeopyle. Fig. 3. Polysphaeridium subtile Davey and Williams, orientation unknown.
Fig. 4. Exochosphaeridium pseudohystrichodinium (Deflandre) emend. Davey, lateral view. Fig. 5. Forma
A, lateral view, showing globular central body. Fig. 6. Polysphaeridium subtile Davey and Williams,
?lateral view showing apical boss. Fig. 7. Oligosphaeridium anthophorum (Cookson and Eisenack)
Davey, lateral view. Fig. 8. Oligosphaeridium anthophorum (Cookson and Eisenack) Davey, ventral
view. Fig. 9. Exochosphaeridium sp. A, lateral view. Fig. 10. Exochosphaeridium cf phragmites Davey
et al., dorsal view. Fig. 11. Oligosphaeridium pulcherrimum (Deflandre and Cookson) Davey and
Williams, lateral view. Fig. 12. Oligosphaeridium pulcherrimum (Deflandre and Cookson) Davey and
Williams, lateral view. Fig. 13. Tanyosphaeridium variecalamwn Davey and Williams, lateral view of
specimen with many processes per plate area. Fig. 14. Tanyosphaeridium variecalamwn Davey and
Williams, lateral view of specimen with few processes per plate area.

PLATE 86

HARLAND, Campanian microfossils

684

PALAEONTOLOGY, VOLUME 16

Cleistosphaeridium diver sispinosum Davey, Downie, Sarjeant, and Williams 1966

Plate 86, fig. 1

1966 Cleistosphaeridium diversispinosum Davey et al. ', 167, pi. 10, fig. 7.

Considerably more specific variation than that observed by Davey et al. (1966)
is present within this species. The processes are variable in thickness, sinuosity,
length, and in the structural diversity of their extremities. This species was previously
recorded from the Eocene. Its affinities are unknown.

Figured material. Loc. 10, 5 (1) at 95 0-43-4.

Dimensions. Range: Length 35-0 (48-75) 57-0 ju.; breadth 43-0 (52-25) 67-0 /x, processes
range in length from 5 to \9 p. Nine specimens observed.

Cleistosphaeridium sp. A

Plate 85, fig. 17

Description. Chorate cyst, subspheroidal in shape, made up of two wall layers; the
periphragm makes up the processes. The test carries granules but no trace of tabula-
tion could be discerned, as the processes appear randomly dispersed. The processes
slender to latispinous, curved to sinuous, cylindrical to tapering. Most processes end
in an equal distal bifurcation but others are acuminate and oblate. Archeopyle apical,
of the A type.

Figured material. Loc. 3, 3 (1) at 98-0-46-5.

Dimensions. Range: Length 35-0 (39-0) 43-0 p.; breadth 39-0 (40-5) 42-0 p, processes
range in length from S to \ \ p. Two specimens observed.

Remarks. This species is similar to 1 Cleistosphaeridium flexuosum Davey et al. 1966
but differs in process length and in the nature of the process extremities. Its affinities
are unknown.

Genus coronifera Cookson and Eisenack emend. Davey 1969u

Type species. Coronifera oceanica Cookson and Eisenack 1958; O.D.

ICoronifera oceanica Cookson and Eisenack f958
Plate 85, fig. 13

1958 ICoronifera oceanica Cookson and Eisenack; 45, pi. 12, figs. 5, 6.

Description. Proximate cyst, spheroidal in shape. Test composed of two wall layers
closely adpressed of which the periphragm alone makes up the processes. Cyst wall
microgranulate and covered by numerous processes, seemingly at random. The
processes do not connect to the interior of the cyst and are slender, curved, cylindrical
to tapering, and distally acuminate. No tabulation could be discerned. The cyst
carries a single large antapical process which is latispinous, erect, distally open with

HARLAND: CAMPANIAN MICROPLANKTON

685

an entire or denticulate margin. This process is very characteristic of the species and
genus. No archeopyle observed.

Figured material. Loc. 1, 28 (1) at 106-7-46 0.

Dimensions. Range: Length 33 0 (37-5) 42 0 /j,; breadth 3 1 0 (40-5) 50-0 ju, the pro-
cesses range in length from 4-10 /x. Two specimens observed.

Remarks. The specimens under consideration differ from those of Davey (1969fl) in
lacking any kind of sutural ridges or clear apical processes. This is probably a conse-
quence of specific variation. Lack of the presence of an archeopyle casts some doubt
on the identification of this species. The presence of an A archeopyle would suggest
that these specimens be placed in the genus Diphyes Cookson 1965. C. oceanica had
a geological range Albian-Cenomanian. It probably has gonyaulacacean affinities
by virtue of the single antapical process, suggesting the possession of a single antapical
plate. Millioud (1969) reported the presence of a precingular archeopyle in a new
species of Coronifera from the Upper Hauterivian of Angles, SE. France, indicating
that Coronifera should be attributed to the hystrichodinioid lineage.

Forma A

Plate 86, figs. 2, 5, text-fig. 1 1

Deseription. Chorate cyst, ovoidal to elongate in shape. Test consists of a thick (LO-
TS microns) endophragm and a thin periphragm. Test smooth to microgranulate.
Endoblast appears made up of discrete chambers giving the cyst a globate appearance.
Two whorls of lobes present separated by a cingular groove in the form of a laevo-
rotatory helicoid. Processes intratabular, reflecting a possible tabulation of ?7" and
5 or 6"'. They do not connect to the interior of the cyst and are constructed of

periphragm. No antapical processes present.
Archeopyle apical with an attached operculum,
possibly of the Aa type.

Figured material. Loc. 1, 36 (3) at 107-2-35-9.

Dimensions. Range: Length 32 0 (37-5) 43 0 [x;
breadth 25-0 (29-5) 34 0 p, processes range in
length from 20 to 25 p. Two specimens observed.

Remarks. These cysts are peculiar in the struc-
ture of the central body which appears to be
made up of discrete chambers. Two whorls are
present, one at either side of the cingulum.
A single chamber is centrally placed on the
ventral surface of the cyst and it may be equi-
valent to reflected plate 1" . The laevorotatory
helicoidal cingulum divides the two whorls. It is
of interest to note that no antapical process is
present.

TEXT-FIG. 1 1. Forma A. Semidiagrammatic
sketch, xc. 1200.

686

PALAEONTOLOGY, VOLUME 16

There is, however, a possible accessory sulcul process. Also of interest is the opening
in reflected plate T" . It appears that the archeopyle has been formed by partial loss
of the apical plate series with a single opercular piece remaining. It was difficult to
observe the exact relationship of this operculum ? to the archeopyle. Evitt {^pers.
comm.) considers that the lobate nature of this cyst is due to deformation as the result
of the growth of pyrite sphaerules. Its affinities are unknown.

Genus oligosphaeridium Davey and Williams \966b

Type species. Oligosphaeridium complex (White) Davey and Williams 1966ft; O.D.

Oligosphaeridium anthophorum (Cookson and Eisenack) Davey 1969u

Plate 86, figs. 7, 8

1958 Hystrichosphaeridium anthophorum Cookson and Eisenack; 43, 44, pi. 11, figs. 12, 13.
1969a Oligosphaeridium anthophorum (Cookson and Eisenack) Davey; 147, 148, pi. 5, figs. 1-3.

These specimens compare well with those of Cookson and Eisenack (1958) except
that they are smaller and the processes appear to be a little more slender. The species
had a previously described geological range Oxfordian-Albian, and has gonyaula-
cacean affinities with the hystrichosphaeridioid lineage.

Figured material. Loc. 3, 7 (3) at 105-7-39-7. Loc. 9, 37 (1) at 92-0-28-2.

Dimensions. Range: Length of long axis 24 0 (33-3) 45-0 p.; length of short axis 2T0
(31-8) 43-0 p, the processes range in length from 12 to 34 p. Sixteen specimens
observed.

Oligosphaeridium pulcherrimum (Deflandre and Cookson) Davey and Williams

\966b

Plate 86, figs. 11, 12

1955 Hystrichosphaeridium pulcherrimum Deflandre and Cookson; 270, 271, pi. 1, fig. 8, text-
figs. 21, 22.

1966ft Oligosphaeridium pulcherrimum (Deflandre and Cookson) Davey and Williams; 75, 76,
pi. 10, fig. 9, pi. 1 1, fig. 5.

These specimens agree closely to those of Deflandre and Cookson (1955) except
in size; the Bearpaw material being smaller. The geological range of this species is
Valanginian-Lower Eocene. It has gonyaulacacean affinities with the hystricho-
sphaeridioid lineage.

Figured material. Loc. 11, 9 (1) at 103 0-44-6. Loc. 9, 39 (2) at 101-3T9.

Dimensions. Range; Length of long axis 20-0 (34-9) 43-0 p', length of short axis 23-0
(32-9) 40 0 p, processes range in length from 9 to 37 p. Eifty specimens measured,
from a studied population of seventy-four.

Genus polysphaeridium Davey and Williams 1966/t
Type species. Polysphaeridium subtile Davey and Williams 1966ft; O.D.

HARLAND: CAMPANIAN MICROPLANKTON

687

Polysphaeridium subtile Davey and Williams 19666

Plate 86, figs. 3, 6

I966^> Polysphaeridium subtile Davey and Williams; 92, pi. 11, fig. 1.

These specimens compare well with those of Davey and Williams (19666) except
in the morphology of the distal tip of the processes; those of Davey and Williams
(19666) are more serrate. There is, however, a certain amount of variability in the
Bearpaw specimens suggesting that these morphological variations may all be
encompassed in the concept of this species. P. subtile has a geological range of
Coniacian-Middle Miocene, and has gonyaulacacean affinities probably with the
hystrichosphaeridioid lineage.

Figured material. Loc. 3, 7 (2) at 107-2-42-6. Loc. 9, 39 (2) at 97 0-32-3.

Dimensions. Range: Length of long axis 22 0 (38-5) 50 0 p,', length of short axis 30 0
(35-9) 47 0 ju,, processes range in length from 5 to 14 p. Twenty specimens observed.

Genus tanyosphaeridium Davey and Williams 19666

Type species. Tanyosphaeridium variecalamum Davey and Williams \966b~, O.D.

Tanyosphaeridium variecalamum Davey and Williams 19666

Plate 86, figs. 13, 14

19666 Tanyosphaeridium variecalamum Davey and Williams; 98, 99, pi. 6, fig. 7, text-fig. 20.

A tentative tabulation from specimens where the processes appear to be restricted
to one per plate area is 6" , 6c, 6-7'", Ip, These specimens compare well in all
respects to those figured by Davey and Williams ( 1 9666). This species had a previously
recorded geological range of Albian-Cenomanian. The possession of a single reflected
antapical plate suggests a gonyaulacacean affinity for this cyst, probably with the
hystrichosphaeridioid lineage.

Figured material. Loc. 9, 39 (3) at 106 0-38 0, and 109 0-40-9.

Dimensions. Range: Length 25 0 (3T4) 37 0 p; breadth 15 0 (22-3) 42 0 p, length of
processes ranges from 6 to \3 p. Seven specimens observed.

Cyst-Family exochosphaeridiaceae Sarjeant and Downie emend. Davey 1969c
Genus exochosphaeridium Davey, Downie, Sarjeant and Williams 1966

Type species. Exochosphaeridium phragmites Davey, Downie, Sarjeant, and Williams 1966; O.D.
Exochosphaeridium cf. phragmites Davey, Downie, Sarjeant, and Williams 1966

Plate 86, fig. 10

1966 Exochosphaeridium phragmites Davey, Downie, Sarjeant, and Williams; 165, 166, pi. 2,
figs. 8-10.

Description. Chorate cyst, subspheroidal in shape, made up of periphragm and
endophragm closely adpressed; the former makes up the processes. Test granulate.

688

PALAEONTOLOGY, VOLUME 16

A large apical process present and is one-sixth to one-fifth the length of the cyst,
carries granules, branched, and distally acuminate. The other processes appear
randomly distributed on the test and are slender, solid, tapering, and distally
acuminate. No tabulation seen. The archeopyle precingular of the P or 2P type.

Figured material Loc. 9, 39 (3) at 106 0-35-5.

Dimensions. Range: Length 52 0 (62 0) 84 0 breadth 43 0 (50-3) 63 0 fx, processes
vary in length from 2 to 8 |Li. Six specimens observed.

Remarks. Except in lacking the pitted nature of the central body and possessing
smaller processes, the Bearpaw specimens compare favourably with those of Davey
et ai (1966). This species had a previously recorded range of Albian-Cenomanian
and has possibly gonyaulacacean affinities in the apteodinioid lineage.

Exochosphaeridium pseudohystrichodinium (Deflandre) emend. Davey 1969u

Plate 86, fig. 4

1937 Hystrichosphaeridium pseudohystrichodinium Deflandre; 73, pi. 15, figs. 3, 4.

1969fl Exochosphaeridium pseudohystrichodinium (Deflandre) emend. Davey; 163, 164, pi. 11,
figs. 4, 5.

These specimens compare well with those of Davey (1969u) except for the detail of
the pitted test. All specimens attributed to this species were granulate. The posses-
sion of a slight cingulum in some suggests that this species should be transferred to
Trichodinium Eisenack and Cookson 1960 but as this is the exception rather than the
rule it is assigned as above. It may, however, prove necessary in the future to treat
Exochosphaeridium as a junior synonym of Trichodinium. This species has a geological
range of Cenomanian-Eocene, and has possibly gonyaulacacean affinities with the
apteodinioid lineage.

Figured material Loc. 11,5 (2) at 110 0-3L7.

Dimensions. Range: Length 53 0 (60T) 710 /li; breadth 44 0 (53-3) 69 0 p., processes
range in length from 10 to 20 /x. Six specimens observed.

Exochosphaeridium sp. A

Plate 86, fig. 9

Description. Chorate cyst, subspheroidal in shape, made up of periphragm and
endophragm closely adpressed; the former making up the processes. Test smooth.
The apical process large, up to one-fifth of the cyst length and branched. The processes
solid to membranous, cylindrical, erect to curved and distally oblate to bifurcate.
In plan the membranous processes give a reticulate pattern to the cyst surface.
Tabulation not observed nor the archeopyle.

Eigured material Loc. 5, 2 (1) at 1010-33-6.

Dimensions. Range: Length 50 0 (62-8) 98 0 /x; breadth 43 0 (49-8) 60 0 p., processes
range in length from 5 to 10 /x. Five specimens observed.

Remarks. These specimens are unlike all other previously described species of
Exochosphaeridium but the scarcity of specimens precludes the erection of a new

HARLAND: CAMPANIAN MICROPLANKTON

689

species. The distinguishing feature is the apparent reticulate pattern on the cyst
surface. It has possibly gonyaulacacean affinities with the apteodinioid lineage.

Cyst-Family areoligeraceae Evitt emend. Sarjeant and Downie 1966
Genus cyclonephelium Deflandre and Cookson emend. Cookson and Eisenack

1962

Type species. Cyclonephelium compactum Deflandre and Cookson 1955; O.D.

The original diagnosis of this genus was emended by Cookson and Eisenack
(1962) and later by Williams and Downie (1966) to correct the interpretation of cyst
orientation and to describe fully the range of process structure. The later emendation
of Williams and Downie (1966) is nearly word for word the same as that of Cookson
and Eisenack (1962).

Cyclonephelium distinctum Deflandre and Cookson 1955

Plate 87, figs. 1, 4

1955 Cyclonephelium distinctum Deflandre and Cookson; 285, 286, pi. 2, fig. 14, text-figs. 47, 48.

This species is recorded for the first time from Campanian rocks. Its previously
recorded range being Hauterivian-Santonian. A large degree of process variation is
observed for this species. C. distinctum is characterized by the isolated nature of the
process. C. compactum Deflandre and Cookson 1955 is very similar except the pro-
cesses are lamella-like. C. distinctum has gonyaulacacean affinities with the areo-
ligeroid lineage.

Figured material. Loc. 13, 5 (1) at 97 0-43-3. Loc. 10, 3 (1) at 108 0-34-6.

Dimensions. Range: Length 37 0 (48-9) 84 0 jx\ breadth 39 0 (5 T9) 87 0 jj,, processes
range in length from 2 to 18 |U. Thirty-eight specimens observed.

Cyst-Family spiniferitaceae Sarjeant 1970
Genus spiniferites Mantell ex Loeblich and Loeblich 1966

Type species. Spiniferites ramosus (Ehrenberg) Mantell 1854; S.D.

Spiniferites ramosus var. ramosus (Davey and Williams) Sarjeant 1970

Plate 87, fig. 7

1966fl Hystrichosphaera ramosa var. ramosa Davey and Williams; 33, 34, pi. 1, figs. 1, 6, pi. 3,
fig. 1, text-fig. 8.

This variety has a known geological range of Middle Barremian-Ypresian. It has
gonyaulaeacean affinities with the gonyaulacoid lineage.

Figured material. Loc. 9, 39 (1) at 99 0-38-2.

Dimensions. Range: Length 33 0 (39-2) 46 0 ju.; breadth 22 0 (30-3) 35-0 p., processes
range in length from 6 to p. Thirteen specimens observed.

690

PALAEONTOLOGY, VOLUME 16

Spiniferites ramosus var. multibrevis (Davey and Williams) Sarjeant 1970

Plate 87, fig. 3

1966a Hystrichosphaera ramosa var. multibrevis Davey and Williams; 35-37, pi. 1, fig. 4, pi. 4,
fig. 6, text-fig. 9.

This variety compares well with those figured by Davey and Williams (1966a).
It is characterized by its short gonal and sutural processes; it has a geological range
from Hauterivian-Eocene. Its affinities are gonyaulacacean with the gonyaulacoid
lineage.

Figured specimen. Loc. 11, 1 (1) at 103 0-32-4.

Dimensions. Range: Length 20 0 (33-9) 42 0 jli; breadth 18 0 (26-75) 35-0 /x, processes
range in length from 3 to 12 /x. Twenty specimens observed.

Spiniferites ramosus var. granosus (Davey and Williams) Sarjeant 1970

Plate 87, fig. 8

1966a Hystrichosphaera ramosa var. granosa Davey and Williams; 35, pi. 4, fig. 9.

These specimens compare well with those of Davey and Williams (1966a) except
in differences of process morphology probably due to specific variability. This
variety had previously been recorded only from the Eocene. Its affinities are
gonyaulacacean with the gonyaulacoid lineage.

Figured material. Loc. 3, 3 (3) at 96-S-53-5.

Dimensions. Range: Length 39-0 (43-25) 46-0 p.; breadth 32-0 (34-5) 38-0 p, processes
range in length from 9 to 19 |ix. Five specimens observed.

Spiniferites cf. porosus (Manum and Cookson) comb. nov.

Plate 87, fig. 5

1964 Hystrichosphaera poro.sa Manum and Cookson; 11, 12, pi. 2, figs. 1-5, text-fig. 2.

EXPLANATION OF PLATE 87

All figures at a magnification of x 600 unless otherwise stated.

Fig. 1. Cyclonephelium distinctum Deflandre and Cookson, lateral view. Fig. 2. Defiandrea tripartita
Cookson and Eisenack emend, Cookson and Manum, dorsal view. Fig. 3. Spiniferites ramosus var.
multibrevis (Davey and Williams), lateral view. Fig. 4. Cyclonephelium distinctum Deflandre and
Cookson, lateral view, showing the operculum. Fig. 5. Spiniferites cf. porosus (Manum and Cookson),
lateral view. Fig. 6. Defiandrea korojonensis Cookson and Eisenack, dorsal view. Eig. 7. Spiniferites
ramosus var. ramosus (Davey and Williams), lateral view. Fig. 8. Spiniferites ramosus var. granosus
(Davey and Williams), lateral view. Fig. 9. Defiandrea spectabilis Alberti, lateral view of specimen with
a conical epitract. Fig. 10. Defiandrea echinoidea Cookson and Eisenack, ventral view. Eig. 1 1. Spini-
ferites ramosus var. membranaceus (Rossignol), lateral view. Eig. 12. Achomosphaera cf hyperacantha
(Deflandre and Cookson) Davey et al., lateral view. Fig. 13. Spiniferites ramosus var. gracilis (Davey
and Williams), lateral view. Fig. 14. Defiandrea macrocysta Cookson and Eisenack, lateral view.
Fig. 15. Defandrea spectabilis Alberti, dorsal view of specimen with a bell-shaped epitract.

PLATE 87

12

HARLAND, Campanian microfossils

15

692

PALAEONTOLOGY, VOLUME 16

Description. Proximo-chorate cyst, spheroidal to ovoidal in shape, made up of
periphragm and endophragm closely adpressed. The endophragm may be thickened;
the periphragm makes up the processes. Cyst smooth. Tabulation present with the
fields delimited by sutural ridges; no specimen, however, seen in which the tabulation
could be deciphered. Processes hollow, latispinous, erect, buccinate, open distally,
fenestrate, and digitate. One or two specimens observed in which the cylindrieal
sutural processes were present. Archeopyle precingular of type P, formed by the loss
of plate 3".

Figured material. Loc. 3, 10 (2) at 109-4-44-8.

Dimensions. Range: Length 33 0 (38-8) 46 0 jx', breadth 22-0 (32-5) 40 0 /u,, processes
range in length from 8 to 18 |U. Seventeen specimens observed.

Remarks. This species is morphologically similar to H. porosa Manum and Cookson
1964 but in certain specimens a similarity with H. perforata Davey and Williams 1966
was apparent. It may be that there is a complete morphological range between these
two species. H. porosa had a recorded geological range of Aptian-Turonian. Its
affinities are gonyaulacacean with the gonyaulacoid lineage.

Spiniferites ramosus var. gracilis (Davey and Williams) Sarjeant 1970

Plate 87, fig. 13

1966a Hystrichosphaera ramosa var. gracilis Davey and Williams; 34, 35, pi. 1, fig. 5, pi. 5, fig. 6.

These specimens compare well with those of Davey and Williams (1966fl) except
that in the Bearpaw assemblages there is more variability in process length; this
may be an indication of a morphological trend from the ramosus type to the gracilis
type. The geological range of this variety is Cenomanian-Miocene. Its affinities are
gonyaulacacean with the gonyaulacoid lineage.

Figured material. Loc. 3, 3 (1) at 94-0-47-3.

Dimensions. Range: Length 28 0 (34-5) 49 0 fi; breadth 21-0 (30-5) 38 0 p., processes
range in length from 7 to 18 /x. Thirteen specimens observed.

Spiniferites cf. membranaceus (Rossignol) Sarjeant 1970

Plate 87, fig. 1 1

1964 Hystrichosphaera furcata var. membranacea Rossignol; 86, pi. 1, figs. 4, 9, 10, pi. 3, figs. 7, 12.
1966a Hystrichosphaera ramosa var. membranacea (Rossignol) Davy and Williams; 37, pi. 4,
figs. 8, 12.

1967 Hystrichosphaera membranacea (Rossignol) Wall; 102, 103, pi. 14, figs. 14, 15, text-fig. 2.
1970 Spiniferites membranaceus (Rossignol) Sarjeant; 76.

These specimens compare well with those of Davey and Williams (1966u) but
differ from those of Rossignol (1964) in lacking the two large dorsal antapical pro-
cesses. S. ramosus var. membranaceus had a previously recorded geological range of
Eocene-Recent, and its affinities are gonyaulacacean with the gonyaulacoid lineage.

Figured material. Loc. 11, 1 (1) at 96 0-29-3.

Dimensions. Range: Length 30 0 (35-6) 40 0 p; breadth 22 0 (27-8) 38 0 p, processes
range in length from 8 to 18 /x. Ten specimens observed.

HARLAND: CAMPANIAN MICROPLANKTON

693

Genus achomosphaera Evitt 1963

Type species. Achomosphaera ramulifera (Deflandre) Evitt 1963; O.D.

Achomosphaera cf. hyperacantha (Deflandre & Cookson) Davey and Williams 1969

Plate 87, fig. 12

1955 Hystrichosphaera hyperacantha Deflandre and Cookson; 264, 265, pi. 6, fig. 7.

1967 Hystrichosphaera hyperacantha (Deflandre and Cookson) Wall; 100, pi. 14, fig. 3.

1969 Achomosphaera hyperacantha (Deflandre and Cookson) Davey and Williams; 4.

Description. Proximo-chorate cyst, spheroidal to ovoidal, consisting of two closely
adpressed wall layers. Cyst surfaces smooth. Periphragm alone makes up the hollow
processes, which do not connect to the interior of the cyst. They are slender and
taeniate, erect, cylindrical to tapering; generally trifurcate with bifid tips. The
processes are, like the test, smooth. Sutural ridges absent in general although one or
two faint lines may be seen in some specimens. Archeopyle not observed.

Figured material. Loc. 13, 13 (2) at 94 0-32-6.

Dimensions. Range: Length 37-0 (38 0) 39-0 pt; breadth 32 0 (32-5) 33 0 /n, length of
processes 6 to \5 fx. Two specimens observed.

Remarks. This species has recently been formally transferred to the genus Achomo-
sphaera by Davey et al. (1969). The observed specimens were smaller than the original
specimens of Deflandre and Cookson (1955). Wall (1967) considers this species as
being a robust variety of H.furcata (Ehrenberg) Wetzel. It is, however, probably best
regarded as a morphotype within the ' Spiniferites complex’. Aehomosphaera hypera-
cantha has a recorded geological range of Lower Miocene?-Holocene. The geological
range should only be extended with some reserve on the evidence of only two
specimens. It is characterized, in particular, by the nature of its trifurcate processes.
It is almost certainly a gonyaulacacean dinoflagellate of the gonyaulacoid lineage.

Cyst-Family deflandreaceae Eisenack emend. Sarjeant and Downie 1966
Genus deflandrea Eisenack emend. Williams and Downie 1966

Type species. Deflandrea phosphoritica Eisenack 1938; O.D.

Williams and Downie (1966) state that this genus is represented by many species
that clearly overlap with regard to their morphology. It is probably best to regard
this genus as embracing a complex of morphotypes that show geographical and
evolutionary intergradation ; but at any one level in the stratigraphic column, a num-
ber of morphotypes may be recognized. The species described below are regarded
in this light. Vozzhennikova (1967) created two new genera, Chatangiella and
Australiella, and emended the genus Deflandrea in her treatment of Deflandrea-Mke.
cysts. The present author is reluctant to follow this scheme at the present time.

Deflandrea spectabilis Alberti 1959

Plate 87, figs. 9. 15

1959 Deflandrea spectabilis Alberti; 99, pi. 9, figs. 7, 8.

This species is common in the Bearpaw Formation, it has a large specific variation

694

PALAEONTOLOGY, VOLUME 16

as interpreted from the assemblages studied, and it is not difficult to visualize that
with more of a conical epitract it would appear very similar to D. cooksoui as figured
by Clarke and Verdier (1967); but with a bell-shaped epitract it is closer to the holo-
type of D. cooksoni Alberti (1959). Vozzhennikova (1967) regards this species as
a member of the genus Australiella. D. spectabilis has a geological range of Santonian-
Campanian. Manum (1963) reported a peridiniacean tabulation for certain species
of Deflandrea, and Wall and Dale (1968) place it in the deflandreoid lineage.

Figured material. Loc. 5, 1 (3) at 103 0-44-3. Loc. 11, 1 (1) at 95-0-35-9.

Dimensions. Range: Length 58 0 (65-5) 87-0 jn; breadth 32-0 (41-7) 50 0 ju,. Fifty
specimens measured, out of a studied population of seventy-four.

Deflandrea korojonensis Cookson and Eisenack 1958

Plate 87, fig. 6

1958 Deflandrea korojonensis Cookson and Eisenack; 27, pi. 4, figs. 10, 11.

This species may be distinguished from D. spectabilis by its over-all shape and the
absence of any type of tabulation. It forms a distinct morphotype within these
assemblages. It is, however, close to D. bakeri, Deflandre and Cookson 1955, the
major difference being the nature of ‘ornamentation’, D. bakeri having a punctate
test. Vozzhennikova (1967) regards this species as a member of the genus Australiella.
D. korojonensis has a geological range of Campanian-Maestrichtian. It has peri-
diniacean affinities with the deflandreoid lineage.

Figured material. Loc. 13, 5 (1) at 93 0-44-2.

Dimensions. Range; Length 70 0 (97-4) 132 0 /u,; breadth 40 0 (56-6) 82 0 Thirteen
specimens observed and measured.

Deflandrea echinoidea Cookson and Eisenack 1960

Plate 87, fig. 10

1960 Deflandrea echinoidea Cookson and Eisenack; 2, pi. 1, figs. 5, 6.

This species forms a distinct morphotype within the Bearpaw Formation by virtue
of its ‘ornamentation’, but it is otherwise morphologically similar to D. spectabilis.
D. echinoidea has a geological range of Albian-Campanian. It has peridiniacean
affinities with the deflandreoid lineage.

Figured material. Loc. 1 1, 7 (2) at 102 0-47-7.

Dimensions. Range: Length 60 0 (67 0) 71 0 ju.; breadth 45 0 (48-3) 50 0 fx. Two speci-
mens observed.

Deflandrea tripartita Cookson and Eisenack emend. Cookson and Manum 1964

Plate 87, fig. 2

1960 Deflandrea tripartita Cookson and Eisenack; 2, pi. 1, fig. 10.

This species is similar to D. victoriensis but differs in that it lacks all traces of
tabulation. Vozzhennikova (1967) regards this species as a member of the genus

HARLAND: CAMPANIAN MICROPLANKTON

695

Australiella. D. tripartita has a geological range of Turonian-Campanian. It has
peridiniacean affinities with the deflandreoid lineage.

Figured material. Loc. 4, 13 (2) at 107-8-30-9.

Dimensions. Range; Length 75 0 (940) 105 0 /it; breadth 35 0 (48-6) 620 fx. Six
specimens observed and measured.

Deflandrea macrocysta Cookson and Eisenack 1960

Plate 87, fig. 14

1960 Deflandrea macrocysta Cookson and Eisenack; 3, pi. 1, figs. 7, 8.

This species forms a distinct morphological type in Bearpaw assemblages. In
general it appears quite close to Trithyrodinium evittii Drugg 1967 but differs in
possessing an intercalary archeopyle formed by the loss of a single plate. It has
peridiniacean affinities with the deflandreoid lineage.

Figured material. Loc. 1, 13 (2) at 99-2-39 0.

Dimensions. Range; Length 440 (67-6) 78 0 fx; breadth 360 (47-6) 540 fx. Ten
specimens observed and measured.

Cyst-Family pseudoceratiaceae Eisenack emend. Sarjeant and Downie 1966
Genus odontochitina Deflandre 1935

Type species. Odontochitina operculata (Wetzel) Deflandre 1946 = Odontochitina silicorum Deflandre
1935; O.D.

Odontochitina operculata (Wetzel) Deflandre 1946

Plate 88, fig. 1

1933 Ceratium (Euceratium) operculatum Wetzel; 170, pi. 2, tigs. 21, 22.

1946 Odontochitina operculata (Wetzel) Deflandre; 238, figs. 1016-19.

This was the only species of Odontochitina observed in the Bearpaw assemblages.
It is distinguished from O. costata Alberti emend. Clarke and Verdier 1967, which
in many respects it closely resembles, by lack of striations on the horns. It is uncertain
whether the horns of Odontochitina are equivalent to horns as seen in many modern
species of Ceratium or whether they should be more correctly termed processes. Its
affinities are unknown.

Figured material. Loc. 9, 39 (2) at 99 0-32-9.

Dimensions. Range; Endoblast; length of long axis 42 0 (53-9) 7T0 /x, length of short
axis 39 0 (49-9) 68 0 fx. The horns range in length from 54 0-200 0 [x. Thirty-one
specimens observed.

In addition to these species the following were noted, but were only represented
by single specimens. They are, therefore, not herein described or figured but are
included in the primary data and in the range charts.

Cribroperidinium sp.

Pareodinia sp.

696

PALAEONTOLOGY, VOLUME 16

Komewuia cf. glabra Cookson and Eisenack 1960

Canningia cf. rotundata Cookson and Eisenack 1961

Exochosphaeridium bifidum (Clarke and Verdier) Clarke et al. 1968

Cyclonephelium cf. paucispimmi Davey 1 969

Spiniferites cornutus (Gerlach) var. A

Deflaudrea gramdifera Manum 1963

D. sp.

Hexagoiiifera chlamydata Cookson and Eisenack \962b

INCERTAE SEDIS

Group ACRITARCHA Evitt 1963

Subgroup ACANTHOMORPHiTAE Downie, Evitt, and Sarjeant 1963
Genus baltisphaeridium Eisenack emend. Downie and Sarjeant 1963

Baltisphaeridium sp. A. (PI. 88, fig. 5.) Loc. 11, 9 (1) at 105 0-38-0. Thirty-one
specimens.

Genus Micrhystridium Deflandre emend. Downie and Sarjeant 1963

Micrhystridium sp. A. (PI. 88, figs. 2, 3.) Loc. 2, 1 (1) at 93-0-45-4 and Loc. 3, 13 (2)
at 108 0-35-6. Twenty-three specimens.

Micrhystridium sp. B. (PI. 88, fig. 7.) Loc. 3, 13 (2) at 109 0-39-4. Twenty-seven
specimens.

Micrhystridium sp. C. (PI. 88, fig. 4.) Loc. 3, 10 (3) at 93-5-38T. Seven specimens.
Micrhystridium sp. D. (PI. 88, fig. 6.) Loc. 3, 10 (2) at 94-6-43-2. Nine specimens.

Subgroup HERKOMORPHiTAE Downie, Evitt and Sarjeant 1963
Genus cymatiosphaera Wetzel emend. Deflandre 1954

Cymatiosphaera sp. A. (PI. 88, fig. 8.) Loc. 3, 1 (1) at 106-0^7-L Ten specimens.

INTERPRETATION OF THE BEARPAW PHYTOPLANKTONIC RECORD

The Bear paw assemblage. The Bearpaw Formation contains a distinct assemblage of
dinoflagellate cysts and acritarchs. In particular it is characterized by the presence of
the genera Defiandrea, Diconodinium, and Lejeunia. To characterize the assemblage
further it is necessary, though diffieult, to give some idea of the relative proportions
of certain of the cysts present. In a qualitative sense, therefore, the following three
categories are used: ‘common’, ‘occasionally common’, and ‘rare’.

EXPLANATION OF PLATE 88

All figures at a magnification of x600 unless otherwise stated.

Fig. I . Odontochitina operculata (Wetzel) Deflandre, lateral view, showing the archeopyle and the nature
of the horns. Fig. 2. Micrhystridium sp. A, general view to show the cyst habit. Fig. 3. Micrhystridium
sp. A, general view, showing the nature of the processes. Fig. 4. Micrhystridium sp. C, general view,
showing the nature of the processes. Fig. 5. Baltisphaeridium sp. A, general view, showing the cyst
habit. Fig. 6. Micrhystridium sp. D, general view, showing the nature of the processes. Fig. 7. Micrhy-
stridium sp. B, general view, showing the cyst habit. Fig. 8. Cymatiosphaera sp. A, general view, show-
ing the polygonal fields on the central body and the nature of the processes.

PLATE 88

HARLAND, Campanian microfossils

698

PALAEONTOLOGY, VOLUME 16

The Bearpaw assemblages studied always had high proportions of the following
cysts: Defiandrea spectabUis Alberti, Diconodinium firmum sp. nov., and Oligo-
sphaeridium pulcherrimum (Deflandre and Cookson) Davey and Williams. These
species are regarded as being ‘common’. Some of the assemblages, in addition to
those mentioned above, contain high proportions of Canningia senonica Clarke and
Verdier, Microdinium irregulare Clarke and Verdier, Lejeunia ampla sp. nov., L.
tricuspis (Wetzel), Odontochitina operculata (Wetzel) Deflandre, Cyclonephelium
distinctum Deflandre and Cookson, Hystrichosphaeridium tubiferum var. brevispinum
Davey and Williams and Defiandrea korojonensis Cookson and Eisenack. These
species are regarded as ‘occasionally common’. All other species recovered from the
Bearpaw Formation are ‘rare’.

Comparisons with other assemblages are diflicult to make as little work has been
published with respect to Campanian microplankton. Cookson and Eisenack (1960)
described some types from Western Australia with some similar Defiandrea species
to those from the Bearpaw Formation. In addition they recorded a unique collection
of species. Clarke and Verdier (1967) recorded very few species of dinoflagellate cysts
and acritarchs from the Campanian of the Isle of Wight. Species common to their
assemblage and the present assemblage are Exochosphaeridium bifidum (Clarke and
Verdier) Clarke et ai. 1968, Cycionepheiiwn distinctum, and Odontochitina opercuiata.
Vozzhennikova (1967), in her tables of diagnostic species, lists the following for
the Campanian of Kazakhstan: Gymnodinium kasachstanium Vozzhennikova,
Austraiieiia cooksoni (Alberti) Vozzhennikova, A. granuiifera (Manum) Vozzhenni-
kova, Aibertia curvicornis Vozzhennikova and Cooksonieiia manumi Vozzhennikova.
Similarities exist to the Bearpaw assemblages especially with regard to the Defiandrea
species. Oltz (1969) recorded the presence of Defiandrea ah', microgranuiata Stanley,
Defiandrea sp., Hystrichosphaeridium aff. tubiferum (Ehr.) Deflandre, Hystricho-
sphaeridium, cf. Gonyauiacysta, Forma ‘A’, Forma ‘B’, Forma ‘C’, and Paieo-
tetradinium sp. from the Bearpaw Formation of east central Montana. His assemblage
appears similar to that described in this work but a full comparison is not possible
as he failed to place his specimens in formal taxa. Recently Davey {\969b, 1969c)
described dinoflagellate cysts from the Campanian of South Africa. His assemblages
do not appear comparable except for the presence of Diconodinium spp. and Exocho-
sphaeridium bifidum.

Locai biostratigraphy. The Bearpaw Formation is interpreted as containing a number
of informal microplankton assemblage zones. At Fethbridge three informal
assemblage zones are recognized from the distribution of the contained micro-
plankton. These have been labelled I to III on fig. 12. The primary data, i.e. number
of specimens of each organic-walled microplankton species per assemblage, on
which figs. 12 and 13 are based may be obtained from the author on request.

The first assemblage zone encompasses a body of rock contained between 10 feet
above the base of the Bearpaw to approximately 160 feet above the base of the
formation.

The second informal assemblage zone consists of a body of rock 190 feet
above the base of the formation to approximately 260 feet above the base of the
formation.

HARLAND: CAMPANIAN MICROPLANKTON

699

The third assemblage zone consists of a body of rock from 305 feet above the base
of the Bearpaw to approximately 540 feet above the base of the formation.

In the Cypress Hills sections only two informal assemblage zones were initially
recognized, as part of the Manyberries Member could not be sampled because of
poor exposure and lack of stratigraphic control. A lower assemblage zone was recog-
nized (see fig. 13), and an upper assemblage zone was also present although its full
limits could not be ascertained.

Recently the author, through the courtesy of Dr. J. H. Wall of the Research
Council of Alberta, has examined a part of the core of water borehole RCA Thelma
(Lsd 14, Sec. 31, Tp. 6, R. 2, W 4th Mer) to complete that part of the Cypress Hills
sections not originally studied. The additional data has been added to all the relevant
diagrams and the samples and slides have been deposited in the Palynological Col-
lections at the Institute of Geological Sciences, Leeds, and registered as SAL 1720-
SAL 1726. The samples from RCA Thelma confirm the presence of informal assem-
blage zone II in the Cypress Hills based on the use of the semi-quantitative procedures,
i.e. percentages of dinoflagellates and acritarchs, the number of dinoflagellate cyst
species and on the gonyaulacacean ratio discussed below, but not on the ranges of
the organic-walled microplankton (see fig. 13).

The recognition of the majority of these informal microplankton assemblage zones
rests on the vertical distribution of the microplankton in the Bearpaw Formation.
A clear correlation by species inspection between these two areas is not possible but
both areas do reflect, in the distribution of the dinoflagellate cysts and acritarchs, the
transgression and various environmental changes of the Bearpaw sea.

Palaeoemironment of the Bearpaw. The Lethbridge section of the Bearpaw Formation
was examined for its foraminiferan content by Anan-Yorke (1969). He recognized
six cycles of water-depth fluctuations and it was suggested that salinity changes
accompanied these fluctuations. An attempt has been made to recognize these
fluctuations using dinoflagellate cysts and acritarchs.

A record was kept of the percentage of dinoflagellate cysts and acritarchs present
in the total palynomorph population in each sample examined. This information is
shown in column A of figs. 12 and 13. In addition, in those samples that were studied
in detail, i.e. those where the percentage of dinoflagellates and acritarchs rose to
10% or more, a record of the number of dinoflagellate species was kept, and this is
shown in column B of the two range charts. Column C records the gonyaulacacean
ratio, for each of the samples studied in detail. The gonyaulacacean ratio is simply
the number of species that have a gonyaulacacean affinity divided by the number of
species having a peridiniacean affinity. If we assume that conditions have not radi-
cally altered from today, then it appears that in an open marine environment the
number of gonyaulacacean dinoflagellate species is relatively higher than the number
of peridiniacean dinoflagellate species (Schiller 1937). We must assume that this is
reflected in the cyst populations. In Wall (1967) the calculated gonyaulacacean ratio
is 18 0 for cysts collected from deep-sea cores in the Caribbean, and Wall and Dale
(1968) has a calculated gonyaulacacean ratio of 0-44 for a near-shore cyst population
at Woods Hole, Massachusetts. Freshwater assemblages have high proportions of
peridiniacean dinoflagellates and low proportions of gonyaulacacean dinoflagellates

K

Diconodinium arcticum
Oligosphaeridiun' pulcherrimum
Lejeunia parva
Oeflandrea koro)onensis
Baltisphaeridium sp. A
Diconodinium firmum

Hyslrichosphaendium tubiferum var. brevispinum

Cyclonephelium distinctum

Oeflandrea macrocysta

Hystnchosphaeridium cf arbori&pinum

Polyspbaeridium subtile

Eiochosphaeridium sp. A

Spiniferites ramosus var. membranaceus

Spiniferites ramosus var. gracilis

Micrhystridium sp. A

Spiniferites ramosus var. granosus

Tanyosphaeridium variecalamum

Oeflandrea spectabilis

Cymatiosphaera sp. A

Dinogymnium longicornis

Lejeunia ampla

Spiniferites ramosus var. multibrevis
Spiniferites cf. porosus
Micrbysiridium sp. B
Cleistosphaeridium divers ispinosum
^ Coronifera oceanica
Spiniferites cornutus var. A
Odontochitina operculata
Micrhystridium sp. C
Hystnchosphaeridium dowlingii
Oligosphaeridium anthophorum
Spiniferites ramosus var. ramosus
Lejeunia tricuspis
Micrhystridium sp. 0
Exochosphaeridium cf. phragmites
Hystrichosphaeridium salpingophorum
Microdinium cf. irregulare
Cleistophaeridium sp. A
^ Membranosphaera cf. maastrichtica
Spinidinium clavum
Hexagonifcra chlamydata
Oeflandrea tripartita
Exochosphacndium cf. bifidum

TEXT-FIG. 12. Vertical ranges of the dinoflagellate cysts and acritarchs recovered from the Lethbridge
sections of the Bearpaw Formation together with the proposed informal biostratigraphical zonation.
Column A— Percentages of phytoplankton per sample.

Column B— Number of dinoflagellate cyst species per sample.

Column C— Gonyaulacacean ratio.

HARLAND: CAMPANIAN MICROPLANKTON

701

as may be seen in Eddy (1930) and Thompson (1947, 1950). The gonyaulacacean
ratio was calculated for each of the samples studied in detail, using cysts for which
their natural affinities are known or reasonably assured. The results are shown in
column C of the range charts.

An inspection of these three columns (A-C) from the Lethbridge section reveals
that an increase in the percentages of the dinoflagellates and acritarchs is accompanied
by an increase in the number of species and by an increase in the gonyaulacacean
ratio. This relationship may be used to pinpoint the appearance of possible open
marine conditions. The author has no justification in setting definite limits on this
environment, in terms of salinity, temperature, etc., so the term open marine is used
only in a relative sense. These open marine conditions may be indicative of periods
of maximum extent of the Bearpaw sea such that the shoreline is distant with the
accompanying normal salinity for ocean waters. It might, however, indicate periods
when the nutrient content of the sea was optimum for the phytoplankton. Using this
information it may be interpreted that an initial flooding or transgression in the
Lethbridge area is represented by the lowermost 150 feet of the formation with
optimum open marine conditions between 60 and 100 feet above the base of the
formation. A second period of open marine conditions is represented from approxi-
mately 180-240 feet above the base of the formation and a third period is represented
from approximately 310-350 feet above the base of the formation. Many more minor
fluctuations are apparent in column A, but on an individual basis these are difficult
to explain and may indeed be entirely spurious.

In comparing the results from the dinoflagellate cysts and acritarchs with those
from the foraminiferans a close similarity is evident. Anan-Yorke (1969) categorizes
his zones as follows: (5) 450 feet to Ryegrass Member — lagoonal. (4) 200^50 feet —
deeper water but not as deep as in 2. (3) 115-200 feet— lagoonal, brackish. (2) 56-
115 feet— open marine. (1) Basal 56 feet— lagoonal, brackish. Anan-Yorke’s zone 2
corresponds quite well with the time of optimum conditions for the initial flooding
of the Bearpaw sea as documented by the dinoflagellates and acritarchs. These fossils
also pick out a second period of optimum conditions at the level of the Magrath
Member which Anan-Yorke unfortunately failed to examine for foraminiferans
because of lack of samples. Dinoflagellate and acritarch evidence suggests fluctuating
conditions for zone 5 of Anan-Yorke. In the Cypress Hills the period of initial
flooding can be recognized together with the other open marine periods. A correla-
tion of the marine palaeoenvironments is thus achieved.

Although the use of the gonyaulacacean ratio appears to be a useful technique it
may, in this case, more precisely indicate that the Deflandrea spp. prefer a near-shore
reduced salinity situation; as the majority of the peridiniacean cysts studied were
species of the genus Deflandrea. Certain over-all limitations do exist. The palyno-
logist, in studying microplankton, is looking at the cyst and not the motile stage of
the life cycle. To what extent does the cyst population reflect the true population of
these organisms in their natural habitat? Under what conditions do cysts form and
what factors have affected their distributions in the sediments from which they are
extracted? These two largely unanswered questions clearly point out the present
limitations of all dinoflagellate and acritarch research.

10 20 30

Deflandrea spectabilis

Diconodinium arcdcum

Odontochitina operculata

Oligosphaendium pulcherrimum

Cyclonephelium distinctum

Hystrichosphaeridium salpingophorum

Spiniferites ramosus var. ramosus

Deflandrea korojonensis

Deflandrea macrocysta

Diconodinium firmum

Micrhystndium sp. B

Dinogymnium longicornis

Hystrichosphaeridium cf. arbonspinum

Hystrichosphaeridium tubiferum var. brevispinum

Lejeunia tricuspis

Cleistosphaendium diversispinosum

Leieunia ampla

Eiochosphacndium cf. phragmites
Spiniferites cf. porosus
Microdinium cf. irregulare
Spinidinium clavum
Canningia cf. rotundata
Exochosphaendium pseudohystrichodinium
Baltisphaeridium sp. A
Hystrichosphaeridium dowlingii
Dinogymnium cf. albertii
Komewuia cf. glabra
Micrhystridium sp. D
Achomosphacra hypcracantha
Polysphaeridium subtile
Spiniferites ramosus var. gracilis
Oligosphaeridium anthophorum
Tanyosphaeridium variecalamum
Deflandrea granulifcra
Micrhystridium sp. A
Deflandrea tripartita
Spiniferites ramosus var. multibrevis
Canningia senonica
’ Coronifera oceanica
Cleistosphaendium sp. A
Cribroperidinium sp. A
Spiniferites ramosus var. membranaccus
Deflandrea cchinoidca
Micrhystridium sp. C
’ Uvatodinium cf. nasalum
Cymatiosphacra sp. A
?Membranosphaera cf. maasirichlica
Aptcodinium sp. A
Cyctonephclium cf. paucispinuni

TEXT-i iG. 13. Vertical ranges of the dinoflagellate cysts and acritarchs recovered from the Cypress Hills
sections of the Bearpaw Formation together with the possible equivalent informal biostratigraphical units
to those proposed for the Lethbridge area. Column explanations the same as those given for text-fig. 12.

HARLAND: CAMPANIAN MICROPLANKTON

703

Acknowledgements. This work forms part of a doctoral thesis presented to the University of Alberta.

1 acknowledge the advice given to me by Drs. C. R. Stelck, G. Playford, and W. R. Evitt. Special thanks
are due to Dr. J. H. Wall of the Research Council of Alberta for introducing the author to the Bearpaw
Formation and for supplying many samples. Mr. R. Anan-Yorke kindly allowed the use of his unpublished
results. Financial aid was provided by the University of Alberta, by the National Research Council of
Canada, and by the Geological Survey of Canada. Finally, I would like to thank my wife, Patricia, for her
encouragement and patient assistance through all the stages of this work.

REFERENCES

ALBERTI, G. 1959. Zur Kenntnis der Gattung Deflandrea Eisenack (Dinoflag.) in der Kreide und im Alltertiar
Nord- und Mitteldeutschlands. Geol. Staatsinst. Hamburg, Mitt. 28, 93-105.

1961. Zur Kenntnis Mesozoischer und Alltertiaren Dinoflagellaten und Hystrichosphaerideen von

Nord und Mitteldeutschland sowie einigen Anderen Europaischen Gebieten. Palaeontographica, Abt. A,
116, 1-58.

ANAN-YORKE, R. 1969. A microfaunal study of the Bearpaw Formation, Lethbridge area, Alberta. Unpubl.
M.Sc. thesis, Univ. of Alberta, 1-126.

CALDWELL, w. G. E. 1968. The late Cretaceous Bearpaw Formation in the South Saskatchewan river valley.
Sask. Research Council, Geology Div. Kept., 8, 1-89.

CLARKE, R. F. A. and VERDIER, J. p. 1967. An investigation of microplankton assemblages from the Chalk
of the Isle of Wight, England. Verh. K. Ned. Akad. Wet. 24, 1-96.

DAVEY, R. J., SARJEANT, w. A. s. and VERDIER, J. p. 1968. A note on the nomenclature of some Upper

Cretaceous and Eocene dinoflagellate taxa. Taxon, 17, 181-183.

COOKSON, I. c. 1965. Cretaceous and Tertiary microplankton from south-eastern Australia. Proc. Roy.
Soc. Victoria, 78, 85-93.

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

Ibid. 70, 19-79.

1960. Upper Mesozoic microplankton from Australia and New Guinea. Palaeontology, 2,

243-261.

1961. Upper Cretaceous microplankton from the Belfast No. 4 bore, south-western Victoria.

Proc. Roy. Soc. Victoria, 74, 69-76.

1962fl. Some Cretaceous and Tertiary microfossils from Western Australia. Ibid. 75, 269-273.

19626. Additional microplankton from Australian Cretaceous sediments. Micropaleontology,

8, 487-507.

and HUGHES, N. f. 1964. Microplankton from the Cambridge Greensand (Mid-Cretaceous). Palaeonto-
logy, 7, 37-59.

and MANUM, s. 1964. On Deflandrea victoriensis n.sp., D. tripartita Cookson and Eisenack and related

species. Proc. Roy. Soc. Victoria, 77, 521-524.

DAVEY, R. J. 1969fl. Non-calcareous microplankton from the Cenomanian of England, Northern France
and North America. Part I. Bull. Br. Mus. nat. Hist. (Geol.), 17, 105-180.

19696. Some dinoflagellate cysts from the Upper Cretaceous of Northern Natal, South Africa.

Palaeont. afr. 12, 1-23.

1969c. The evolution of certain Upper Cretaceous hystrichospheres from South Africa. Palaeont.

afr. 12, 25-51.

1970. Non-calcareous microplankton from the Cenomanian of England, Northern France and

North America. Part II. Bull. Br. Mus. nat. Hist. (Geol.), 18, 333-397.

and WILLIAMS, G. l. 1966u. The genera Hystrichosphaera and Achomosphaera. In davey, r. j. et al..

Studies on Mesozoic and Cainozoic dinoflagellate cysts. Ibid. Supplement 3, 28-52.

19666. The genus Hystrichosphaeridium and its allies. In davey, r. j. et al.. Studies on Mesozoic

and Cainozoic dinoflagellate cysts. Ibid. 53-105.

downie, c., SARJEANT, w. A. s. and WILLIAMS, G. L. 1966. Fossil dinoflagellate cysts attributed to

Baltisphaeridium. In Studies on Mesozoic and Cainozoic dinoflagellate cysts. Ibid. 157-175.
1969. Appendix to ‘Studies on Mesozoic and Cainozoic dinoflagellate cysts’. Ibid.

1-24.

704

PALAEONTOLOGY, VOLUME 16

DEFLANDRE, G. 1934. Sur les microfossiles d’origine planctonique conserves a I’etat de matiere organique
dans les silex de la craie. C.R. Acad. Sci., 199, 96^968.

1935. Considerations biologiques sur les micro-organismes d’origine planctonique conserves dans

les silex de la craie. Bull. Biol. Fr. Belg. 69, 213-244.

1937. Microfossiles des silex cretaces II, Flagelles incertae sedis. Hystrichosphaeridees. Sarcodines.

Organismes divers. Ann. Paleont. 26, 51-103.

1954. Systematique des Hystrichosphaerides : sur I’acception du genre Cvmatiosphaera O. Wetzel.

Soc. Geol. Fr., C.R. Sonim. 12, 257-258.

1966. Addendum a mon memoire; Microfossiles des silex Cretaces. Cahiers de Micropaleontologie,

Arch. orig. Centre Docum. C.N.R.S. 419, 1-9.

and COOKSON, e c. 1955. Fossil microplankton from Australian Late Mesozoic and Tertiary sediments.

Aust. J. Mar. Freshwr. Res. 6, 242-313.

DIESING, K. M. 1866. Revision der Prothelminthen. Abtheilung: Mastigophoren. K. Akad. fViss. Wien,
Math-Nat. CL, Sitzber. 52, 287-401.

DOUGLAS, R. J. w. 1942. New Species of Inoceramus from the Cretaceous Bearpaw Formation. Trans. Roy.
Soc. Canada, Ser. 3, 36, 59-66.

DOWLING, D. B. 1917. The southern plains of Alberta. Geol. Survey Canada Mem. 93, 1 200.

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

EViTT, w. R. and sarjeant, w. a. s. 1963. Dinoflagellates, hystrichospheres and the classification of the

acritarchs. Stanford Univ. Publ., Geol. Sci. 7, No. 3, 3-16.

DRUGG, w. s. 1967. Palynology of the Upper Moreno Formation (Late Cretaceous-Paleocene), Escarpado
Canyon, California. Palaeontographica, Abt. B. 120, 1-71.

EDDY, s. 1930. The freshwater armored or thecate dinoflagellates. Trans. Am. Micro. Soc. 44, 277-
305.

EHRENBERG, c. G. 1832. Beitrage zur Kenntnis der Organisation der Infusorian und ihrer geographischen
Verbreitung, besonders in Sibirien. Abh. preuss. Akad. Wiss. 1830, 1-88.

EiSENACK, A. 1938. Die Phosphoritknollen der Bersteinformation als Uberleiferer tertiaren Planktons.
Phys.-dkon. Ges. Konigsb., Schr. 70, 181-188.

1958. Mikroplankton aus dem norddeutschen Apt nebst einigen Bemerkungen fiber fossile Dino-

flagellaten. N. Jb. Geol. Paldont., Abh. 106, 383-422.

-- 1967. Katalog der fossilen Dinoflagellaten, Hystrichosphdren und verwandten Mikrofossilien., Band 1
Dinoflagellaten. E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart, 1-895.

and COOKSON, i. c. 1960. Microplankton from Australian Lower Cretaceous sediments. Proc. Roy.

Soc. Victoria, 72, 1-11.

EVITT, w. R. 1963. A discussion and proposals concerning fossil dinoflagellates, hystrichospheres and
acritarchs. Proc. nat. Acad. Sci. 49, 158-164, 298-302.

1967. Dinoflagellate studies II. The archeopyle. Stanford Univ. Publ., Geol. Sci. 10, No. 3, 1-82.

CLARKE, R. F. A. and VERDiER, J. p. 1967. Dinoflagellate studies III. Dinogymnium acuminatum n. gen.,

n. sp. (Maastrichtian) and other fossils formerly referable to Gymnodinium Stein. Stanford Univ. Publ.,
Geol. Sci. 10, No. 4, 3-27.

FOLiNSBEE, R. E., BAADSGAARD, H. and LiPSON, J. 1960. Potassium-argon time scale. Internat. Geol. Congress,
XXI Session Rept, 7-17.

1961. Potassium-argon dates of Upper Cretaceous ash falls, Alberta, Canada. Annals New

York Acad. Sci. 91, 352-359.

FRASER, R. J., MCLEARN, F. H., RUSSELL, L. S., WARREN, P. S. and WICKENDEN, R. T. D. 1935. Gcology of SOUthcm
Saskatchewan. Geol. Survey Canada, Mem. 176, 1-137.

FURNiVAL, G. M. 1950. Cypress Lake map-area, Saskatchewan. Geol. Survey Canada, Mem. 242, 1 161.
GERLACii, E. 1961 . Mikrofossilien aus dem Oligozan und Miozan Nordwestdeutschlands, unter besonderer
Berucksichtigung der Hystrichosphaeren und Dinoflagellaten. N.Jb. Geol. Paldont. Abh. 112, 143
228.

GORKA, H. 1963. Coccolithophorides, Dinoflagellates, Hystrichosphaerides et microfossiles incertae sedis
du Cretace superieur de Pologne. Acta Palaeont. Polon. 8, 3-90.

HATCHER, J. B. and .STANTON, T. w. 1903. The stratigraphic position of the Judith river beds and their cor-
relation with the Belly river beds. Science, N.S. 18, 211-212.

HARLAND: CAMPANIAN MICROPLANKTON

705

LANJOUW, J. et al. 1966. International Code of Botanical Nomenclature. International Bureau for Plant
Taxonomy and Nomenclature of the International Association for Plant Taxonomy, Utrecht, 1-75.

LINES, F. G. 1963. Stratigraphy of Bearpaw Formation of southern Alberta. Bull. Canadian Petroleum
Geology, 11, 212-227.

LINK, T. A. and CHiLDERHOSE, A. J. 1931. Bearpaw Shale and contiguous formations in the Lethbridge area,
Alberta. Bull. Amer. Assoc. Petroleum Geologists, 15, 1227-1242.

LOEBLiCH, A. R. JR. and LOEBLiCH, A. R. III. 1966. Index to the genera, subgenera and sections of the
Pyrrhophyta. Stud. trop. Oceanogr. Miami, 3, 1-94.

LORANGER, D. M. and GLEDDiE, J. 1953. Some Bearpaw zones in southwestern Saskatchewan and southern
Alberta. Alta. Soc. Petroleum Geologists, 3rd Ann. Field Conference Guidebook, 158-175.

MANTELL, G. A. 1854. The Medals of Creation-, or, First Le.ssons in Geology and the study of Organic Remains.
2nd edition, Bohn, London, 1-930.

MANUM, s. 1960. Some dinoflagellates and hystrichosphaerids from the Lower Tertiary of Spitzbergen.
Nytt. Mag. Bot. 8, 17-24.

1963. Some new species of Deflandrea and their probable affinity with Peridinium, Norsk. Polarinst.

Arbok., 1962, 55-67.

and COOKSON, i. c. 1964. Cretaceous microplankton in a sample from Graham Island, Arctic Canada,

collected during the second ‘Fram’-expedition (1898-1902) with notes on the microplankton from the
Hassel Formation, Ellef Ringnes Island. Skr. Norska. Vid-Akad. Oslo, Mat. -Naturv. kl. (n.s.), 17, 1-36.

MARTIN, L. J. 1960. Tectonic framework of northern Canada. In raasch, g. o. (ed. ). Geology of the Arctic,
vol. I. University of Toronto Press, Toronto, 442-457.

MiLLiouD, M. E. 1969. Dinoflagellates and acritarchs from some western European Lower Cretaceous type
localities. Proc. First Intern. Conf. Planktonic Microfossils, 2, 420-434.

MORGENROTH, p. 1966. Neue in organischer Substanze erheltene Microfossilien des Oligozans. N. Jb.
Geol. Paldont., Abh. 127, 1-12.

NEVES, R. and dale, b. 1963. A modified filtration system for palynological preparations. Nature, 198,
775-776.

NORRIS, G., MCANDREW, J. H. and SARJEANT, w. A. s. 1965. A descriptive index of fossil Dinophyceae and
Acritarcha. Palaeont. Bull. 40, 1-72.

OLTZ, D. F., JR. 1969. Numerical analysis of palynological data from Cretaceous and Early Tertiary sedi-
ments in east central Montana. Palaeontographica, Abt. B, 128, 90-166.

PIERCE, R. L. 1959. Converting coordinates for microscope-stage scales. Micropaleontology, 5, 377-378.

pococK. s. A. J. 1962. Microfloral analysis and age determinations of strata at the Jurassic-Cretaceous
boundary in the western Canadian plains. Palaeontographica B, 111, 1-95.

REESiDE, J. B. JR. 1957. Paleoecology of the Cretaceous seas of the western interior of the United States.
In LADD, H. s. (ed.). Treatise on Marine Ecology and Paleoecology, vol. 2. Geol. Soc. America, Mem. 67,
1-1077.

ROSSiGNOL, M. 1964. Hystrichospheres du Quaternaire en Mediterranee Orientale, dans les sediments
Pleistocenes et les boues marines actuelles. Rev. Micropaleont. 7, 83-99.

RUSSELL, L. s. 1950. Correlation of the Cretaceous-Tertiary transition in Saskatchewan and Alberta. Bull.
Geol. Soc. America, 61, 21 -Al.

and LANDES, R. w. 1940. Geology of the southern Alberta plains. Geol. Survey Canada, Mem. 221,

1-223.

SAMOYLOVICH, s. A. et al. 1961. Pyl'tsa i spory zapadnoy Sibiri, Yuru-Paleotsen. Trudy vses. neft. nauchno-
issled. geol.-razv. Inst. 177, 1-659.

SARJEANT, w. A. s. 1966. Dinoflagellate cysts with Gonyaulax-iype tabulation. In davey, r. j. et al.. Studies
on Mesozoic and Cainozoic Dinoflagellate Cysts. Bull. Br. Mus. nat. Hist. (Geol.) Supplement 3, 107-156.

1967. The stratigraphical distribution of fossil dinoflagellates. Rev. Palaeobotan. Palvnol. 1, 323-

343.

1970. The genus Spiniferites Mantell, 1850 (Dinophyceae). Grana, 10, 74-78.

and DOWNiE, c. 1966. The classification of dinoflagellate cysts above generic level. Grana palvnol.

6, 503-527.

SCHILLER, J. 1935-1937. Dinoflagellata (Peridineae) in monographischer Behandlung. 1 1. /« rabenhorst, l.
(ed.), Kryptogainen-Flora von Deutschlands, Osterreichs and der Schweiz. 10, Sect. 3, Pt. 2, Nos. 1-4,
1-590.

706

PALAEONTOLOGY, VOLUME 16

STANLEY, E. A. 1965. Upper Cretaceous and Paleocene plant microfossils and Paleocene dinoflagellates and
hystrichosphaerids from northwestern South Dakota. Bull. American Paleontology, 49, 179-384.
THOMPSON, R. H. 1947. Freshwater dinoflagellates of Maryland. Contr. Chesapeake. Biol. Lab. 67, 3-24.

1950. A new genus and new records of freshwater Pyrrophyta in the Desmokontae and Dinophyceae.

Lloydia, 13, 277-299.

vozzHENNiKOVA, T. F. 1963. Typ Pyrrhophyta. In Osnovi Paleontologii, 182, 171-186.

1965. Vvedenye v izuchenye iskopayemyx Perideyvyx vodoroslei. Akad. Nauk. SSSR Sibirskoe

Otledeinie Inst. Geol. Geofiz. 1-156.

1967. Iskopayemiye peridineiyurskikh, myelovikh, palaeogenovikh otlozheniy S.S.S.R. Ibid. 1-347.

WALL, D. 1965. Modem hystrichospheres and dinoflagellate cysts from the Woods Hole region. Grana
palynol. 6, 297-314.

1967. Fossil microplankton in deep-sea cores from the Caribbean Sea. Palaeontology, 10, 95-123.

and DALE, B. 1968. Modem dinoflagellate cysts and evolution of the Peridiniales. Micropaleontology,

14, 265-304.

WARREN, p. s. 1931. Invertebrate paleontology of southern plains of Alberta. Bull. Amer. Assoc. Petroleum
Geologists, 15, 1283-1291.

1934. Paleontology of the Bearpaw Formation. Trans. Roy. Soc. Canada, Ser. 3, 81-100.

1937. A rhynchonellid brachiopod from the Bearpaw Formation of Saskatchewan. Ibid., Ser. 3,

31, 1-4.

and STELCK, c. R. 1958. Continental margins of western Canada in pre-Jurassic time. Alta. Soc.

Petroleum Geologists, 6, 29-42.

WETZEL, o. 1933. Die in organischer Substanze erheltenen Mikrofossilien des baltischen Kreide-Feuersteins.
Palaeontograpliica, Abt. A, 78, 1-110.

WILLIAMS, G. D. and BURK, c. F. JR. 1964. Upper Cretaceous. In mccrossan, r. g. and glaister, r. p. (ed.).
Geological History of Western Canada. Alta. Soc. Petroleum Geologists, Calgary, 169-189.

WILLIAMS, G. L. and downie, c. 1966. Further dinoflagellate cysts from the London Clay. In davey, r. j.
et al.. Studies on Mesozoic and Cainozoic Dinoflagellate Cysts. Bull. Br. Mus. nat. Hist. {Geol.) Supple-
ment 3, 215-236.

WILLIAMS, M. Y. and DYER, w. s. 1930. Geology of southern Alberta and south-western Saskatchewan. Geol.
Survey Canada, Mem. 163, 1-160.

WILSON, G. J. 1967. Some new species of Lower Tertiary dinoflagellates from McMurdo Sound, Antarctica.
N.Z. J. Bot. 5, 57-83.

APPENDIX A: SAMPLE LOCALITIES

Sample localities previously mentioned in the text are listed below with their code numbers and geographical
locations.

Lethbridge Area. Loc. 1 : Lsd. 1, Sec. 2, Tp. 7, R. 22, W.4th Mer.

Loc. 2: Lsd. 9-10, Sec. 33, Tp. 6, R. 22, W.4th Mer.

Loc. 3: Lsd. 15, Sec. 32, Tp. 6, R. 22, W.4th Mer.

Loc. 4; Lsd. 11, Sec. 19, Tp. 6, R. 22, W.4th Mer.

Loc. 5 : Lsd. 10-15, Sec. 24, Tp. 6, R. 23, W.4th Mer.

Loc. 6: Lsd. 12, Sec. 34, Tp. 9, R. 23, W.4th Mer.

Loc. 7: Lsd. 1-2, Sec. 32, Tp. 9, R. 23, W.4th Mer.

Cypress Hills Area. Loc. 8: Lsd. 6, Sec. 31, Tp. 11, R- 2, W.4th Mer.

Loc. 9: Lsd. 12, Sec. 14, Tp. 11, R. 3, W.4th Mer.

Loc. 10: Lsd. 6, Sec. 32, Tp. 5, R. 4, W.4th Mer.

Loc. 1 1 ; Lsd. 7, Sec. 32, Tp. 5, R. 2, W.4th Mer.

Loc. 12: Lsd. 12, Sec. 5, Tp. 6, R. 2, W.4th Mer.

Loc. 13: Lsd. 5, Sec. 25, Tp. 6, R. 3, W.4th Mer.

Loc. 14: Lsd. 1-2, Sec. 7, Tp. 8, R. 3, W.4th Mer.

REX HARLAND

Institute of Geological Sciences
Ring Road, Halton
Leeds, LSI 5 8TQ

Typescript submitted 10 July 1972

ON THE MODE OF BRANCHING IN A NEW
SPECIES OF CLONOGRAPTUS

by D. E. JACKSON

Abstract. The precise nature of dichotomous branching in pyritized specimens of Clonograptus aureus sp. nov. of
Tremadocian age is described and is compared with Arenigian species of Loganograptus and Goniograptus from
Scandinavia. The manner whereby new branches are generated is analogous to the proximal end development of
Didymograptus minutus.

Although the mode of development of first-order stipes has been worked out in
considerable detail for several multiramous and pauciramous graptolites, little is
known about the precise way in which higher order stipes branch laterally or dichoto-
mously. This new material from the Road River Formation in Yukon Territory
is therefore of great interest because its exquisitely pyritized rhabdosome indicates
how second-, third-, and fourth-order branches are generated. The absence of bithecae
in this species indicates that at least one line of clonograptids lost this dendroid
characteristic before the close of the Tremadocian.

SYSTEMATIC DESCRIPTION

Family anisograptidae Bulman 1950
Genus clonograptus Hall and Nicholson 1873
Clonograptus aureus sp. nov.

Text-figs. 1 A-c, 2a, 3 a-c

Material. Three incomplete rhabdosomes are available preserved in full relief in pyrite. Illustrated specimens
comprising GSC 27096 (Holotype) and 27098 as well as 27097 and 27099 come from the Road River
Formation in the Upper Canyon on Peel River, Yukon Territory (65° 56' N., 134° 51' W.).

Horizon. Upper part of Tremadocian Bryograptus-Clonograptus Zone of Jackson and Lenz (1962, p. 33).
The approximate stratigraphic relationship is known to be 60 metres below the base of Tetragraptus
approximatus Zone and 120-150 metres above the top of the Staurograptus Zone. In 1969 the writer recol-
lected at this stratigraphic level and found Adelograptus cf. victoriae (T. S. Hall), Adelograptus{l) antiquus
(T. S. Hall), Dictyonema pulchellum T. S. Hall, and 1 Tetragraptus decipiens T. S. Hall. Such a fauna is
clearly indicative of an La 2 age in Australia.

Derivation of name. From Latin aureus = golden : referring to colour of pyritized rhabdosome.

Description. Rhabdosome not seen to exceed 25 mm across with dichotomous
branching to 5th-order. The funicle is 2-5-2-9 mm long, details of proximal end
development not seen. Second-order branches 2-3 mm long diverge so that distally
they enclose angles of 110 to 120°; each branch probably composed of 3-4 thecae
(see text-fig. 1b). Third and fourth-order branches 3 mm-5-5 mm long and 0-3 mm
wide in dorsal view; one fourth-order branch (Text-fig. Ic) preserved in profile is
0-6 mm across thecal aperture, free ventral wall of thecae 10 mm long, concave, and
inclined at 20°-30°. The thecal rate on third-order stipes is 4 in 4 mm. Associated

[Palaeontology, Vol. 16, Part 4, 197.3, pp. 707-711.]

(n-f4f

(n+3)'’

(n+2)‘’"'

(n+lf

dlchotomy(2)

TEXT-FIG. 1a-c. Clonograptus
aureus sp. nov. from beds of
Tremadocian age in the Road
River Formation on Peel River,
Yukon Territory, a, dorsal view
of rhabdosome of GSC 27096
and location of dichotomies
illustrated in Text-figs. 1 b, c.

S = sicula. B, dorsal view of
dichotomies 2, 3, and 4, xl5.
c, dorsal and lateral views of
dichotomy I, X 15. Note: two
small dots on diagrams repre-
sent 1 mm. B

JACKSON: BRANCHING IN CLONOGRAPTUS

709

TEXT-FIG. 2a. Pyritized sicula GSC 27098 associated and possibly conspecific with Clonograptus aureus
sp. nov. ; X 33. b, Schematic thecal diagram illustrating dicalycal nature of th (n-|- 1)**; c, proximal end of
Loganograptus kjerulfi Herrmann from Galgeberg, paratype 73123, Palaeontological Museum Oslo, x 13.

fragmented stipes on same bedding plane believed to belong to this species have
a profile width of 0-8 mm across thecal apertures and 1 1 thecae in 10 mm. Bithecae
are apparently absent in all specimens. The details of the sicula are somewhat un-
certain. Specimen GSC 27098 may represent the sicula of this species (Text-fig. 2a)
in which case it is typically dendroid, consisting of a parallel-sided tube with thl*
originating near apex presumably in the prosicula.

710

PALAEONTOLOGY, VOLUME 16

Mode of branching. The precise way in which stipes undergo dichotomous division
is best illustrated at points 1, 2, and 3 of text-fig. 1a. Text-figs. 1 b, c are enlarged
sketches of these branching points and the process is perhaps most clearly shown in
dichotomy 2. The third theca on this second-order stipe which is here labelled ‘n’
(analogous to Jaanusson’s ‘accessory theca’ in Goniograptus) is slightly fatter than
the preceding theca and in mid-length curves to the right through about 40°. About
one-third of the way along the convex side of theca n arises theca («+!)*’ which
immediately diverges from theca n to form an angle of about 60° with it. Theca
(n-\~2f is derived from the base of theca (n + 1)^ and in nearly all cases the junction
is marked by a constriction (see text-fig. 1b). Theca (n+l)^’ also gives rise to (n + 2)”
and is therefore dicalycal. Recognition of dicalycal thecae elsewhere on the rhabdo-
some is facilitated by the fact that theca (n + 2)’^ lies considerably closer to the point
of dichotomy than does theca {n^3y and also by the tendency for the (n-{- 1)^ bear-
ing stipe to diverge at a slightly greater angle from the parent stipe. The paired stipes
arising from theca n are of unequal length due to their thecal composition. For
example the left stipe which develops from theca (/?+!)'’ is composed of (/?T 1)^ +
(n + 2)*’ + (n + 3)^ + (n + 4)*’ + ^ (n“) = 4^ thecae, whereas the right stipe is composed
of (n + 2)^ + (A7 + 3)‘^ + (n + 4)=^+^ (rd) = 3^ thecae (excluding the metathecal portion
of theca n).

The position of the dicalycal theca in each dichotomy can be on either the left or
right side of thecae n, id, etc. The dispositions of these dicalycal thecae on indivi-
dual specimens are plotted in text-fig. 3.

ABC

TEXT-FIG. 3a-c. Dispositions of dicalycal thecae in specimens of Clonograptm aureus sp. nov. a-c are
GSC 27096, 27097, and 27099 respectively. S = sicula, L — left-handed, R ^ right-handed.

Although GSC 27096 indicated a highly organized pattern in the direction the
dicalycal thecae were distributed in each of the four quadrants of the colony it does
not apparently hold true for GSC 27099.

Comparisons. The mode of branching observed by the writer in Loganograptus
kjerulfi Herrmann from the Arenigian at Galgeberg, Norway (see text-fig. 2b) is
of identical type. However, the pattern of the positioning of the dicalycal thecae in
PMO Paratype 73123 can be seen to be left-handed for second-order stipes and left-
handed for one-third-order stipe. This is comparable to GSC 27097 but not to
GSC 27096 or 27099.

The described mode of dichotomous branching also closely resembles that sug-
gested by Jaanusson (1965) for Goniograptus sp. from Jamtland. However, two

JACKSON: BRANCHING OF CLONOGRAPTUS

71 1

important differences in stipe anatomy exist. Firstly, in Goniograptus the positions
of the dicalycal thecae are alternately right-handed and left-handed along each main
stipe whereas in Clonograptus the pattern is not regular. A second difference is that
whereas in Clonograptus aureus sp. nov. the prothecal segment of {n-\-2)'"^ (see text-
fig. 1b) is abnormally long by comparison to (u+ 1)*’ the relationship in Goniograptus
sp. is reversed.

In summary, the manner of stipe dichotomy in Clonograptus, Loganograptus, and
Goniograptus and perhaps many other multiramous genera can be compared with
proximal end development in dichograptids. When theca n is made analogous to
theca F then the development is of isograptid type in which the dicalycal theca
(u+l)*^ = thF and theca n, like thF, forms from the first theca of the other stipe.
Among the three subtypes of development discussed by Bulman (1955), Didymo-
graptus rninutus with its single crossing canal seems to afford the best comparison.

Acknowledgements. I gratefully acknowledge the facilities provided by Dr. G. Henningsmoen during his
visit to the Palaeontological Museum, Oslo, in 1971. Also I am indebted to Professor O. M. B. Bulman
for helpful discussions and comments on the draft of the manuscript. The work was supported by a travel
grant from the Open University. Specimens described in this paper have been deposited with the Geological
Survey of Canada, Ottawa.

REFERENCES

BULMAN, o. M. B. 1955. Treatise on Invertebrate Palaeontology. Part V, Graptolithina. Univ. Kansas Press.
JAANUSSON, V. 1965. Two multiramous graptoloids from the Lower Didymograptus Shale of Scandinavia.
Geol. For. Stockh. Fork. 86, 413-432.

JACKSON, D. E. and LENZ, A. c. 1962. Zonation of Ordovician and Silurian Graptolites of Northern Yukon,
Canada. Bull. Amer. Ass. Petrol. Geol. 46, 30-45.

DENNIS E. JACKSON
Department of Earth Sciences
The Open University
Bletchley

Revised typescript received 2 January 1973 Bucks.

■V

^•v

VV'

<1 ii

, h

'. ^ ‘ ' '.’H Vie- -■ ' ' *

^ tAiiJii'.iwiJi.' ' • frit' . •“ I

* ■'■'>■ , ■•i ; win ' <•

■i'- . r (• ^•>- „■,. --fv

^ W|^ '.1^

• ^ i -u.-

-> .yii'/, ■

'•'1

.-.^ ■ -1

.« '

*/i~ ,., ■

i ,vn^i '

■ti. ■

# ■>• -"TV *

■in*j&if*W'*'yifiH-'

■li.r \i\

NOTES ON OPEN NOMENCLATURE
AND ON SYNONYMY LISTS

by S. C. MATTHEWS

Abstract. Rudolf Richter’s proposals on practice in open nomenclature and on annotated synonymy lists are
described and briefly criticized.

Plus quam leges valent boni mores

Tacitus, quoted by Richter in 1930

Authors of palaeontological papers can discover a great deal of instruction in
the International Code of Zoological Nomenclature (Stoll and others 1961). But the
Code sets a limit on its provisions ; it does not intend in any way to impinge on the
individual taxonomist’s exercise of his judgement (see the ‘Preamble’ to the Code).
It is therefore necessary to seek elsewhere for guidance on matters such as open
nomenclature (a device whereby an author expresses his judgement of his own
material) and synonymy lists (the means by which an author concisely expresses his
judgements of earlier opinions on the taxonomic problem he is handling). A highly
explicit set of recommendations on these matters is available in Rudolf Richter’s
(1948) ‘Einfiihrung in die Zoologische Nomenklatur’. His proposals have never been
as well known as they deserve among English-speaking palaeontologists. A French
translation, it may be noted, is available as Traduction no. 1448, prepared by Departe-
ment Documentation du B.R.G.M., B.P. 6009, 45 — Orleans — 02, France. This
present article, intended to bring Richter’s views to a wider public, draws freely on
what Richter wrote in 1948 (especially pp. 45-56). It is not a direct translation (some
things are deleted, and there are certain interpolations) and it makes no claim to
carry anything of the authority of Richter’s original. It is offered in order to promote
discussion of two points of nomenclatural technique, in the hope that authors might
become more familiar with the nature of certain devices they commonly employ, and
hoping too that greater consistency of practice might emerge. Richter’s proposals
are recommended by the fact that they have to a large extent become standard in the
German palaeontological literature— this article may have some incidental useful-
ness in explaining the meaning of a system of annotation which regularly appears in
papers published in the major German palaeontological journals, but whose signi-
ficance is not widely understood in other countries.

Those who are entirely unfamiliar with Richter’s work may find it useful to read
Stubblefield (1957).

OPEN NOMENCLATURE

Richter introduced his discussion of open nomenclature by considering the
problem of dealing with a specimen whose identity cannot exactly be determined.
If it is too hastily referred to a known species or genus, a previously clear taxonomic

[[’alaeontology, Vol. 16, Part 4, 197.4, pp. 713-719.]

E

714

PALAEONTOLOGY, VOLUME 16

concept may be diminished. If one refuses altogether to identify it, potentially useful
information may be left unemployed. If one decides to propose a new species or
genus to contain the specimen (a lesser error, in Richter’s view, than would arise by
referring the specimen to a previously well-established taxon— Richter, epigrammatic
here as elsewhere in his writings, observed that though spoiling the work, this pro-
cedure would not damage the tool, the standard form) a feeble name might result,
and proposals of feeble new names should not be encouraged.

Open nomenclature was developed as a remedy against such weaknesses of the
taxonomic method. It operates by attaching to known species or genera those
specimens whose identity is uncertain. The method offers a clear expression of the
fact of uncertainty, and also some indication of the degree of uncertainty involved.
This is not a matter of abdicating taxonomic responsibilities. It is instead an especially
perspicacious form of nomenclature that is involved. In contrast to closed nomen-
clature, with its firmly established and strictly defined names, it remains open to
whatever possibility of improvement future findings might bring. By giving taxonomy
a means of stretching (in an entirely honest and proper way) the limits of existing
knowledge, it by itself indicates where improvements are needed and in which
direction they might be sought. It permits us to build any such improvements into
nomenclature left open for that purpose, and this without any upset of established
names.

The signs. The signs employed in open nomenclature are in essence nomenclatural
and make up an integral part of the name. It should therefore be understood that they
are fundamentally different from signs attached to synonymy lists (see below).

1 . Signs for uncertainty at family or higher level.

The highest category touched by uncertainty, and above which certainty begins,
has the designation ‘incerta’. Examples:

Incertae familiae:
Incerti subordinis:
Incerti ordinis:
Incertae sedis :

Family uncertain

Suborder uncertain (order known)
Order uncertain (class known)
Class uncertain

The requirements at other levels are handled in the same way: Incertae subfamiliae;
Incertae superfamiliae.

2. Signs for uncertainty at genus or subgenus level.

If the attribution to an established genus is uncertain a ? is placed behind the name
of the genus. Examples:

Agenusl album Anton (? Anton)

Agenusl album (Anton) (?Anton)

Agenusl album (Anton) (?Bruno)

In the first case it was Anton himself who, at the time of the establishment of the
species album, attributed it with a question mark to the genus Agenus; for behind the
species-name, the name of the author is not in parentheses, and this, according to
ICZN Article 1 1 , signifies that the original generic assignment has remained unaltered.
In the second and third cases Anton had originally assigned the species without

MATTHEWS: OPEN NOMENCLATURE

715

question to the genus Agenus, and it was in later publications that the assignment
came into doubt. In the second case it was Anton himself who expressed this doubt,
and so added to his authorship of the name album also the authorship of the open
nomenclature. In the third case it was Bruno who was responsible, and he is the
author of the open nomenclature. Because in these latter two cases the generic assign-
ment is no longer unequivocally the one proposed by Anton when he established the
species, the name of the author of the species-name appears in parentheses (Article 11).

Uncertainty surrounding the subgenus can be dealt with in a corresponding way.
Example :

Agenus (Agenusl) album (Anton) (?Bruno)

3. Signs for uncertainty at species or subspecies level.

(i) When attribution to an established species is possible, but cannot be thought
certain, a ? is placed behind the name of the author of the species. In a subsequent
citation, the author of the open nomenclature, with the sign he introduced (here : ?),
is added in parentheses. Examples:

Agenus album Anton? (?Anton)

Agenus album Anton? (?Bruno)

In the first case, Anton himself had assigned a specimen, with some question, to
his species album, and in the second case it was Bruno who did this. There are no
circumstances in which it would be correct to place a ? between a species-name and
the name of its author. These two names (species + author) make up a nomenclatural
entity, which nothing should be allowed to divide.

Year of publication is also relevant, e.g. in the case where a certain author has at
different times made distinctions between forms but has given them all the same name,
assigning each of them, with a ?, to some particular species. These forms, which
could of course eventually prove to be specifically different from one another and
must be cited individually, may be distinguished for one another by the year.

The corresponding treatment can be given where it is attachment not to the species
but to a subspecies that is to be shown to be uncertain. Example:

Agenus album striatum Caesar? (?Bruno)

It also happens (especially in ornithology and herpetology) that the subspecies can
be firmly fixed even although the attribution to a species remains under question (in
most cases lateral replacement of species is involved). Examples:

Agenus albuml striatum Caesar (?Caesar)

Agenus albuml striatum Caesar (?Bruno)

(ii) If instead of attribution to an established species, only a possibility of com-
parison with that species should be indicated, cf. (abbreviation of the Latin word
confer) is placed in front of the species-name. Authors’ names, repetition of the sign
introduced (here: cf.), year, and distinction between several forms can be inserted
as in (i). Example:

Agenus cf. album Anton (cf. Bruno)

716

PALAEONTOLOGY, VOLUME 16

The corresponding treatment may be given to a subspecies where this is the subject
of a comparison only, as compared with the species attribution, which is firm.
Example :

Agenus album cf. striatum Caesar (cf. Bruno)

(iii) If a specimen shows itself to represent a new species, whose formal establish-
ment is, however, not yet justifiable, one can, in the interim, associate it with some
related, known species, before whose name n. sp., aff. (abbreviation of nova species,
affinis) will be inserted. Authors’ names, restatement of the sign inserted (here;
n. sp., aff.), and distinction of several forms can be introduced as in (i). Example;
Agenus n. sp., aff. album Anton (n. sp., aflf. Bruno).

Association with a known subspecies can be done in the same way. Example ;
Agenus album n. subsp., aff. striatum Caesar (n. subsp., aff. Bruno).

If any author has several different forms to compare with a known species in this
way, it is useful practice to identify each of them by a lower-case letter. These letters
have an advantage over names in that they introduce no question of priority and so
impose no burden. They can be used, with just as much exactitude as resides in a
species-name, for temporary characterization of a particular form. They are placed,
together with n. sp., behind the genus-name, or in the case of subspecies, together
with n. subsp., behind the species-name. Examples;

Agenus n. sp. a, aff. album Anton (n. sp. a, aff. Bruno)

Agenus n. sp. b, aff. album (n. sp. b, aff. Bruno)

Agenus album n. subsp. a, aff. striatum Caesar (n. subsp. a, aff. Bruno)

If the form is to be treated as a new species, not yet to be defined, and incapable of
being associated with any established species, then one writes simply n. sp., or if
several forms are involved, n. sp. a and n. sp. b (there are, quite possibly, many people
to whom ‘open nomenclature’ means no more than this particular provision).

(iv) If, again, the question of a relationship with some established species is unclear,
and yet an indication of the possibility of such a relationship is desirable (on taxo-
nomic, geographic, or stratigraphic grounds) then n. sp. aff.? is placed in front of the
name of the established species. Example;

Agenus n. sp. aff.? album Anton (n. sp. aff. ?Bruno)

An unclear relationship with a subspecies can be treated likewise. Example;

Agenus album n. subsp., aff. ? striatum Caesar (n. subsp., aff. ?Bruno)

(v) If the form might equally well belong to a known as to a new species, sp. (or
more fully, sp. inc., or sp. indet.— abbreviations of; species incerta, indeterminabilis)

[Editorial note; it is the practice in Palaeontology to put the noun before the adjective, e.g. sp. nov., subsp.
nov.]

MATTHEWS: OPEN NOMENCLATURE 717

is placed behind the genus-name. The corresponding practice at subspecies level is to
place subsp. inc. behind the species name. Examples :

Agenus sp. Anton, or Agenus sp. inc. Anton
Agenus album Anton subsp. inc. Bruno

4. Signs for uncertainty of both genus and species.

If both the genus and the species are uncertain, the appropriate signs all appear.
Examples :

Agenusl cf. album Anton (cf. Bruno)

Agenusl album Anton? (?spec. Bruno)

Agenusl n. sp., alf. album Anton (n. sp., aff. Bruno)

Agenusl n. sp., aff. ? album Anton (n. sp., aff. ?Bruno)

Agenusl sp. inc. Bruno

SYNONYMY LISTS

Non-nomenclatural signs. The signs attached to entries in a faunal list or synonymy
list are not integral parts of the zoological names and have nothing to do with formal
nomenclature. They stand to the left of the name and belong neither to the name
itself nor to the author of the name. Nor do they belong among the signs that stand
within or to the right of the name. They express the judgements of the author of
the list.

Signs attached to the synonymy list. The author of a synonymy list uses his signs to
make qualifying comments on these cases he cites in his list as synonyma of the species
whose name appears at the head of the list.

Anyone who wishes to carry forward a piece of research must check the existing
information. He needs the whole literature on the subject. In order to assist such an
inquirer one must strive to make the synonymy lists as near complete as possible, and
yet at the same time try to find means of making them as serviceable and as readable
as possible. By such means one can rid one’s text of pointless information and dis-
cussion. The following signs, first proposed in 1924, have subsequently had much use.

1 . Signs which should obviate needless searches.

1881 Year in italics : this work has a mention of the species, but without description
or illustration. It may be ignored by anyone who wishes to check merely the
morphological information, rather than the total data arising out of the
occurrence.

1881 Year in roman: the work contributes to our knowledge of the species. If such
a reference includes an illustration, it may help a later inquirer to give an
indication of the anatomical parts figured, e.g. in arthropods S = illustration
of the whole carapace, ^ = cephalon, = abdomen.

[cop. Anton 1856] : the illustration is not a new one, merely a repetition of one
already produced by Anton in 1856. Someone who is familiar with the figure
in the earlier work need not feel obliged to examine the repeat.

718

PALAEONTOLOGY, VOLUME 16

2. Signs which indicate the degree of confidence with which particular items in the

list are referred to the species under discussion :

*1881 * in front of the year : with publication of this work the species can be regarded

as valid under the terms of Article 1 1 of the ICZN (earlier mentions of this
name are to be regarded as nomina nuda).

.1881 .in front of the year: we accept responsibility for attaching this reference to
the species under discussion.

1881 No sign in front of the year: we have no right expressly to accept responsi-
bility for attaching this reference to the species under discussion; but at the
same time we have no cause to doubt such an allocation.

71881 ? in front of the year : the allocation of this reference must be subject to some

doubt because of the way in which it was presented (e.g. if the species-name
concerned included at that time several forms now treated as separate
species).

vl881 V in front of the year: vidimus! We have checked the deposited specimens
that relate to the work cited, and on their evidence we have chosen the addi-
tional sign used. v*1881: we have seen the type of the species, v.1881:
because of the evidence of the deposited specimens we are able to take
responsibility for this assignment, or vl881 : we do not accept responsibility.
v?1881 : the condition of the original specimens is such that no clear decision
is possible.

(1881) year in parentheses: the year of publication is uncertain.

An example of a synonymy list :

Agenus album Anton, 1900.

Agenus viride Aulus. — Bruno, Monogr. Agenidae, S. 12 Taf. 3 Fig. 2.

Agenus nigrum Anna. — Berta, Bibl. Ind., S. 20.

Agenus album n. sp. — Anton, Fauna Bras., S. 35 Taf. 2 Figs. 1-4 ^ ^ .

Begenus cf. cinereum Aulus. — Caesar, Reiseber., S. 10 [vix S. 12],

Agenus album Anton. — Anton, Neue Beob., S. 25, Tabelle 4.

Agenus? album Anton (?Bruno). — David, Orientierung, S. 30 Taf. 9 Fig. 3 ^ [kop.
Anton 1900 Fig. 1].

Agenus caeruleum n. sp. — Emil, Ubersicht, S. 6 Taf. 5, Fig. 2 w . [non Fig. 1 = Agenus
fuscum Felix.]

A critical synonymy list like this one, Richter observed, may be in itself a piece of
scientific work, approaching the state of a detailed monograph in its critical complete-
ness as well as in the range of information on which it draws. A compilatory synonymy
list, which brings together every mention of a zoological name available in the litera-
ture, but involves no expense of effort on deposited material nor any exercise of
Judgement, is not a scientific work, although it could usefully supply raw material for
one. A transcriptive synonymy list, which does nothing other than repeat earlier lists,
serves merely to put a surface gloss, easily penetrated, on something that is quasi-
scientific, and in effect no more than a waste of paper. Once a dependable list has been
published, it is sufficient at a later date simply to make a reference to it, adding any
necessary supplementary material. The supplementary entries will then serve toward
preparation of a revised synonymy list.

V . 1895
? 1907

V* 1900
V 1902
1908

V . 1910

V ? 1914

MATTHEWS: OPEN NOMENCLATURE

719

Concluding comment. It is now twenty-five years since the second edition of Richter’s
‘Einfiihrung’ appeared. Much of what he suggested then has become firmly estab-
lished in Germany. One or two of his proposals have faltered. His practice of citing
the name of the author of a piece of open nomenclature, for example, is not often seen
in the literature. His method of recording anatomical parts figured in references (see
‘synonymy lists’ above) could be applied in a simple way in the case of arthropods,
but it would be difficult to contrive similar schemes for other groups— it would
certainly be difficult to think of anything that could conveniently be rendered in type.
The annotations he proposed for synonymy lists have otherwise proved acceptable
and practicable. Rabien (1954) has proposed two additional signs. These are

(?)1881 (?) before the year: it is probable that the reference applies to the species

under discussion, but this cannot be established with certainty (e.g. in a
case in which the originals could not be checked, and the illustrations and
descriptions were insufficient to justify firm identification, yet in which the
identification could be considered probable for reasons stated in detail in
remarks which would follow).

p.l881 p before the year: partim: the reference applies only in part to the species
under discussion. If attached to any of the other signs of the synonymy list,
p would indicate that the sign applies only up to a certain limit to the work
cited. Example: vp before the year: the deposited specimens have been
checked, and some only of them belong to the species under discussion.

Struve (1966, p. 125) too has made some supplementary proposals. These have not
achieved any great currency, but they may, like Rabien’s, be found useful in cases
where it is necessary to make more specific comment than is provided for in Richter’s
own system.

Acknowledgements. The author is grateful to Professor Dr. D. Meischner (Gottingen), Professor Dr.
K. Krommelbein (Kiel), and Professor S. Simpson (Exeter) for their comments on a draft of the paper.
P. M. Sadler has kindly joined in discussions of many of the points involved.

REFERENCES

RABIEN, A. 1954. Zur Taxionomie und Chronologic der oberdevonischer Ostracoden. Abh. hess. Landesamt.
Bodenforsch. 9, 268 pp.

RICHTER, R. 1948. Einfiihrung in die Zoologische Nomenklatur. Kramer, Frankfurt a. M., 252 pp. (2nd
edition).

STOLL, N. and others (eds.). 1961. International Code of Zoological Nomenclature. International Trust for
Zoological Nomenclature, London, 176 pp.

STRUVE, w. 1 966. Beitrage zur Kenntnis devonischer Brachiopoden, 1 5 : Einige Atrypinae aus dem Silurium
und Devon. Senckenberg. leth. 47, 123-163.

STUBBLEFIELD, c. J. 1957. Profcssor Rudolf Richter. Bull. zool. Nom. 13, 139-141, 1 pi.

S. C. MATTHEWS
Department of Geology
University of Bristol
Queen’s Building
University Walk
Bristol, BS8 ITR

Manuscript received 14 September 1972

. ’m

> '

ft

f'ipi»M 'io'4.-j/?T<*' i’#'-

■ ' _— . \ ,*■ ' * r> .i.

"f ^ •? ‘M : . rttif

1. '■ V V, :/•,,# yj . ,J ,^. ■.,. ; (.iyrr.k'* V, W'j^'v.fi'i,’,), 4^5 1

i

M t-

■ vi-»‘» " ’" ■ ■

■ \ V ^-i)v ■ ' . ■ f .

= „ ' v'. .:' , , . •

•.'» V'>

,^. .'‘ ' i^--

; I I ' j'

■ '"■'* ' >■ ’ ‘ 'v;.- I'j.*-

r:J-|

■■ . WS !r»:

'. >. "
'■H

VittCj.

.' ’ "jS 'fKU.ii n'fdhii^j‘1 hi !«<.•

i - V ‘I'i '■' ' H-=ki i ‘>

<■•," MTUl--’

•i. :,I>\ ■ i?',* ( oil *

!. ?'f.l» -'V;. ,

J<\. ' ■'^i

■ ^ ■" ^

' r^W^4riii-<

:f.; .,, '-W<- - ■

:, “'O. ■'tv.'',

'V-. ■( , t

■ ;■ <n.i' -;''''• .

..! !'. - •. 'i'

l/-t it ■■ ' ;-I- t’VI 'vLy.' - '■

,, III

»

1 ' ■

i -i<:

■ ir *ii

»■ ,. /

'■■ '-Ifi" ^ ■// '.'

X¥'->»‘

• • ij"/

■ ;■■■:'» t- 1 ri/,

i-' '•

WYLEYIA: A NEW BIRD HUMERUS FROM THE
EOWER CRETACEOUS OF ENGLAND

by c. j. o. HARRISON and c. a. walker

Abstract. An imperfect humerus found in the Weald Clay (Lower Cretaceous) of Henfield, Sussex, represents the
earliest British bird known, Wyleyia valdensis gen. et sp. nov. It is compared with other reptile and bird humeri
from the Jurassic and Cretaceous.

In July 1964 Mr. J. F. Wyley found an imperfect humerus in the Weald Clay (Lower
Cretaceous) at Henfield, Sussex. It is reminiscent of certain advanced archosaur
humeri; but, having compared it with the humeri of pterosaurs (from which it is
completely distinct) and of small dinosaurs such as Hypsilophodon Huxley 1869 and
Deinonychus Ostrom 1969, we are confident that it represents a bird. The particular
bird-like characters which it shows are as follows :

1. There is a transverse ligamental furrow on the palmar surface just below the
articulating area of the proximal end.

2. The deltoid crest is blade-like, large, and very thin.

3. There is some indication of a capital groove on the anconal surface.

4. The shaft is slender, hollow, and thin-walled.

5. There is a distinct groove (brachial depression) for the insertion of the brachialis
anticus muscle.

The only other Lower Cretaceous bird remains known are Gallornis straeleni
Lambrecht 1931 from France (assigned by Brodkorb in 1963 to the Phoenicopteridae),
the gaviiform Enaliornis Seeley 1876 from the Cambridge Greensand of England,
and indeterminate feathers from Australia.

SYSTEMATIC DESCRIPTION
Wyleyia valdensis gen. et sp. nov.

Plate 89; text-figs. 1-2

Holotype. Unique specimen in British Museum (Nat. Hist.) No. A 3658: an imperfect right humerus.

Diagnosis. Humerus similar in character to both Archaeopteryx von Mayer and
Ichthyornis Marsh, but smaller. Proximal end of humerus relatively smooth and
flattened. Deltoid crest extending for nearly one-third of length of humerus apparently
making only slight angle with plane of the proximal palmar surface ; palmar surface
of crest itself with small longitudinal ridge on distal half and a very small curved
depression at distal end; lower profile of crest forming angle of about 45° with
external profile of shaft. No evidence of prominent bicipital surface. Bicipital crest
large, relatively thick, and rounded along internal edge. Ligamental furrow present
just below head on palmar surface. Shaft curved with some torsion (may be due to

[Palaeontology, Vol. 16, Part 4, 1973, pp. 721-728, pi. 89.]

E

TEXT-FIG. I. Wyleyia valdenis gen. et sp. nov. Holotype: right humerus (BMNH A 3658). x2. A, palmar
aspect; b, anconal aspect; c, internal aspect; d, external aspect; e, proximal aspect.

crushing), widening distally. Impression for brachialis anticus muscle nearer external
side of shaft, fairly deep, broad (half as wide as shaft), and elongated.

Description. The specimen is a damaged right humerus, with the distal end missing
altogether. The proximal end also is eroded and the outer part of the deltoid crest
and the proximal part of the bicipital crest have been broken away. The head itself
is damaged, but appears to have been prominent and rounded; there is some indica-
tion of a capital groove. Just below the head, on the palmar surface, there is a broad,
shallow, transverse depression which is probably the remains of the ligamental
furrow. The palmar surface is relatively smooth and slightly concave; the anaconal
surface is rounded, especially towards the internal side, but tapers down on the

EXPLANATION OF PLATE 89

Wyleyia valdemis gen. et sp. nov. Holotype: right humerus (BMNH A 3658). Stereopairs, x2. a, palmar
aspect; b, anconal espect; c, internal aspect; d, external aspect; e, proximal aspect.

PLATE 89

HARRISON and WALKER, Wyleyia

724

PALAEONTOLOGY, VOLUME 16

external side into the thin flange of the deltoid crest. The deltoid crest is long, about
a third of the probable total length of the bone, and distally curves in rather abruptly
to the shaft; its outer edge is broken, but on the palmar surface a small longitudinal
ridge is apparent distally, running near and parallel to the edge. There is also a very
small curved depression at the distal end of the crest, just next to the shaft. On the
internal side the bone curves out evenly to form a bicipital crest, but lacks any
prominent bicipital surface. The bicipital crest is thick and is curved on the anconal
surface with no suggestion of a narrow flange indicated in the illustrations of the
humeri of Ichthyornis. There is a small ridge where a median crest would be expected
to occur, but no suggestion of a pneumatic fossa or foramen.

The centre of the shaft is slightly crushed, but allowing for this, its anconal profile
is still slightly convex in lateral view; the shaft begins to widen towards the distal
break. Also visible in this area, near the external side of the shaft, is the scar of the
brachialis anticus muscle. It is fairly deep, broad (half as wide as the shaft), and
elongated; its edges are relatively steep and the impression tapers proximally, appear-
ing to terminate in a smaller and slightly deeper pit. There is some torsion in the
shaft, so that the two ends of the bone were inclined to each other at a considerable
angle (text-fig. 2c); some of it might be due to crushing, but it was in any case much
greater than in a modern bird (text-fig. 3c).

A B

TEXT-FIG. 2. Wyleyia valdensis gen. et sp. nov. Holo-
type: right humerus, a, reconstruction of proximal
aspect; B, reconstruction of palmar aspect, xlf;
c, diagram showing approximate angle between
proximal (1) and distal (2) articulations. Abbrevia-
tions used in this and other text-figures : as, anconal ;
be, bicipital crest ; brdp, brachial depression ; dc, del-
toid crest ; dpc, deltopectoral crest ; h, humeral head ;
it, internal tuberosity; If, ligamental furrow.

C

Measurements (in millimetres);

Total length as preserved 42-4

Length of deltoid crest 17-4

Length of bicipital crest 10-3

Width of shaft just distal to deltoid crest (internal-external) 4-5

Width of shaft just distal to deltoid crest (palmar-anconal) 3-3

Greatest width preserved at distal end 5-3

HARRISON AND WALKER; NEW BIRD HUMERUS

725

A B

TEXT-FIG. 4. lOniitlwcheirus clifti. Proximal part of
left humerus (BMNH 2353). a, proximal aspect;
B, palmar aspect, xf. Abbreviations: see legend to
text-fig. 2.

TEXT-FIG. 3. Cathartes aura (Turkey Vulture).
Left humerus, a, proximal aspect; b, palmar
aspect, X approx. 3; c, diagram showing
approximate angle between proximal (1) and
distal (2) articulations. Abbreviations: see
legend to text-fig. 2.

TEXT-FIG. 5. Hypsilophodon foxii. Right humerus
(BMNH R 196). a, proximal aspect; B, palmar
aspect, X f ; c, diagram showing approximate angle
between proximal (1) and distal (2) articulations.
Abbreviations: see legend to text-fig. 2.

726

PALAEONTOLOGY, VOLUME 16

Occurrence. Weald Clay, Lower Cretaceous, Henfield, Sussex. The ostracods in the
Henfield deposits indicate that they lie about 500-600 feet below the top of the
Weald Clay. This would establish the horizon as being in the middle of the Reeves
Group II, in the Barremian Substage of the Neocomian Stage.

COMPARISONS

Because of the early Cretaceous age of Wyleyia it was felt necessary to compare it
with humeri of certain archosaurs and early birds.

lOrnithocheirus clifti (Mantell). The humerus compared (BMNH 2353 and 2353a;
text-fig. 4) is from the Wealden of Sussex. It resembles a bird humerus in having a thin
wall and a hollow interior, but differs considerably in its general anatomy. The shaft
is shorter and more massive with a large distal articulation. The articular head is
fundamentally different in being saddle-shaped and concave in lateral view. The
deltopectoral crest (deltoid crest in birds) also differs in being a thick projection rather
than a thin blade, projecting at a much more acute angle to the plane of the
proximal end.

Hypsilophodon foxii Hulke. The humerus of this Wealden ornithopod from the Isle
of Wight belonged to a small specimen (BMNH R 196; text-fig. 5). It differs from
Wyleyia in several characters. Seen from above, the articulating area of the humeral
head is more massive and triangular and not crescentic in shape. The bone lacks the
transverse ligamental furrow on the palmar surface, the capital groove on the anconal
surface, and the well-marked brachial depression at the distal end. The shaft is rather
short and stout and has undergone a considerable amount of torsion, rather more
than in Wyleyia, so that the two ends of the humerus are almost at right angles to
each other. The deltopectoral crest is not blade-like but thick and blunt. Although
there is some torsion present in Wyleyia which is certainly more than is found in
modern birds, it is much less than is found in the majority of archosaurs. Some of the
torsion present in the shaft of Wyleyia could be due to crushing.

Deinonychus antirrhopus Ostrom. Comparison was made with a cast of a humerus
(AMNH 3015; text-fig. 6) of this Lower Cretaceous theropod from Montana. Like
Hypsilophodon, it lacks the ligamental furrow of Wyleyia and the well-marked
brachial depression; indeed, the brachialis anticus muscle has left no indication of its
presence. Further, the deltopectoral crest lies in a different plane.

Archaeopteryx lithographica von Meyer. Comparison was made with the holotype
(BMNH 37001; text-fig. 7) from the Upper Jurassic of Bavaria. Like Wyleyia,
Archaeopteryx lacks most of the distinct prominences and ridges generally found in
modern flying forms. The whole bone is slender and shows about the same amount
of torsion in the shaft. The proximal end bears a simple head, slightly swollen and
laterally elongated with a rounded articulation surface. The deltoid crest, however,
diverges less abruptly from the shaft and is set almost at right angles to the palmar
surface.

B

HARRISON AND WALKER: NEW BIRD HUMERUS

727

B

TEXT-FIG. 7. Archaeopteryx lithographica. Right
humerus (BMNH 37001). a, proximal aspect;
B, palmar aspect, x 1 ; c, diagram showing
approximate angle between proximal (1) and
distal (2) articulations. Abbreviations: see
legend to text-fig. 2.

TEXT-FIG. 6. Deinonychus antirrhopus. Right
humerus (AMNH 3015). A, proximal aspect;
B, palmar aspect, x^; c, diagram showing
approximate angle between proximal ( I ) and
distal (2) articulations. Abbreviations: see
legend to text-fig. 2.

A

C

TEXT-FIG. 8. Ichthyornis dispar. Right humerus (after ?

Marsh, 1880). a, proximal aspect; b, palmar aspect, i —

X 1^; c, diagram showing approximate angle between 2

proximal ( 1 ) and distal (2) articulations. Abbreviations :
see legend to text-fig. 2.

be

a s

728

PALAEONTOLOGY, VOLUME 16

Hesperornis Marsh and Enaliornis Seeley. These flightless birds had very reduced
wings. The humerus of Hesperornis, from the Upper Cretaceous of Kansas, is long
and slender with few prominent features ; the humerus of Enaliornis (Lower Cretaceous
of England) is unknown.

Ichthyornis Marsh. Plates in Marsh (1880; text-fig. 8) and a specimen in the BMNH
A 905 were used for this comparison, although humeri of I. victor Marsh were dis-
torted by crushing. The humeri of this Upper Cretaceous bird from Kansas resemble
Wyleyia in the proportions of the various parts (including the length of the deltoid
crest), the large deltoid crest projecting in approximately the same plane as the palmar
surface, and the relatively flat nature of the proximal portion. They differ, however, in
that (a) the bicipital crest does not project so far and is reduced in thickness, (b) the
impression of the branchialis anticus muscle is shallow proximally and lies nearer
the internal side of the shaft than the external.

Comparison with other Mesozoic material and Recent birds suggests that the new
humerus is also that of a bird, showing the greatest resemblances to Archaeopteryx
and Ichthyornis.

The fact that the humeri of Archaeopteryx, Ichthyornis, and Wyleyia are similar
would not, however, seem sufficient reason to place Wyleyia in either the Archaeo-
pterygiformes or Ichthyornithiformes, which would imply an affinity with the corre-
sponding Jurassic or Cretaceous forms. It therefore seems advisable to consider the
new genus incertae sedis until further evidence of affinity is forthcoming.

Acknowledgements. We wish to thank Mr. J. F. Wyley for making this specimen available for description;
Dr. A. J. Charig (British Museum (Natural History)) for criticizing the manuscript; Miss M. L. Holloway
for the line drawings; and Messrs. T. Parmenter and C. Keates for taking the photographs.
Abbreviations. BMNH— British Museum (Natural History); AMNH — American Museum of Natural
History.

BRODKORB, p. 1963. Catalogue of fossil birds. Bull. Fla. St. Mas. biol. Sci. 7, 179-273.

HUXLEY, T. H. 1870. On Hypsilophodon foxii, a new dinosaurian from the Wealden of the Isle of Wight.
Q. Jl. geol. Soc. Lond. 26, 3-12, pis. 1-3.

MARSH, o. c. 1880. Odontornithes: Monographs on the extinct toothed birds of North America. United
States Geological Exploration of the Fortieth Parallel, 7, xv-201, pis. 1-34.

MANTELL, G. A. 1844. Medals of creation. London.

MEYER, H. VON 1861. Archaeopteryx lithographica (Vogel-Feder) und Pterodactylus von Solenhofen. Neues
Jb. Miner. Geol. Paldont. 678-679.

OSTROM, J. H. 1969. Osteology of Deinonychus antirrhopus, an unusual theropod from the Lower Cretaceous
of Montana. Bull. Peabody Mus. nat. Hist. 30, 1-165, figs. 1-83.

SEELEY, H. G. 1876. On the British fossil Cretaceous birds. Q. Jl. geol. Soc. Lond. 32, 496-512, 2 pis.
LAMBRECHT, K. 1931. Gallornis straeleni n. g. n. sp., ein Kreidevogel aus Frankreich. Bull. Mus. r. Hist,
nat. Belg. 1 (30), 1-6, figs. 1-3.

CONCLUSIONS

REFERENCES

C. J. O. HARRISON

C. A. WALKER

Sub-Department of Ornithology Department of Palaeontology

Typescript received 13 November 1972

British Museum (N.H.)
Tring, Herts.

British Museum (N.H.)
London, S.W. 7

A PROBLEMATICAL DINOFLAGELLATE FROM
THE TERTIARY OF VIRGINIA AND
MARYLAND

by DEWEY M. MCLEAN

Abstract. Inversidinium exilimurum, gen. et sp. nov., a problematical dinoflagellate recovered from the Aquia
Formation (Upper Paleocene) of the Virginia-Maryland Coastal Plain, displays an atypical peridinioid outline
characterized by a truncated antapex which ruptures to form a hitherto unreported type of antapical excystment
apparatus; antapical archeopyles are exceedingly rare among the dinoflagellates.

In the Cretaceous and Tertiary dinoflagellate and acritarch assemblages of the
Virginia-Maryland Coastal Plain are problematical palynomorphs of which the
morphological interpretation and taxonomic treatment cannot be unequivocally
assessed. A cyst recovered from the marine Aquia Formation (Upper Paleocene),
displays characteristics which suggest affinities with the Pyrrhophyta ; however, inter-
pretation of these characters is subjective. Its outline compares most closely with that
of a peridinioid but is atypical by having a truncated antapex which lacks antapical
horns; the excystment apparatus (= archeopyle?) does not conform in shape or
position with any thecal plate, or combination of plates, known to the author; and
features of the periphragm which seemingly correspond in position to a cingulum and
sulcus are reflected not by indentations, but by convex-outward folds. Orientation of
the cyst is subjective. By comparison with a general peridinioid outline, the pointed
tip of the periblast is considered apical and the truncated end showing the excystment
feature, antapical. In addition, convex-outward folds of the periphragm which corre-
spond in position to a cingulum and sulcus seemingly reflect dorsal and ventral
surfaces, respectively. Following this orientation, the cyst is interpreted to have a
hitherto unreported type of antapical excystment apparatus. Antapical archeopyles
are exceedingly rare among dinoflagellates.

All samples are from the Aquia Formation (Upper Paleocene) of the Virginia-
Maryland Coastal Plain from outcrop localities along the Potomac River south of
Washington, D.C. They are;

Locality 1. Prince Georges County, Maryland. Reference: U.S. Geol. Svy. Anacostia, Md.-D.C. quad.,
7-5 minute series; geog. coords. 38° 45' 10" N. Lat., 76° 59' 15" W. Long. Approx. 45 feet (14 m) of lower-
most Aquia glauconitic quartz sands are exposed about 0-5 mile (0-80 km) west of Friendly, Maryland,
along the stream occurring immediately south of, and paralleling, the Old Fort Road.

Locality 2. Stafford County, Virginia. Reference: U.S. Geol. Svy. Passapatanzy, Va.-Md., quad.,
7-5 minute series; geog. coords. 38° 22' 15" N. Lat., 77° 17' 50" W. Long. This is the type locality of the
Aquia Formation. Approximately 70 feet (21 m) of Aquia glauconitic quartz sands are exposed in the
bluffs along the south bank of Aquia Creek, about 0-5 mile south-east of the Maryland-Virginia Monu-
ment No. 37.

Locality 3. Stafford County, Virginia. Reference; U.S. Geol. Svy. Passapatanzy, Va.-Md., quad.,
7-5 minute series; geog. coords. 38° 20' 35" N. Lat., Long, 77° 17' 17" W. Long. Approximately 35 feet
( 10 m) of Aquia glauconitic quartz sands are exposed in the bluffs along the south bank of Potomac Creek,
from 0-5 and 0T5 mile west of the Maryland-Virginia Monument No. 35.

[Palaeontology, Vol. 16. Part 4, 197.4, pp. 729-7.42, pi. 90.]

F

730

PALAEONTOLOGY, VOLUME 16

Standard acid maceration techniques were utilized for all samples. Palynomorphs were concentrated by
use of ZnBr (sp. gr. = 2 0), and were darkened for study and photomicrography by acetolysis. Slides are
stored at Stanford University and are assigned Stanford University Paleontological Type Collection
( = SUPTC) numbers. Coordinates are measurements in millimetres to the right (R) or left (L) and toward
the top ( + ) or bottom (— ) of the slide from an index cross engraved on the coverslip near its lower left
corner.

SYSTEMATIC PALAEONTOLOGY
Division pyrrhophyta Pascher
Class DiNOPHYCEAE Paschcr
Genus inversidinium nov.

Derivation of name. Latin, inversus, inverted, with reference to the antapically oriented archeopyle.

Type species. Inversidinium exilimurum sp. nov.

Diagnosis. Bi-layered cyst; periblast outline peridinioid with pointed apex and
truncated antapex; lacks indications of tabulation; endoblast outline variable.
Convex-outward folds of the periblast reflect cingulum and sulcus. Excystment
apparatus (= archeopyle?) antapical; forms by rupture of anatapical tips of peri-
blast and endoblast. Periphragm and endophragm externally smooth to granulose.

Remarks. Inversidinium exilimurum, gen. et sp. nov., is assigned to the Pyrrhophyta
on the basis of its generalized peridinioid shape, excystment apparatus, and features
of the periphragm that seemingly reflect a cingulum and sulcus.

Inversidinium exilimurum sp. nov.

Plate 90

Derivation of name. Latin, exilis, thin or meagre; mums, wall-in reference to the thin, transparent nature
of the periphragm and endophragm.

Holotype. PI. 90, figs. 1-2. Loc. 2, sample 3384, SUPTC 10079 (R26 0, +12-5).

EXPLANATION OF PLATE 90

Inversidinium exilimurum, gen. et sp. nov. ; all specimens are from the Aquia Lormation (Upper Paleocene)

of the Virginia-Maryland Coastal Plain.

Pigs. 1-2. Holotype, x920. 1, Shows details of ventral longitudinal ridge; dorsal transverse ridge is
shown in optical section on left side of specimen. 2, Shows general peridinioid outline of periblast and
triangular outline of endoblast; note truncated antapex of periblast and complete absence of antapical
horns. Dimensions: LxW = 50x40 /xm. Loc. 2, sample 3384, SUPTC 10079 (R26 0, -I-12-5).

Pigs. 3, 6. Two focus levels of one specimen oriented with antapex facing the observer and ventral surface
facing up. 3, Transverse optical section at level of dorsal transverse ridge, x 1340. The bulge facing up
in the photograph shows in cross section the ventral longitudinal ridge. 6, Antapical tip of periblast,
X 1710. Maximum width of specimen 35 /xm. Loc. 1, sample 3370, SUPTC 10080 (R14-2, +4-3).

Pigs. 4-5. Two views of one specimen, x925. 4, Shows optical section of triangular endoblast; note
ruptured antapex of both periblast and endoblast. 5, Phase contrast, shows details of ruptured antapex
of both periblast and endoblast. Dimensions: LxW .= 54 x 45 /xm. Loc. 2, sample 3390, SUPTC 10081
(R190, fl20).

Pigs. 7-9. Several focus levels of one specimen oriented with antapex facing observer and ventral surface
facing up, x2000. 7, Transverse optical section at level of dorsal transverse ridge showing optical
section of ventral longitudinal ridge. 8, Pocused on extreme antapical tip of periblast. 9, Antapical view
of periblast focused slightly deeper into specimen than in fig. 8, showing what is interpreted to be a
circular excystment aperture ( archeopyle?). Maximum width of specimen 40 /xm. Loc. 2, sample
3392, SUPTC 10082 (R24-3, + 1-5).

PLATE 90

McLEAN, Inversidinium

732

PALAEONTOLOGY, VOLUME 16

Diagnosis. As for genus.

Description. Periphragm transverse section elliptical at widest part and triangular at
antapex. Endoblast dorso-ventral outline roundly triangular with elliptical transverse
section at widest part and triangular cross section at antapex; endoblast occupies
hypotractal portion of periblast. A narrow convex-outward fold 1-3 (xm wide traverses
periblast dorsal surface at its widest part ; a similar fold extends longitudinally along
the periblast ventral midline from the antapex to about midway between the widest
part of the periblast and the apex. Excystment apparatus (= archeopyle?) forms by
posterior tip of endophragm breaking away and posterior tip of periphragm ruptur-
ing irregularly. Periphragm and endophragm less than 1 ^m thick, transparent, and
externally smooth to granulose.

Dimensions. Holotype L X W = 50 x 40 /^m. Observed range (eleven specimens measured) : length 45-60 fj.m
(mean 51 /xm); width 32-52 ;u,m (mean 41 ju.m).

Remarks. The narrow convex-outward folds on the dorsal and ventral surfaces of the
periblast are commonly folded secondarily such that the resultant deformed folds
have the appearance of furrows. Rotation of such specimens into both lateral and
polar views, so that the features could be examined in cross section, confirmed that
they are convex-outward folds that had been folded secondarily and are not furrows.
Rupture of the periphragm along the crests of the folds has not been observed.

Comparison with similar species. Inversidinium exilimurum, gen. et sp. nov., is unique
among the Pyrrhophyta with the possible exception of Wetzeliella {Rhomb odiniuml)
minuscula Alberti, 1961, which resembles the new species in appearance. They differ
in details concerning the antapex and shape of the endoblast. According to Alberti’s
description, W. {R7) minuscula exhibits two antapical protrusions separated from
one another by means of a longitudinal split; I. exilimurum, on the other hand, has
a sharply truncated antapex that lacks any vestige of protrusions (horns) and has an
outward-convex fold along its ventral midline instead of a split. The endoblast of
W. (R?) minuscula is rhombohedral in outline and nearly fills the periblast whereas
that of /. exilimurum is triangular in dorso-ventral outline and occupies the hypo-
tractal portion of the periblast.

W. {Rhombodiniuml) minuscula lacks the intercalary type archeopyle typical of
Wetzeliella, thereby raising questions concerning its assignment to Wetzeliella.
Should later investigations show it to have an antapical excystment apparatus, it
should then be transferred to Inversidinium.

Occurrence. Loc. 1, less than 1% of the phytoplankton content through all but the upper 20 feet of the
section; Loc. 2, less than 1% of the phytoplankton content throughout the section; Loc. 3, less than 1%
of the phytoplankton content of one sample taken 20 feet above the base of the section.

Acknowledgements. It is a pleasure to express my thanks to Dr. W. R. Evitt of Stanford University who
offered many helpful suggestions during this study which was made possible by National Science Founda-
tion Grants GB 4702 and GB 471 1 (to Evitt).

REFERENCES

ALBERTI, G. 1961. Zur Kenntnis mesozoischer and alttertiarer Dinoflagellaten und Hystrichosphaerideen
von Nord- und Mitteldeutschland sowie einigen anderen europiiischen Gebieten. Palaeontographica,
116 (A), 1-58, 12 pis.

DEWEY M. MCLEAN

Department of Geological Sciences
Virginia Polytechnic Institute
Blacksburg Virginia, 24061 U.S.A.

Typescript received 14 December 1972

THE APPLICATION OF ELECTRON
MICROSCOPY TO PALAEONTOLOGY

by M. D. MUIR and w. l. diver

The Palaeontological Association and British Micropalaeontological Group were
joint sponsors of the Symposium ‘The Application of Electron Microscopy to
Palaeontology’ held on 11 September 1972, as a part of the 5th European Electron
Microscope Congress, EMCON 72. The following four papers form part of the
proceedings of the Symposium and illustrate how both the transmission electron
microscope (TEM) and the scanning electron microscope (SEM) can be used in a
wide range of palaeontological applications to give information on a variety of
specimens that would be unobtainable by any other means. Most speakers emphasized
that the reliable interpretation of electron micrographs requires supporting evidence
from a thorough light optical microscope investigation.

In the general discussion that followed the meeting, it became clear that combined
studies using all available techniques on a limited amount of material offered much
greater returns in terms of understanding than large numbers of ‘pretty pictures’
produced using only one technique. Several novel preparation techniques were
presented, all of them giving significantly better results than conventional methods.
The short discussions printed here illustrate some of the interesting points raised at
the meeting.

It was evident at the meeting that applications of electron microscopy (both TEM
and SEM) will increase rapidly in all branches of palaeontology as access to equip-
ment becomes possible for more and more palaeontologists. It is to be hoped that all
new palaeontological electron microscopists will show such care in specimen pre-
paration, instrument operation, and interpretation of results as was evident in the
papers presented by the contributors to the EMCON 72 Symposium.

Department of Geology
Royal School of Mines
Imperial College
Prince Consort Road
London, SW7 2BP

[Palaeontology, Vol. 16, Part 4, 197,t, p. 733.]

"MS'

«■.. ■

■ -.. ■ V"'"„. ^'' y'::n%-i.v

Ha;,

^ -

'• n\ I -

. , V ■ : '',■ h' ify

’ I ■' K ■’

’-■ . VI X

W«- '•:•/ .j: ^'ib

j' 'W-Jlf I

f(.1 ‘♦‘Jjiif^ gi . •••{.4>m'. 'j

■ ■ , ’ t- y

, ; ^ ^'v /■. yH^arx '

, •' ..♦ t \w. .M

■ ■■■ ‘

• ' : 'V '

.?r»i |XI» n'fkiWi

J - •■ ^' t '■■•■*'•• 't ■ . -n

if 'w':'

■ V .

jf ■ -M- ^

MORPHOLOGY AND EVOLUTION OF THE
EYE IN UPPER CAMBRIAN OLENIDAE

(TRILOBITA)

by E. N. K. CLARKSON

Abstract. The eyes of selected olenid species from Scandinavian concretionary shales were examined with the
scanning electron microscope. Though these eyes are small, many previously unknown details were visible, including
the ‘peripheral zone’ of Olenus wahlenhergi and other genera. Reconstructions, prepared by camera-lucida techniques,
show the eye and the whole cephalon of certain species.

In early olenids the visual surface was dehiscent in the adults and is preserved only in meraspids; in later genera
the ocular suture became fused and the visual surface was retained. Details of lens distribution and manner of emplace-
ment are described in Peltura minor, P. scarabaeoides, and Ctenopyge (Mesoctenopyge) similis. Evolutionary changes
in the structure and shape of the eye are clear in different lines of descent. Some of the observed modifications are
thought to be due to paedomorphosis.

Some comments on the mode of life of olenids are also given.

THE EYES OF CAMBRIAN TRILOBITES

At the end of the Cambrian there was a major crisis in the history of the trilobites.
Most of the rather undifferentiated Upper Cambrian stocks became extinct and were
replaced, first by a number of short-lived Tremadoc groups, and then soon after-
wards by several very distinct suborders which came to dominate the Ordovician
trilobite fauna (Whittington 1966).

This late Cambrian crisis had far-reaching effects on the evolution of trilobites.
Certain morphological features which had remained rather conservative during
Cambrian times became much more diversified and novel kinds of functional
organization came into being. Amongst the characters affected was the visual system,
and the new trilobite stocks of the early Ordovician evolved eyes exhibiting greater
variety than those possessed by their Cambrian forebears. Not only did the primitive
holochroal organization, which was already established in the earliest Cambrian
trilobites, become modified in many different ways, but there first appeared an
entirely new kind of visual organ, the schizochroal eye (which may have been derived
from a holochroal ancestral pattern by paedomorphosis according to Clarkson
(1971). This kind of eye is probably confined to the suborder Phacopina, which per-
sisted from Arenig to Famennian times.

Though some of the many different kinds of eye in Ordovician and later trilobites
have been quite extensively studied there is at present so little information on the eyes
of Cambrian genera that we do not have a comprehensive picture of the evolution of
the eye in trilobites. One good reason for this is that the eyes in adult specimens of
Cambrian trilobites are not very often preserved, though in a few cases intact lenti-
ferous surfaces have been reported. Thus Walcott (1910) noted lenses in the eyes of
meraspids of the Lower Cambrian Olenellus gilbert i Meek and Opik (1961, p. 57)
and later Jell (1970, p. 306) and Jago (1972, p. 230) discussed the presence of lenti-
ferous surfaces in Cambrian pagetiids, where the eye has a ‘schizochroal’ appearance.

[Palaeontology, Vol. 16, Part 4, 1973, pp. 735-763, pis. 91-95.]

736

PALAEONTOLOGY, VOLUME 16

I am not aware that any visual surfaces are known to be preserved in Middle
Cambrian trilobites, but amongst the Upper Cambrian fauna certain genera with
intact eyes occur sporadically, and different kinds of eyes may be preserved within
particular groups such as the family Olenidae, which are the subject of the present
study. Lindstrom (1901, p. 29, pi. Ill) in his monograph on trilobite eyes figured the
visual surfaces of the olenid genera Peltura, Sphaerophthalmus, and Ctenopyge,
illustrating highly magnified lenses, thin sections, and the librigenae in position on the
cranidium. As a matter of historical interest, he regarded the olenids as the oldest
known ‘oculate genera’, and did not think that earlier trilobites had functional eyes.
Upper Cambrian trilobite eyes were also described by Opik (1967) in his monograph
of the Mindyallan fauna of Queensland, where, amongst others, the large and well-
preserved eyes of Blountia mindycrusta Opik were illustrated.

Opik (1967, p. 54) discussed the question of the preservation of the eye in Cambrian
trilobites very thoroughly. Noting that the visual surface is rarely preserved, he sug-
gested that in life the visual surface had been bounded by a peripheral circumocular
suture, and that during ecdysis or after death the whole lentiferous area would fall out
and not be preserved. This suture comprised the palpebral suture and the ocular suture
(text-fig. 2a), which ran along the upper and lower borders of the visual surface
respectively, meeting at the front and rear. He pointed out that in post-Cambrian
trilobites, the lower part of the circumocular suture or ocular suture became fused,
so that during ecdysis, the visual surface separated only along its contact with the
palpebral lobe. The visual surface adhered to the librigena and thus stood a much
higher chance of preservation. Some of the Upper Cambrian trilobites also had non-
functional ocular sutures and, as Opik pointed out, fusion of the visual surface with
the librigena became reasonably common in Upper Cambrian times in unrelated
groups, and is not a character of phylogenetic significance. The only alternative possi-
bility is that in many Cambrian trilobites the visual surface was so delicate that its
preservation in any case would be unlikely; but then one would expect there to be
a ragged edge to the eye-socle, to which the visual surface was attached, and this is
not so. I therefore agree entirely with Opik’s suggestions, adding that there is some
evidence of the ocular suture having been functional only in adult trilobites. The
meraspids of O. gilbert i Meek described by Walcott (text-fig. la) and sometimes very
small holaspids of Ordovician Flexicalymene species from the Waynesville forma-
tion, Ohio, have intact visual surfaces with visible lenses. Adult individuals of these
species, however, never have visual surfaces preserved, although in mature Flexi-
calymene the lower surface of the palpebral lobe, along the line of the palpebral
suture, may be denticulate as if elongate prisms or ienses’ had originally rested
there.

In the course of the present study, I found some small but complete visual surfaces
in meraspids of Olenus wahlenbergi Westergard. The retention of the visual surface
in Parabolina and other derivatives of Olenus may be seen as an example of paedo-
morphic development, in which the ocular suture, which had been functional only
in the adult, was even there dispensed with. The role of paedomorphosis in the evolu-
tion of olenid and other eyes is discussed later.

CLARKSON: OLENID EYES

737

TEXT-FIG. 1. a. Olenellus gilberti Meek. Meraspid figured by Walcott (1910, pi. 36, fig. 4c; pi. 43, fig. 5, 6)
in oblique lateral view with the individual lenses visible. Where the visual surface has been broken away
at X, impressions of the lenses are left on the underlying matrix. Lower Cambrian, Ptarmigan Pass, Alberta.
Smithsonian Institution Catalogue number 56828L

b. Olenus wahlenbergi. Lateral view of cephalon reconstructed, showing the ‘eye-indices’ of Struve
(1958). A = length of visual surface. H = Distance from posterior edge of eye to posterior border furrow.
G = preglabellar to occipital furrow. Cn = preglabellar furrow to rear edge of occipital ring.

THE OLENIDS

Because so many Cambrian trilobites had functional ocular sutures, it is unlikely
that we shall ever obtain a good record of the evolution of the most ancient trilobite
eyes. Studies of the detailed morphology of the eyes of single species can, however,
contribute towards this end, and when the phylogeny of Cambrian trilobites becomes
better known these may be seen more clearly in an evolutionary perspective. In
addition, it is fortunate that there is one family at least, the Olenidae, where the
phylogeny is well known and in which material for study is so well preserved and
abundant that at least some features of the evolutionary history of the visual system
within this family can be elucidated.

The Olenidae are a geographically widespread family, which arose early in Upper
Cambrian times and abounded to the end of the. Tremadoc. A few genera persisted
into the Ordovician, and Triarthrus until the close of the Middle Ordovician. Olenid
faunas are best known in Scandinavia where they have been the subject of many
studies culminating in the major monographs of Westergard (1922) and Hennings-
moen (1957); they are common throughout the Acado-Baltic province and in the
Tremadoc of South America (Harrington and Leanza 1957), and there are isolated
occurrences elsewhere.

In the alum shales of the Oslo region and the old quarries of Andrarum, in Scania,
there occur stinkstone concretions with vast numbers of disarticulated olenid frag-
ments, frequently with very fine structure preserved, and in full relief with no trace
of flattening. In the early genera, Olenus, Leptoplastus, Eurycare, and others, the
lenses are preserved only in small meraspids. The visual surface in adults is unknown,
but even so there remains, at least in the best-preserved adults of Olenus, a wealth of
interesting detail on the palpebral lobe and the lower rim or eye-socle, which sug-
gests that the whole region peripheral to the visual surface may have been a highly
sensory zone. Later genera, which include Sphaerophthalmus, Ctenopyge s.l., Peltura,

738

PALAEONTOLOGY, VOLUME 16

and Parabolina, retained the visual surface, often with excellent details of the lenses
and peripheral zones. Though no details of subsurface layers in the eye are preserved,
but only the lenses, it is hoped that the present study will be a useful contribution to
olenid morphology in general, and to the understanding of the evolution of trilobite
visual systems. Because the olenid faunas of Scandinavia are so well known I have
made extensive reference to Westergard (1922) and Henningsmoen (1957) in which
full accounts of morphology and complete synonymies are given. Following
Henningsmoen both proposed international and local Norwegian zones are given,
e.g. Olenus wahlenbergi occurs in Zone II (2a/3).

METHODS AND TECHNIQUES

Since all the material available to me consisted of disarticulated fragments the
work on olenid eyes fell naturally into two parts. The first task was to reconstruct the
cephalon with the cranidium and librigenae fitting together as they were originally
assembled in the living animal. This was to show the visual surface (where present)
in its original relationship to the palpebral lobe, and the eye in its true relationship
to the cephalon. The second phase of the work was the detailed study of the visual
surface and the bordering regions (palpebral lobe and eye-socle) with the scanning
electron microscope (SEM). With this information, certain inferences could be made
about the evolution of the eye in the family, though it was not possible to study all
the genera.

Technique of reconstruction. The reconstructions were made from cranidia and
librigenae, accurately drawn with a Wild-Heerbrugg microscope with an M5 draw-
ing tube or ‘camera-lucida’. For each species several undamaged or nearly complete
cranidia were drawn in dorsal, lateral, and frontal views; the plane of the palpebral
lobe being normally taken as horizontal. High magnification drawings were also
made in oblique lateral view. Where the specimen was slightly damaged appropriate
details could be filled in with reference to other cranidia.

Librigenae of equivalent size were also selected. Each was drawn in an orientation
where its camera-lucida image fitted the drawing of the reconstructed cranidium,
with the slope of the cheek region, and the edges of the librigena and fixigena match-
ing all the way along the suture. In the final drawing the dimensions of the parts of
the reconstructed cephalon were the same in all the different views.

Scanning electron microscopy. Both gold-palladium and aluminium coatings were
used; the latter were found to be equal to the former in conducting properties. Visual
surfaces, palpebral lobes, and eye-socles were all examined in different orientations,
to build up a complete picture of the eye. Where the lenses had become detached from
the eye in some areas, their total thickness was apparent. Unexpected features visible
with the SEM were the remarkable peripheral zones in the eye of Olenus wahlenbergi,
which are recorded in detail below.

Deposition of specimens. AH olenid specimens used in this study are in the collections of the following
institutes; Palaeontologisk Museum, Oslo (P.M.O.); Geology Department, University of Lund (LO);
British Museum (Natural History), London (BMNH); Grant Institute of Geology, Edinburgh (Gr. I.).

CLARKSON: OLENID EYES

739

OLENID EYE MORPHOLOGY
Subfamily oleninae

Olenus s.s. is the earliest olenid genus and seems to have been the rootstock of the
whole family. The eye of O. wahlenbergi, described below from exceptionally well-
preserved material, is representative of early olenids as a whole and eyes of this kind
were retained by various later genera which include Leptoplastus and Eurycare. Later
leptoplastines, however, had eyes of modified form.

In adults of Olenus the visual surface has not been found, but small meraspids
have been found with the visual surfaces still intact, so that some details of their
structure can be determined. In the adults, fine details of the palpebral lobes and eye-
socles remain, and these can be reconstructed in their original relationship.

Only two genera of the Oleninae, Olenus and Parabolina, have been studied.
Parabolina is probably a derivative of Olenus (see p. 744) in which the adult has
a visual surface of similar kind to that in the meraspids of Olenus. The palpebral lobe,
moreover, though inflated and of peculiar form, is confluent with the ocular ridge,
again as in immature specimens of Olenus. These two factors amongst others are
suggestive of a paedomorphic origin for the eye of Parabolina, a situation paralleled
by Peltura and more distantly by other olenids.

Olenus wahlenbergi Westergard, 1922

1922 Olenus Wahlenbergi n. sp; Westergard, p. 128, pi. IV, figs. 5-14.

1 957 Olenus wahlenbergi Westergard 1 922 ; Henningsmoen, p. 1 1 0, pi. 3 (with complete synonymy).

Plates 9 1 , figs. 1 -6 ; 92, figs. 1 -4 ; text-fig. 2 a-j

Material. Twenty-two blocks of topotype material from Andrarum, Zone II (2a^). Gr I 5514-5536.

Remarks. The gross morphology described by Westergard is supplemented by my reconstruction which
shows the librigenae in place (text-fig. 2, c-e). Both lateral and frontal views show how the genal spines
were in life directed horizontally and may be interpreted as props for supporting the cephalon on the sea
floor. The anterior arch (Clarkson 19666) is well developed, though it might have been partially blocked
by the hypostome.

Development. Though the ontogenetic development of O. wahlenbergi has not been documented in detail.
Strand’s (1927) description of ontogeny in O. gibbosus show a broadly comparable mode of development.
Many larval stages of O. wahlenbergi are present in the material which I studied, those figured in text-
fig. 1 g, h, and j being close in morphology to Strand’s stages 8 or 9 (length 0-70. —0-71 mm) and 1 1 (length
1 mm). Strand remarks upon the presence of continuous eye-ridges from the earliest stages, though the
severance of these from the palpebral lobe in later development was not noted. The equivalent stages to
Strand’s stages 8 and 9 are here referred to as meraspids, following Whitworth (1970).

Structure of the cuticle. At high magnifications (over x 1000), the external surface of the palpebral lobe,
eye-socle, and other parts of the exoskeleton can be resolved into raised polygons, all of the same general
size and of semi-regular form (PI. 91, fig. 3). Such polygons also underlie the ridges of the alimentary
prosopon. They are found only on the external surface and have no internal expression. They seem to be
similar to the ‘cell polygons’ of modem arthropod cuticles (Dalingwater, in press); each of the under-
lying cells which secretes the cuticle contributes a single ‘tile’ to the mosaic which forms the whole cuticle,
and its form is retained on the outside of the cuticle. Though cell-polygons are present in other olenids
they have not been found so clearly preserved as in O. wahlenbergi. Fractured cuticular surfaces show
radial structures though these have not been investigated further.

Eye-morphology : Meraspid eyes. The visual surface is present in meraspids where the length of the eye

CLARKSON; OLENID EYES

741

does not exceed 0-45 mm. In these the external surface of the lentiferous area is smooth, though exami-
nation of internal surfaces shows the undersides of the lenses, which are small and weakly convex. Pre-
servation of these small structures is not particularly good, and therefore details of their structure and
arrangement are indeterminate. The eye-socle is distinct from the visual surface, though at this stage in
development does not exhibit the vertical ridges of the adult (Plate 91, fig. 2).

In meraspids the palpebral lobe is at first very narrow, and is connected to the ocular ridge (text-fig. Ig).
Later it widens and eventually, when the eye has attained a length of more than 0-75 mm, it becomes
separate from the ocular ridge. The specimen illustrated in text-fig. 2j, in which the ornament of the fixigena
is still pustulose, shows the beginnings of separation of the palpebral lobe from the ocular ridge. In adult
specimens, the pustulose ornament is replaced by the ridges of the prosopon (Opik, 1961), and the ocular
ridge does not connect with the palpebral lobe, but is separated from it by a pronounced channel. It is
noteworthy that the visual surface in juveniles is first of all directed more anteriorly, and only later com-
mands a more lateral field of view.

Adult eyes. Sometime after the meraspid stages illustrated in text-fig. 2 h, j, the ocular suture must have
been effective, for there is never any trace of the visual surface thereafter. The visual surface must have
been reniform and of moderate height, though not spherical or globular. It was set opposite S2 with its
posterior edge set slightly further from the mid-line so that the long axis of the eye (line joining the anterior
and posterior edges) made an angle of about 10° with the exsagittal plane. This contrasts with the situation
in meraspids where the equivalent angle may be up to 45°. Eye-indices (Struve 1958): A/G 37%, A/Gn 30%,
H/A 108% (text-fig. \b). The palpebral lobe is reniform, separated from the fixigena by a distinct palpebral
furrow, depressed centrally, and rising anteriorly and posteriorly to low elevations (the rear elevation is the
more prominent). From these elevations the surface plunges very steeply and the palpebral lobe narrows
as it joins the eye-socle. The surface of the palpebral lobe is rather smooth but becomes highly ornamented
in the outer region near the facial suture. Two separate elements can be distinguished. The first kind of
structure (PI. 92, figs. 1 -4) consists of thin elongated ridges, nearly normal to the outer edge of the palpebral
lobe and especially prominent on its anterior and posterior elevations. On these raised areas the ridges
resemble alimentary prospon and bifurcate as they approach the edge. In the outer central part of the lobe,
which lies between the two elevations, the ridges are less prominent and anastomose, forming an area of
irregular polygons (PI. 92, fig. 2), again confined to the outer part of the lobe.

Secondly, there are a number of peculiar swellings, usually elongate, situated along, or sometimes be-
tween, some of the ridges. They sometimes show a well-developed crystalline structure (PI. 92, fig. 2), but
are otherwise of indeterminate morphology. These alone have some similarity of appearance to the much
more highly developed corrugated surface of the eye-socle, and might have had a similar function. They
might have been the sites of glands or sensory organs. In addition, the surface of the palpebral lobe, like
that of the rest of the cuticle, has many round pits, possibly the openings of perpendicular canals in the
cuticle (Dalingwater, in press).

The eye-socle is a prominent band, which could on superficial inspection be taken for the visual surface
itself. The true shape of the eye-socle was determined by excavating inverted librigenae, which retained their
upper edges within the rock matrix and were more likely to possess intact anterior and posterior edges than
specimens with the dorsal surface uppermost. From librigenae such as that figured in text-fig. 1l\ it was

TEXT-FIG. 2. Olenus wahlenbergi (Westergard 1922). Zone II, Andrarum, Scania.

a, b. Reconstructions of the eye region of a medium-sized adult in antero-lateral and dorsal views show-
ing the visual surface missing because of the functional ocular-suture, ‘s’ marks the position of peripheral
(possibly sensory) zones on the palpebral lobe and eye-socle. Mainly from Gr. I. 5521.

c. Part of specimen showing undamaged anterior horn of the eye-socle, lying ventral side uppermost and
excavated from above. Gr. I. 5526.

d, e,f. Reconstructions of complete cephalon in dorsal, frontal, and lateral views from Gr. I. 5521 and
5522.

g. Early meraspid, slightly damaged anteriorly, approximating Strand’s (1927) stages 8 or 9. Gr. I. 5523.

h. Librigena of meraspid still retaining the visual surface. Gr. I. 5524.

j. Cranidium of meraspid of about the same size, approximating Strand’s stage II. Gr. I. 5525.

742

PALAEONTOLOGY, VOLUME 16

seen that these edges curved upwards sharply to meet the descending edges of the palpebral lobe, which are
slightly recessed where they meet the eye-socle. Text-fig. 2 {a, b), showing the reconstructed eye, was con-
structed from camera-lucida drawings of a perfectly preserved palpebral lobe, and a nearly perfect librigena
of a similar-sized specimen.

The vertical ridges on the eye-socle are confined to a median horizontal band, above which the socle
thins abruptly (PI. 91, figs. 4-6). These ridges are more or less vertical and parallel, but sometimes lie
obliquely and anastomose with neighbours. The ridges of the palpebral lobe and eye-socle, though not
really similar in appearance, form a continuous zone peripheral to the visual surface and may have been
the sites of accessory sensory organs; a concept discussed in more detail later on. Since their function is not
proved, it is convenient to refer to the whole complex of ridges and grooves as the ‘peripheral zone’, and
this term is used hereafter. Many other trilobites have a similar peripheral zone, sometimes in the form of
ridges and grooves, sometimes as tubercles, and sometimes as funnel-shaped pits. The existence of such
a zone in Olenus is the earliest recorded occurrence, and it is of interest that it should apparently be much
less well developed, at least in external expression, in later olenid derivatives.

Parabolina spinulosa (Wahlenberg 1821)

1821 Entomostracites spinulosus; Wahlenberg, p. 38, pi. 1, fig. 3.

1854 Parabolina spinulosa Wahl.; Angelin, p. 46, pi. XXV, fig. 9; pi. XXVII, fig. 3.

1922 Parabolina spinulosa (Wahlenberg); Westergard, p. 134, pi. VI, figs. 14-20.

1957 Parabolina spinulosa (Wahlenberg); Henningsmoen, p. 126, pi. 1, fig. 2; pi. 3, fig. 12 (with
complete synonymy).

Plate 92, figs. 5, 6; text-fig. 3 a, b

Material. Five blocks from Westergard’s collection. University of Lund, Andrarum. Zone iii {2b). LO
4527-31.

Remarks. The morphology of this species is very well known and I have not attempted complete restora-
tions, but only antero-lateral views primarily to show the eye and the alimentary prosopon, drawn with
a camera-lucida.

Eye-morphology. As no juveniles were available for examination this description is based upon adults.
The eye is small, and set relatively close to the anterior border. It lies obliquely, and the long axis (line
connecting the anterior and posterior edges) makes an angle of some 20° to the exsagittal plane. Eye-
indices; A/G 25%, A/Gn 19%, H/A 270%. By contrast with an adult Olenus, the ocular ridge contacts the
glabella and runs laterally and slightly backwards, expanding to become confluent with the swollen pal-
pebral lobe. The palpebral lobe which is defined by a shallow palpebral furrow, is smooth with no evidence
of a peripheral zone, nor is there any indication of such a zone on the (very narrow) eye-socle. The visual
surface is reniform and not strongly curved, so that it subtends a rather restricted field of view directed
antero-laterally. The external corneal surface is smooth and structureless (PI. 92, figs. 5, 6), and the lenses
below, which seem to be welded to the lower corneal surface, are poorly preserved, but their lower surfaces
are weakly convex as in the case of meraspids of Olenus. Ridges of the alimentary prospon radiate from
near the base of the eye, branching towards the cephalic border and anastomosing towards the rear of the
librigena.

EXPLANATION OF PLATE 91

Figs. 1-6. Olenus wahlenbergi (Westergfird 1922). Zone II. Andrarum, Scania. 1, Meraspid cephalon
showing confluence of palpebral lobe and ocular ridge. Gr. I. 5514, x90. 2, Meraspid. External mould
of visual surface with some parts of the cornea and underlying lenses still adherent. Gr. I. 5515, x 175.
3, External mould of the surface of an adult cephalon, with cell-polygons and tubercles. The full thick-
ness of the cuticle is seen on the left. Gr. I. 5516, x 220. 4, External mould of eye-socle (inverted) showing
faint striations of the sensory zone and undamaged upper rim. Gr. I. 5517, x65. 5, Outer surface of
eye-socle, with prominent striations. Upper rim damaged. Gr. I. 5518, xl20. 6, Same, showing
striations highly magnified, x 2400. Bar 5 (u.m.

PLATE 91

CLARKSON, Olenus eyes

744

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 3. Parabolina spinulosa (Wahlenberg 1821). Zone III, Andrarum, Scania.

a. Eye region reconstructed from cranidium and librigena with a near-perfect visual surface, b. Whole
cephalon restored. From LO 4259, 60.

Features of the eye of Parabolina which bear resemblance to the eyes of meraspids of Olenus suggest that
the Parabolina eye was derived from the eye of Olenus by paedomorphosis. These include: (i) The retention
of the visual surface and obsolescence of the ocular suture; (ii) Confluence of the palpebral lobe with the
ocular ridge; (iii) The anterior position of the eye and its high inclination to the exsagittal plane, and pos-
sibly (iv) the absence of surface features of the peripheral zone. Other features, such as the inflation of the
palpebral lobe, are not paedomorphic and have a separate origin.

Subfamily pelturinae

As far as is known, all Pelturinae have eyes of the same general kind. They are
small, placed far forward, shaped so as to cover only the anterior hemisphere of
vision, and are normally preserved with the visual surface intact. The palpebral lobe
is swollen and connected to the glabella by an unbroken ocular ridge, though this
may become indistinct near the glabella. In such material of Protopeltura as was avail-
able for study, the eye was not well preserved, and there is less certainty about ocular
morphology. Some of Westergard’s figures of various species of Protopeltura (1922,
Taf. XIV, figs. 4, 27), suggest that the visual surface is absent whereas others (Taf.
XIV, fig. 20; Taf. XV, fig. 1) seem to indicate its presence; but as he also figured
species of Peltura both with and without the visual surface, its absence in some

EXPLANATION OF PLATE 92

Figs. 1-4. Olenus wahlenbergU’WestergkTd \922). Zone II. Andrarum, Scania. 1 , Palpebral lobe, anterior
region with structures of sensory zone. Gr. I. 5519, x 190. 2, The same showing ‘prosopon’ and ‘sensory
nodes’, x 935. Bar = 5 /nm. 3, Palpebral lobe of another specimen, anterior region with structures of
‘sensory’ zone. Gr. I. 5520, x 130. 4, The same, magnified x500.

Figs. 5, 6. Parabolina spinulosa (Vl'dUenhcTg \S2\). Zone III. Andrarum, Scania. 5, Librigena with eye.
LO 4527, x46. 6, The same, magnified xl25.

PLATE 92

CLARKSON, Olenus and Parabolina eyes

746

PALAEONTOLOGY, VOLUME 16

specimens of both genera suggest breakage rather than the presence of an ocular
suture. Indeed, it seems fairly certain that Protopeltura had an eye similar to that of
Peltura.

Certain morphological characters of the eyes of Pelturinae can be interpreted, as
with Parabolina, as being paedomorphic in origin. Into this category fall the small
size, forward position and inclination of the long axis of the eye, the unbroken
ocular ridge, the retention of the visual surface in the adult, and the over-all similarity
of structure to the eyes of meraspids of Olenus. Henningsmoen (1957, p. 1 14) pointed
out the resemblance between the earliest Parabolina species, P. brevispina and
Protopeltura, suggesting that the two are closely related descendants of Olenus. The
similarity of eye structure in Peltura and Parabolina accords with this relationship;
presumably this kind of eye arose once only.

The two species discussed below were selected as having eyes representative of
Pelturinae, and both of them displayed excellent structural details showing the
arrangement of lenses on the visual surface.

Peltura scarabaeoides scarabaeoides (Wahlenberg 1821)

1821 Entomostracites scarabaeoides ', Wahlenberg, p. 41, pi. 1, fig. 2.

1854 Peltura scarabaeoides Wahl.; Angelin, p. 45, pi. XXV, fig. 8.

1922 Peltura scarabaeoides (Wahlenberg); Westergard, p. 173, pi. XV, figs. 12, 13, 18.

1957 Peltura scarabaeoides scarabaeoides (Wahlenberg 1821); Henningsmoen, p. 237, pi. 2,
fig. 1 ; pi. 6; pi. 25, figs. 6, 13, 14; pi. 26, figs. 1, 2.

1958 Peltura scarabaeoides (Wahlenberg 1821); Whittington, p. 200, pi. 38, figs. 1-18.

Text-fig. 4 a, c

Material. Three blocks from Slemmestad, Norway Zone Vc (2dy) P.M.O. 29268, 29270, 29272. Also
ontogenetic series BM It. 5516-9. One block from Andrarum, Scania associated with Ctenopyge lin-
narssoni, and Sphaerophthalmus humilis. Zone Vc Gr. I. 20803.

Remarks. Of all the cephalic reconstructions made to show the true position of the eye that presented here
for P. scarabaeoides is the most tentative. Though the restoration of the cranidium posed no problems, it is
very much more difficult to be certain as to how the cranidium and librigena fit together. Several camera-
lucida drawings were made of the cranidium in different orientations. Various librigenae were then succes-
sively examined under the camera lucida microscope so that the image of the librigena could be seen adjacent
to the previously drawn cranidium. Each librigena in turn was then manoeuvred into an orientation such
that the image of the whole cheek region showed a smooth unbroken curve; this was then taken as the most
lifelike construction. Such a reconstruction shows that the librigenae slope down quite steeply at about
45° and that there is a moderately well-developed anterior arch, though this was probably partially blocked
by the hypostoma.

Eye-morphology. The eye is small, placed far forward on the cephalon and quite near the glabella. Its long
axis is inclined at 45° to the exsagittal plane. Eye-indices: A/G 20%, A/Gn 15%, H/A 350%. Whittington
(1958) showed that the palpebral lobe is poorly defined, though present early in ontogeny. Thereafter it
becomes more distinct and the visual surface is present in the smallest known librigenae (ibid., PI. 38,
figs. 12, 13). The material of P. scarabaeoides is not particularly good, and did not photograph well, hence
the drawing (text-fig. 4c); lens arrangement in the early stages is better shown in P. minor. The palpebral
lobe of adult specimens is entirely smooth, swollen near the facial suture, and confluent with the short
ocular ridge which connects with the glabella, though in some specimens the ocular ridge becomes faint
and ill defined towards the glabella. The lenses themselves are not preserved, and the material is found as
internal and external moulds. External moulds show that the cornea must have been entirely smooth and
without any distinct structure, whereas impressions of the lower surfaces of the lenses appear distinctly on
internal moulds. Lindstrom (1901, p. 29, PI. Ill, figs. 35-42) illustrated the fine structure of the eye of

CLARKSON: OLENID EYES

747

TEXT-FIG. 4. a, Peltura scambaeoides scarabaeoides. (Wahlenberg 1821.) Zone Vc, Slemmestad. Norway.
Cephalon in oblique lateral view, restored from Gr. I. 20803.

b, Peltura minor (Brogger 1882). Diagram exhibiting the spatial relationships of the lenses. From P.M.O.
87558 (vide PI. 93, fig. 1). c. Peltura scarabaeoides scarabaeoides (Wahlenberg 1821). Right eye of a large
specimen drawn from photographs and camera-lucida. From BM It. 5519. d. Idealized hexagonal close-
packing system showing geometrical relationships between lens centres typical of peltiirines. Based on 4c.

P. scarabaeoides from material with the lenses preserved. He showed that the lenses are plano-convex with
a smooth upper surface. As Lindstrom saw no trace of an organic junction between the lenses and the cornea,
he regarded them, not as ‘free lenses’, but as inwardly bulging extensions of the cornea like those of Limulus.
My present study has given no evidence for or against this suggestion, neither in Peltura nor in the similar
Olenus meraspids, where the lenses and cornea apparently can only be detached together. It seems more
likely, considering that the olenids are a close-knit group, that the lenses are in fact free structures of plano-
convex form, closely welded to the lower surface of a very thin cornea, which did not show in Lindstrom’s

748

PALAEONTOLOGY, VOLUME 16

sections because of recrystallization. Nevertheless, the extraordinarily wide spacing of the upper lenses in
P. minor, and their curious distribution could accord with either hypothesis, and Lindstrom’s suggestion
should not be discounted.

Peltura minor (Brogger 1882)

1882 Cyclognathus costatus n. sp. var minor -, Brogger, p. 1 10, pi. II, figs. lO-l 1.

1922 Peltura minor (Brogger); Westergard, p. 175, pi. XV, figs. 3-1 1.

1957 Peltura minor (Brogger, 1882); Henningsmoen, p. 235, pi. 6, pi. 25, figs. 2-5.

Plate 93, figs. 1,2; text-fig. 4b

Material. One specimen from Gamlebyen, Oslo, associated with Sphaerophthalmus alatus. Zone Vb (2d|3).
P.M.O. 87558.

Eye-morphology. Only the visual surface is present, preserved as an internal mould. There appears to be
little difference in eye structure between this species and P. scarabaeoides. The lenses are represented by the
impressions of their lower surfaces. There is considerable variation in the spacing of these lenses; those
near the facial suture being very widely spaced, and almost certainly disjunct, whereas those near the lower
margin are somewhat smaller and much closer together, probably being contiguous.

Development oj the eye in pelturines. In young pelturines the lenses have an unusual
pattern of arrangement, unlike that of leptoplastines or indeed of other Cambrian
trilobites. There may have been a similar system in the Oleninae, but the preservation
is not good enough to determine this. This basic pattern is retained, though modified
by the addition of many more lenses in fully grown pelturines. The beautifully pre-
served eye of P. minor, figured in PI. 93, figs. 1, 2, serves as a model showing an early
stage of development ; young P. scarabaeoides eyes are very similar though less well
preserved.

Here the pattern is a form of hexagonal close packing, but the dorso-ventral files
radiate dorsally, diverging in a fan-like manner. The uppermost lenses are the most
widely spaced; they are also somewhat larger than the others. By analogy with other
trilobites these were presumably the first-formed lenses. This odd pattern, with the
files converging as they plunge downwards seems to be adapted to accommodate
more lenses in the lower central part of the visual field, whilst giving wide-ranging
though less intensive coverage elsewhere. The approximate maximum visual range
for this eye, which has fifty lenses, is 0° to 90° (long.) and —20° to 50° (lat.). Vision is
thus entirely confined to the anterior hemisphere, with the main clustering of lens

EXPLANATION OF PLATE 93

Figs. 1,2. /’e/mra (Brogger, 1882). Gamlebyen. Oslo. Zone Vb. (2dj8). 1, Internal mould of visual
surface (right edge = anterior). P.M.O. 87558, x 135 (vide text-fig. 4b). 2, Same, upper part of visual
surface, x 500. Bar = 10 /urn.

Figs. 3-6. C tenopyge {Mesoctenopyge) similis Wennmg^moen \9S1 . SarsGate. Oslo. Zone Vb (2d)3 sim.).

3, Palpebral lobe showing nearly structureless surface (top right = anterior). P.M.O. 87567, xl20.

4, Adult visual surface in lateral view showing dorso-ventral files and eye-socle with faint vertical stria-
tions. P.M.O. 87566, x85. 5, Oblique dorsal view of adult visual surface. P.M.O. 87565, x65.
6, Oblique dorsal view of visual surface of a young specimen (right edge = anterior), (see also text-fig. 5e),
(enlargement of left-hand specimen in PI. 94, fig. 5). P.M.O. 87564, x 120. (Figs. 5 and 6 are illuminated
from the south).

PLATE 93

CLARKSON, Peltura and Ctenopyge eyes

750

PALAEONTOLOGY, VOLUME 16

axes centred on an axis 45° from the sagittal plane, and directed downwards at
10-20° below the ‘equator’, towards the sea floor.

Though no large adults of P. minor were available for study, the fully developed
pelturine system of lens arrangement was seen in mature specimens of P. scara-
baeoides scarabaeoides. Here there are some 180 lenses, arranged in a pattern like
that of P. minor, though modified through growth. Most specimens have some
fortuitous irregularities, like those figured in text-fig. 4c, but apart from these the
lens centres are arranged in a regular geometric sequence, idealized in text-fig. Ad.
This is clearly a hexagonal close-packing scheme, but one in which the distances
between lens-centres decrease arithmetically towards the base of the eye. Three
intersecting component rows are evident, as follows: (a) arching latitudinal rows,
becoming closer together ventrally, {b) a set of files converging ventrally towards the
anterior ventral edge of the eye, vertical near the anterior edge and curving more and
more obliquely towards the posterior edge, (c) an identical set, vertical near the
posterior edge and curving towards the anterior.

The lenses are largest at the top and decrease in size ventrally and it is probable
that their growth is inhibited by the proximity of neighbouring lens-centres, as sug-
gested in my analysis of the eye of Ormathops (Clarkson 1971). One advantage of
having lenses graduated in size is that irregularities in distribution are avoided. When
the lenses are all the same size, as in Ormathops or Ctenopyge, irregularities are
inevitable. The eyes of Ctenopyge are similar to those of the pelturines in that the
dominant files, which are diagonal near the top of the eye, swing into a more nearly
vertical position towards the base but these do not converge in Ctenopyge, and
identical-sized lenses with inevitable irregularities in distribution result (see p. 737).
A full analysis of different systems of lens-packing in trilobites is beyond the scope
of this paper, but it is worth noting that the system exhibited by pelturines is found
also in certain post-Cambrian trilobite groups and is especially distinct in cyclo-
pygids, though here the decrease in spacing may be logarithmic.

Subfamily leptoplastinae

The eye of Leptoplastus stenotus Angelin was less well preserved than that of
O. wahlenbergi in the material studied. Basically, it is of the same general type,
though relatively smaller. The visual surface is absent, and none of the small mera-
spids were preserved showing the visual surface. No detailed structure was visible on
the palpebral lobe or the eye-socle, due to poorer preservation. Eurycare again has
an eye of similar type, with the visual surface missing.

Though the similarities between the eyes of early Leptoplastinae are clear, there
was a great change with the incoming of Ctenopyge and Sphaerophthamus. Not
only was the visual surface retained but there were substantial modifications in the
palpebral lobe and associated regions. The visual surface furthermore departed from
the primeval reniform shape and became larger, and elliptical or nearly spherical,
often projecting laterally from the head. Though the peripheral zone is less clearly
marked than in Olenus, ridges and grooves are still detectable in some cases on the
palpebral lobe and the eye-socle. Preservation of the eyes in the later Leptoplastinae
was good, though less perfect than in O. wahlenbergi. The granular structure seen
at high magnification implies at least some diagenetic alteration.

CLARKSON: OLENID EYES

751

Ctenopyge (Eoctenopyge) modesta Henningsmoen 1957

1922 Ctenopyge flagellifera angusta n. var. (partim); Westergard, p. 185, pi. XI, fig. 6-7.

1957 Ctenopyge (Eoctenopyge) modesta\ Henningsmoen, p. 191, pi. 5; pi. 19, figs. 1-10.

Plate 94, fig. 5; text-fig. 5 a-d

Material. Four blocks from Sars Gate, Oslo, associated with Ct. similis, Protopeltura bidentata, Parabolina
mobergi. Zone Vb (2dj8 sim.) P.M.O. nos. 87564-7.

Remarks. The reconstructions, made from camera-lucida drawings, show how the slender genal spines
emerge on the librigenae opposite the eye, and thence springing away at right angles to the cephalon curve

TEXT-FIG. 5. a-d. Ctenopyge (Eoctenopyge) modesta. Henningsmoen 1957.

u, b, c. Restoration of the cephalon in frontal, dorsal and lateral views, from P.M.O. 87564-7. d. Same,
enlarged, in antero-lateral view.

e. Ctenopyge (Mesoetenopyge) similis eye drawn from stereoscan photograph (vide P1. 93, fig. 6) showing
relationships of the lenses, the different zones, and the development of the dorso-ventral files. Arrows
represent the directions of the emergent dorso-ventral files. P.M.O. 87564.

752

PALAEONTOLOGY, VOLUME 16

backwards and downwards coming to lie in the same plane as the antero-lateral border of the cephalon.
Such a cephalon could be given support to rest upon the sea floor by these spines, and as with many other
trilobites (Clarkson \966b) the base of the eye would then be horizontal.

Eye-morphology. The eye is one-third the total length (sag) of the cephalon, and set opposite SI, high on
the cheek towards the rear. Eye-indices are A/G 44%, A/Gn 34%, H/A 36%. The palpebral lobe, relatively
large and defined by a pronounced palpebral furrow, rises outwards and in some specimens carries pro-
minent radial ridges, all the way round, and normal with the facial suture. A thin ocular ridge connects the
palpebral lobe to the anterior region of the glabella. The visual surface, of elliptical form, is set upon a
vertical eye-socle, about one-fifth the height of the whole eye. In the material to hand there are no vertical
ridges on the socle. Some specimens have an eye of symmetrical form : a regular oblate spheroid truncated
below by the upper edge of the eye-socle, in others the anterior part of the spheroid is depressed, and the
highest curvature is posterior. This may, however, be a preservational feature. The visual field commanded by
such an eye is panoramic, and the visual fields of the two eyes meet, though hardly overlap, in front, above
and behind. Laterally, the limit of vision is directed a few degrees below the equatorial or horizontal plane.

Ctenopyge (Mesoctenopyge) similis Henningsmoen 1957

1922 Ctenopyge erecta n. sp. (partim); Westergard, p. 156, pi. XI, figs. 26-21 .

1957 Ctenopyge {Mesoctenopyge) similis n. sp. ; Henningsmoen, p. 195, pi. 5; pi. 20, figs. 10-14.

Plate 93, figs. 3-6; text-figs. 5e, 6 a-c, e

Material. Four blocks from Sars Gate, Oslo, associated with E. modesta, Protopeltura bidentata, and
Parabolina mobergi. Zone Vb (2d|S sim.) P.M.O. nos. 87564-7.

Remarks. The most striking feature of the reconstructed cephalon is the pair of large genal spines, which
project forwards and curve round to the rear, terminating behind the body. Though such long spines have
sometimes been used in inferring a planktonic mode of life through frictional retardation of sinking, their
orientation, as the front and side views show, is much more suggestive of an adaptation for supporting not
just the cephalon, but also the whole body upon the sea floor. The flattening of these massive spines suggests
their functioning as a gigantic snowshoe giving support to the body when resting on a muddy sea-floor.

Eye-morphology. The eye is one-fifth the total length (sag.) of the cephalon, and set high on the cheek
opposite S2 with its anterior edge about midway between the anterior and posterior borders. Eye-indices:
A/G 33%, A/Gn 38%, H/A 125%. The palpebral lobe is relatively small and narrow, with distinct ridges,
though these are not so deeply impressed as in O. wahlenbergi or E. modesta. The visual surface is very
similar to that of E. modesta though the lenses are relatively smaller, and it commands a similar visual
field. Some specimens have faint vertical ridges on the eye-socle.

Ctenopyge (Mesoctenpyge) tumida Westergard 1922

1922 Ctenopyge tumida n. sp. (partim.); Westergard, p. 155, pi. XI, figs. 15-18.

1957 Ctenopyge {Mesoctenopyge) tumida Westergard 1922; Henningsmoen, p. 198, pi. 5; pi. 20,
fig. 16.

Plate 94. figs. 1^; text-fig. 6d

Material. Five blocks from Naersnes, R^yken, associated with Peltura acutidens. P.M.O. no. 87551-5.
Also two blocks from Sars Gate, Oslo of C. cf. tumida associated with C. angusta. Zone Vb (2d/S). P.M.O.
nos. 29751, 29757.

Remarks. Though the genal spines are less massive, less flattened, and jut out laterally rather than first
being directed anteriorly, they could still have been used for the support of the cephalon and body on the
sea floor. The morphology of this species does, however, seem to be less ideally adapted for that purpose
than does that of M. similis.

Eye-morphology. The eye is of almost identical morphology to that of M. similis, except in position. It is
set further forward and lower down, so that the posterior edge of the eye lies about midway between the
anterior and posterior borders of the cephalon, and the orientation of the ocular ridge and other structures
is correspondingly altered. Palpebral lobes lie opposite S2. Eye-indices: A/G 28%, A/Gn 35%, H/A 150%.

CLARKSON; OLENID EYES

753

TEXT-nG. 6a-c, e. Ctenopyge (Mesoctenopyge) similis Henningsmoen 1957. Zone Vb. Sars Gate, Oslo.
Restoration of the cephalon in dorsal, frontal, and lateral aspects, and (c) in enlarged antero-lateral view
from P.M.O. 87564-87567.

d. Ctenopyge (Mesoctenopyge) tumida. Westergard 1922. Zone Vb. Royken. Cephalon in antero-lateral
view restored from P.M.O. 87551-87555.

754

PALAEONTOLOGY, VOLUME 16

Sphaerophthalmus alatus (Boeck 1838)

1838 Trilobites alatus mh.; Boeck, p. 143.

1857 Sphaerophthalmus alatus Boeck; Kjerulf, p. 92.

1922 Sphaerophthalmus major Lake; Westerg&rd, p. 163, pi. XIII, figs. 9-19.

1957 Sphaerophthalmus alatus (Boeck); Henningsmoen, p. 212, pi. 2, fig. 15; pi. 5; pi. 22, figs.
18-26.

1968 Sphaerophthalmus alatus (Boeck): Rushton, p. 414.

Plate 95, figs. 1,2; text-fig. la-d

Material. Three blocks from Gamlebyen, Oslo, labelled S. major, and associated with Peltura minor. Zone
Vb (2d/3). P.M.O. 87556-8. Also two blocks from Andrarum (old collection) associated with Peltural
acutidens. Zone Vb (2d/3) Gr. I. 20775-6.

Remarks. Three-dimensional material shows that the genal spines of S. alatus (in standard orientation)
spring out laterally from the librigenae and are not bent downwards below the cephalon as in S. humilis.
Even in the best specimens studied, the tip of the genal spine was always broken, but can be restored from
the illustrations of previous authors, especially Henningsmoen and Rushton. The anterior arch is of
moderate height, and the postero-lateral border rises from the genal spine obliquely. Thus the cephalon
could rest upon the sea floor upon the antero-lateral border and the genal spines ; a position impossible for
the related S. humilis.

Eye-morphology. Eye one-quarter the total length of the cephalon, and set high on the cheek, opposite SI.
Eye-indices : A/G 35%, A/Gn 27%, H/A 108%. The palpebral lobe is relatively narrow with a curving outer
edge, and is confluent with the long narrow and backwardly curving ocular ridge. In side view it forms
a nearly semi-circular arch, with a slightly flattened top, where it is widest. The upper surface of the palpe-
bral lobe flares outwards and upwards from the deeply incised palpebral furrow at about 45°. There is
little trace of surface ornament apart from scattered indistinct tubercles, but the granular surface seen at
high magnifications implies that some recrystallization has taken place. The visual surface is closely similar
to that of typical representatives of Ctenopyge, and is a nearly perfect oblate spheroid in form with irregu-
larities in lens distribution typical of all later leptoplastines. The eye-socle is extremely narrow, more so
than in any species previously discussed, though faint vertical striations are visible.

Sphaerophthalmus humilis (Phillips 1848)

1848 Olenus humilis n.s.; Phillips, p. 55, figs. 4-5, p. 347.

1901 Sphaerophthalmus alatus Angelin {sic)\ Lindstrom, p. 29, pi. Ill, figs. 31-34.

1913 Sphaerophthalmus alatus (Boeck sp.); Lake, p. 74, pi. VIII, fig. 1-5.

1957 Sphaerophthalmus humilis (Phillips 1848); Henningsmoen, p. 215, pi. 5; pi. 22, figs. 7, 1 1-15.
1968 Sphaerophthalmus humilis (Phillips); Rushton, p. 415, text-fig. 2, 3a, pi. 78, figs. 11-15.

EXPLANATION OF PLATE 94

Figs. 1-4. Ctenopyge (Mesoctenopyge) tumida Westergird 1922. Naersnes, Royken. Zone Vb (2d/3).
1, Anterior region of small adult eye with some lenses missing. P.M.O. 87552, x 105. 2, Lowermost
lenses and eye-socle of large adult eye; faint vertical striations visible on eye-socle. P.M.O. 87554, x 135.
3, Palpebral lobe, outer central region with striations nearly normal to the outer edge. P.M.O. 87552,
x250. 4, Adult eye in lateral view showing eye-socle. Visual surface damaged. P.M.O. 87553, x75.
Fig. 5. Block with librigenae and eyes of Ctenopyge (Mesoctenopyge) similis Henningsmoen 1957 (left),
and Ctenopyge (Eoctenopyge) modesta Henningsmoen 1957 (centre and right). Sars Gate, Oslo. Zone Vb
(2d sim.). P.M.O. 87564, x26.

Fig. 6. Sphaerophthalmus humilis Phillips 1848. Andrarum, Scania. Zone Vc. Lower central part of eye;
external surface of cornea and internal moulds of lenses. Gr. I. 20706, x 180.

PLATE 94

CLARKSON, Ctenopyge and Sphaerophthalmus eyes

756

PALAEONTOLOGY, VOLUME 16

TEXT-FIG. 7. a-d. Sphaerophthalmns alatus (Boeck 1838). Zone Vb. Gamlebyen, Oslo. Restoration of the
cephalon in dorsal, frontal, and lateral aspects and (d) enlarged in antero-lateral view from P.M.O. 87556-

87558.

Plate 94, fig. 6; Plate 95, figs. 3-6; text-fig. 8a- J

Material. Three blocks from Andrarum, associated with P. scarabaeoides scarabaeoides, Ct. linnarssoni,
and Ct. teretifrons. Zone Vc (2dy). Old collection, Gr. I. 20706, 20803, 5537.

Remarks. S. humilis is an extremely convex trilobite with a very pronounced anterior arch and almost
vertical librigenae. The peculiar attitude of the genal spines in this species was first noted by Rushton
(1968, p. 41 5). He reconstructed the cephalon with steeply sloping librigenae, and ventrally projecting genal
spines, curving in under the cephalon. I have been able to confirm that the genal spines do plunge down-
wards as Rushton described, so that the cephalon could not rest upon the sea floor. In my restoration,
these spines do not curve inwards quite so strikingly, but in all other respects I agree with Rushton.

Eye-morphology. This species is unusual because of the relatively enormous size of the eye, and its very far
posterior position. It is one-third the total length of the cephalon, with its anterior edge opposite SI. Eye-
indices: A/G 38%, A/Gn 30%, H/A 335%. The palpebral lobe, which is jointed to the ocular ridge, is similar
in form to that of S. alatus, though narrower. No surface ornament has been detected in the material
examined. Since the palpebral lobe is placed slightly behind the widest part of the cranidium, the anterior
edge of the visual surface appears to be slightly recessed. The visual surface forms about a third to a half
of a slightly oblate spheroid, with the long axis horizontal. Its edge lies in an exsagittal plane, inclined at
some 1 0° from the vertical. In side view the eye appears nearly globular. Each eye subtended a visual field
whose lower limits are 70° to 80° below the equator, and which just overlap at the front, rear, and above so
as to give an almost entirely panoramic range not found in other olenids, and, indeed, in few other trilo-
bites. Juvenile eyes are of similar form, but have fewer lenses (c. 70 as compared with c. 200). Both in
juvenile and adult eyes the eye-socle is very narrow and shows no definite structure.

In the material to hand the thin biconvex lenses are easily detached. Where they are partially removed
from the matrix each is preserved as a single calcite crystal with a very slightly convex upper surface and
a more strongly convex inner face. Distinct cleavages are visible, from which it can be deduced that the
c-axis of each crystal is normal to the surface. Sometimes the finer detail has been destroyed by recrystalliza-
tion, but the impressions left by the lower surfaces of the lenses show their arrangement very clearly (PI. 95,
figs. 3-4).

CLARKSON: OLENID EYES

757

Development of the eye in later leptoplastines. In most trilobites the first-formed lenses
are emplaced in an initial horizontal row parallel with the facial suture. New lenses
are added below these, in parallel horizontal rows. The new lenses are offset relative
to those above so that there develops an array of lenses arranged in a regular system
of hexagonal close packing. This pattern is most clearly shown in the phacopids,
where the lenses are large and separate; dorso-ventral files can be seen intersecting
with ascending and descending diagonal files (Clarkson I966u). In some phacopid
eyes new small lenses may actually develop in an accessory row above the initial
horizontal row, but this seems to be confined to certain genera only (Beckmann 1951 ;
Clarkson 1966/?). The eyes of Ct. (E.) modesta, Ct. (M.) similis, and Ct. (M.) tuniida
are closely similar to one another, and though there are differences in size and posi-
tion the resemblance in detailed structure is such that though most of the comments
given here are based upon Ct. (M.) similis (PI. 94, fig. 2, text-fig. 5e) they are appro-
priate also to the others. In most respects they apply also to Sphaerophthalmus eyes,
though the latter have more lenses.

All these olenid eyes begin their development in much the same way as phacopids,
though being holochroal the lenses are contiguous and they are all much the same
size. The first-formed lenses lie in a horizontal row following the curve of the facial
suture. When seen from above this row and subsequent rows appear to be concentric
and curving outwards like parallel strings of beads. In the upper (i.e. the oldest) part
of the eye the close-packing system is regular and arranged with respect to the
dominant elements— the horizontal rows. But some distance below the facial suture,
usually after the first half-dozen rows, irregularities are encountered which break up
this clearly defined pattern. Why do these develop?

TEXT-FIG. 8. a-d. Sphaerophthalmus humilis (Phillips 1848). Zone Vc. Andrarum, Scania. Restoration of the
cephalon in (a) dorsal, (b) frontal, and (c) lateral aspects and {d) enlarged in antero-lateral view, from
Gr. I. 20706.

758

PALAEONTOLOGY, VOLUME 16

It has long been established that in trilobites generally new lenses are usually added
only along the lower margin of the visual surface, and that each normally arises
below and directly between two existing lenses of the preceding horizontal row.
Irregularities come into being when extra lenses are intercalated into this system, in
other words when at a few loci two new lenses are emplaced instead of one. If, as
I discussed previously in the phacopid Ormathops (Clarkson 1971), the developmental
system is ‘programmed’ to produce new lenses when a particular spatial threshold
has been reached, then such new intercalated lenses will necessarily be emplaced to
fill the ‘extra space’ as the visual surface expands in circumference. Each of the new
intercalated lenses will in turn act as a focus for lens-initiation in successive horizontal
rows, and the effects of these small, though inevitable irregularities are clearly visible.
Had these olenids possessed lenses graduated in size such irregularities would never
have arisen, but since they are all much the same size, disruptions of the regular pack-
ing system are, as in Ormathops, a geometrical requirement.

The lower third of the eye lies below the ambitus (this term used as in an echinoid)
where the eye has reached its greatest horizontal circumference and thereafter
decreases slightly in diameter. Since the visual surface is no longer increasing no new
lenses are added by intercalation and it is hardly surprising to find another change in
the manner of lens emplacement. What usually happens is that one of the two original
sets of diagonal rows swings into a vertieal orientation, these becoming the vertical
files characteristic of the lower part of the eye, though they are not homologous
with the dorso-ventral files of the phacopids. The lenses in this region are slightly
smaller than the upper ones, accommodating the slight decrease in the diameter of
the eye.

Thus the eyes of olenids of this type have three horizontal zones, an upper regular
zone where the horizontal rows are the dominant elements, a zone of intercalation,
where the horizontal rows are still dominant though there are notable irregularities,
and a lower zone of dorso-ventral files where the system of hexagonal close packing
is based upon one set of diagonal rows which has now become vertical in response
to packing requirements. This system resembles that of the pelturines only in so far
as the diagonal rows change direction and become more vertical. But the pelturine
system is based upon arithmetical change in distances between lens-centres, whereas
in later leptoplastines, distances between lens-centres remain constant, except where
the increasing girth of the eye promotes hiatoses.

EXPLANATION OF PLATE 95

Figs. 1, 2. Sphaerophthalmus alatus (Boeck 1838). Andrarum, Scania. Zone Vb (2d/3). 1, Dorso-lateral
and 2, anterior view of young eye. Gr. I. 20775, x 140. In 1 right edge is anterior.

Figs. 3-6. Sphaerophthalmus humilis {VYi\W\p?, 1848). Andrarum, Scania. Zone Vc. 3, Single lens, partially
detached from matrix, and internal moulds of missing lenses. Gr. I. 20803, x 1200. Bar = 10 ^xm.

4, Internal mould of adult eye with a few lenses still adherent. Left edge anterior. Gr. I. 20803, x60.

5, Very young eye; outer surface Gr. I. 5537, x 200. 6, Internal mould of lenses near base of eye. Gr. I.
20803, x500. Bar = 10 ^m.

PLATE 95

CLARKSON, Sphaerophthalmus eyes

760

PALAEONTOLOGY, VOLUME 16

SUMMARY AND CONCLUSIONS

Evolution of the eye in the Olenidae. In this summary of observations I have largely
followed the phylogenetic scheme of Henningsmoen (1957, Chart 6), and though this
will certainly need to be modified in the light of recent and future observations, it
still forms a useful provisional basis for discussion of phylogeny.

According to Henningsmoen, Olenus, which was the earliest Upper Cambrian
genus, persisting through Zones I and II only, gave rise to three lines of descent. The
most conservative line led to Parabolina (Zone III to Lower Tremadoc), and to other
Olenidae. At about the same time (Zone III), the first pelturine genus Protopeltura
appeared, and was followed later (Zone Vb), by Peltura and other Pelturinae. From
Olenus also descended a third group, the Leptoplastinae, which began with Lepto-
plastus in Zone IV, a genus which gave rise to the elaborate leptoplastines Ctenopyge
and Sphaerophthalmus which flourished in the time of Zone V.

Though the Triarthrinae and various other genera of the established families
appeared and evolved in Tremadoc and later times, none of these have been studied
in detail in my present work : in most of them the eye is small and rarely well preserved,
though some of the Argentinian olenid species, e.g. Parabolina argentina, Saltaspis
steinmanni, have relatively large eyes (Harrington and Leanza 1957, pp. 83, 95).

Opik (1963) included the monogeneric Australian subfamily Rhodonaspidinae
Opik in the Olenidae. Because of the marked similarity of the pygidium of Rhodonaspis
to that of Parabolina, he suggested a close relationship between these two genera, and
noted that Parabolina may not have been derived from Olenus, but was part of another
complex (including Rhodonaspis), which had persisted from the early Upper
Cambrian. The evidence for this rests on the pygidial resemblance alone.

Study of the eye-morphology of Scandinavian genera, however, supports the other
criteria used by Henningsmoen in erecting his phylogenetic scheme; the similarity of
the eyes of Olenus meraspids and adult Parabolina has already been pointed out.
Rhodonaspis has very large eyes of unusual form, having the palpebral lobe and prob-
ably the ocular ridge also as double structures, divided by an ocular striga. The
ocular ridge is separated from the glabella, as in adult Olenus.

Further descriptions and discussions of new olenid material may help to resolve
conflicting suggestions as to olenid phylogeny, and it is to be hoped that the rich
Cambrian successions of Queensland may furnish yet more material of olenids and
related trilobite families. For the moment, however, I have preferred to take a con-
servative view of olenid phylogeny.

The principal conclusions which have emerged from the present studies are as
follows:

(a) The ‘primeval’ olenid eye from which all other kinds ultimately derived is
exhibited by Olenus. Here the ocular suture is functional in the adult, so that the visual
surface is found only in meraspids. Large adults have a highly structured and possibly
sensory zone surrounding the visual surface, well supplied with alimentary prosopon.
This peripheral zone is, however, weakly developed in meraspids. In the latter the
lenses are plano-convex, and are probably welded to the inner surface of the cornea,
sometimes being quite widely spaced.

{b) In both the later Oleninae and in all the Pelturinae, the eyes have many features

CLARKSON; OLENID EYES

761

very similar to those of meraspids of Olenus. These include their small size and forward
position, the inclination of the long axis to the exsagittal plane, the non-functional
ocular suture, structure of the visual surface, poorly developed peripheral zone, and
the confluence of the palpebral lobe with the ocular ridge. All these are suggestive of
a paedomorphic origin for these adult eyes. Pelturine eyes have lenses decreasing in
size ventrally, arranged in a geometrically regular system with logarithmic diminu-
tion of distances between lens-centres. It is probable that the eyes of most Tremadoc
and later olenids are also of this type.

(c) Early Leptoplastinae had eyes like those of Olenus, though their generally
poorer preservation precludes very detailed comparison. The later leptoplastines
Ctenopyge and Sphaerophthalmus retained the visual surface in the adult, and the
eye is often strikingly well developed, though the peripheral zone is not greatly in
evidence. Some features of these eyes may likewise be regarded as paedomorphic.
Eyes of this kind are usually spheroidal, and have thin biconvex lenses underlying
a very thin cornea. They are variable in size, position, and in the shape of the palpebral
lobe and the range of the visual field (the most extreme form being S. humilis). The
visual surface has a peculiar pattern of development, which is very clear in Ct. (M.)
similis, but seems to be constant throughout the group. In this the lenses are all of the
same size, and have a distinct zone of irregularities. Eollowing the extinction of the
last species of Ctenopyge and Sphaerophthalmus before the close of the Upper
Cambrian, no other olenids evolved such remarkably developed visual organs.

Adaptations of the olenid cephalon. Henningsmoen (1957, pp. 70-82) has written exten-
sively about the mode of life and environment of the olenids. He suggested that
although most olenids were capable of swimming above the sea floor, they could also
sojourn for certain periods on the floor of the stagnant Olenid Sea, and were probably
adapted for life in waters with a restricted oxygen content. Further evidence of bottom-
dwelling habits in some olenids is provided by recent trace-fossil analysis (Orlowski,
Radwahski, and Roniewicz 1970; Birkenmajer and Bruton 1971).

The cephalic reconstructions presented here, which were made to show the eye in
its correct relationship to the rest of the cephalon, also seem to indicate, in some cases,
functional adaptations for a benthonic mode of life. Most olenid cephala (with the
notable exception of S. humilis), seem to be well adapted for resting upon the sea
floor. The short genal spines of most species of Olenus, Parabolina, and Leptoplastus,
together with S. alatus, project horizontally from the cephalon, so that the trilobite
could lie on the sea floor, with its cephalon propped in a stable position and having
its anterior arch open. Such support is even more evident in genera with long genal
spines. Thus the genal spines of E. modesta are long and elegantly curved, in such
a manner that the cephalon could be supported on four points; the lowest parts of the
two antero-lateral borders, and the lower surfaces of the genal spines just in front of
their tips. Ct. similis and Ct. tumida have very long flattened and horizontal genal
spines, which would give support over their whole length. In these the height of the
occipital ring above the base level, and the oblique appearance of the postero-lateral
border, when seen from the side suggest that the body was carried high above the
sea floor, a habit which no doubt carried real functional implications. I have else-
where contended (Clarkson 1969) that long spines in the odontopleurid trilobites are

H

762

PALAEONTOLOGY, VOLUME 16

support structures rather than being used to prevent sinking through frictional
retardation. Judging by the structure of the cephalon in the long-spined olenids, the
same principle seems to apply, but it will be necessary to prepare complete lateral
reconstructions with the thorax and pygidium in place before these suggestions can
be fully worked out. The immensely long thoracic spines of Ctenopyge, and their
relationship to the rest of the body still pose intriguing problems.

Clearly 5. humilis was very differently modified, as witness the nearly vertical genal
spines and the very large eye with its greatly expanded visual field. Though the pur-
pose of such adaptations is far from certain, it is evident that even within the confines
of the organization of such a close-knit family as the Olenidae, there was still a sub-
stantial degree of evolutionary plasticity, and the possibility of individual functional
differentiation.

Acknowledgements. I am very grateful to Dr. D. L. Bruton (Paleontologisk Museum, Oslo), Mr. S. F.
Morris (British Museum, Natural History), and Mr. F. J. Collier (Smithsonian Institution, Washington),
for the loan of specimens. I am also glad to record the assistance of Dr. Jan Bergstrom of the University of
Lund for loan of specimens and for much helpful discussion in the early stages of the work. I particularly
appreciate the generosity of the Trustees of the Moray Fund of the University of Edinburgh, who con-
tributed towards the cost of a Wild-Heerbrugg camera-lucida microscope.

All the SEM photographs were taken using a Cambridge ‘Stereoscan’ by Mr. J. Goodall of the Depart-
ment of Engineering, Edinburgh University, to whom I am greatly indebted for his continued skill and
patience. I gratefully acknowledge a grant towards plate cost from the Carnegie Trust for the Scottish
Universities.

REFERENCES

ANGELIN, N. p. 1854. Palaeontologio scandinavica. Pars 1. Crustacea formationis transitionis. Fasc. II.
I-ix, 21-92, pis. XXV-XLI. Holmiae (Stockholm).

BECKMANN, H. 1951 . Zur Ontogenie der Sehflache grossaugiger Phacopiden. Paldont. Z. 24, 126-141, pi. 10.

BiRKENMAJER, K. and BRUTON, D. L. 1971. Some trilobite resting and crawling traces. Lethaia, 4, 303-319,
figs. 1-14.

BOECK, c. 1838. Uebersicht der bisher in Norwegen gefundenen Formen der Trilobiten-Familie. In keilhau,
Gaea Norvegica, I, p. 138-145. Christiana (Oslof

BROGGER, w. c. 1882. Die Silurischen Etagen 2 Und 3 im Kristianagebiet und auf Eker. Universitats-
programm fur 2. Sem. 1882, pp. i-viii, 1-376, pis. I-XII. Kristiana (Oslo).

BRLINNICH, M. T. 1781. Beskrivesle over Trilobiten. K. danske Vidensk. Selsk. Skr. N.S. 1, 1-384.

CLARKSON, E. N. K. 1966a. Schizochroal eyes and vision of some Silurian acastid trilobites. Palaeontology, 9,
1-29, pis. 1-3.

19666. The life attitude of the Silurian trilobite Phacops musheni Salter 1864. Scott. J. Geol. 2, 76-

83, pi. 1.

1969. A functional study of the Silurian odontopleurid trilobite Leonaspis deflexa (Lake). Lethaia, 2,

329-344, figs. 1-7.

1971. On the early schizochroal eyes of Ormatliops (Trilobita, Zeliszkellinae). Mem. Bur. Recherches

Geol. min. 73. (Colloque ordovicien-silurien), 51-63, figs. 1-2, pi. 1.

DALINGWATER, J. (in press). Trilobite cuticle microstructure and chemistry. Lethaia.

HARRINGTON, H. J. and LEANZA, A. F. 1957. Ordovician Trilobites of Argentina. Univ. Kansas (Lawrence)
Spec. Pub. 1, 1-276, figs. 1-140.

HENNiNGSMOEN, G. 1957. The trilobite family Olenidae. Norske Videnskapsakademien (Oslo) Mat.-Naturv.-
Kl. Skr. 1, 1-303, pis. 1-31, figs. 1-19.

JAGO, J. B. 1972. Two new Cambrian trilobites from Tasmania. Palaeontology, 15, 226-237, pi. 44.

JELL, p. A. 1970. Pagetia ocellata, a new Cambrian trilobite from northwestern Queensland. Mem. Qd.
Mus. 15, 303-313, pi. 23-24.

CLARKSON: OLENID EYES

763

KJERULF, T. 1857. liber die Geologie des siidlichen Norwegens. Nyt. Mag. Naturvid. 9, 193-333, pis. I-V.
Christiana (Oslo).

LiNDSTROM, G. 1901. Researches on the visual organs of the trilobites. K. Svensk. Vetensk. Acad. Handl.
34, 1-86, pis. 1-6.

OPIK, A. A. 1961. Alimentary caeca of agnostids and other trilobites. Palaeontology, 3, 410-438, pis. 68-70.

1963. Early Upper Cambrian fossils from Queensland Bur. Miner. Resour. Aust. Geol. Geoplivs. Bull.

64, 1-133, pis. 1-9.

1967. The Mindyallan fauna of northwestern Queensland. Ibid., 74 (2 vols.); v. 1, 1-404, v. 2, 1-167,

pis. 1-67.

ORLOWSKi, s., RADWANSKi, A. and RONiEWicz, p. 1970. The trilobite ichnocoenoses in the Cambrian sequence
of the Holy Cross Mountains. In Trace fossils. Ed. crimes, t. p. and harper, j. c. Geol. J. Special Issue,
3, 345-360, figs. 1, 2, pis. 1-4.

PHILLIPS, J. 1848. The Malvern Hills compared with the Palaeozoic Districts of Abberley, May Hill, Tort-
worth, and Usk. With Palaeontological Appendix. Mem. Geol. Surv. Great Britain, II (Part 1), 1-386,
pis. 1-30. London.

RUSHTON, A. w. A. 1968. Revision of two Upper Cambrian Trilobites. Palaeontology, 11, 410-420, pis.
77-78.

STRAND, T. 1927. The ontogeny of Olenus gibbosus. Norsk, geol. Tidsskr. 21, 49-164, pis. 1, 2.

STRUVE, w. 1958. Beitrage zur Kenntnis der Phacopacea (Trilobita); 1. Die Zeliszkellinae. Senckenberg.
Leth. 39, 165-219. Abb. 1-16, taf. 1-4.

WAHLENBERG, G. 1821. Petrificata Telluris Svecana. Nova Acta. Soc. Regiae Sci. 8, 1-116, pis. 1-4.
WALCOTT, c. D. 1910. Cambrian geology and palaeontology, 6; Olenellus and other genera of the Meson-
acidae. Smithsonian Misc. Coll. 53, 231-422, pis. 23-44.

WESTERGARD, A. H. 1922. Sveriges Olenidskiffer. Sveriges Geol. Undersok. Ser. Ca, 18, 1-205, pis. 1-16.
WHITTINGTON, H. B. 1958. Ontogeny of the trilobite Peltura scarabaeoides from the Upper Cambrian,
Denmark. Palaeontology, 1, 200-206, pi. 38.

1966. Phylogeny and distribution of Ordovician trilobites. Journ. Paleont. 40, 696-737, figs. 1 16.

WHITWORTH, p. H. 1970. Ontogeny of the Upper Cambrian trilobite Leptoplastus crassicornis (Westergard)
from Sweden. Palaeontology, 13, 100-111, pis. 22-24.

E. N. K. CLARKSON
Grant Institute of Geology
University of Edinburgh
West Mains Road

Typescript received 25 October 1972 Edinburgh, EH9 3JW

Discussion on Dr. Clarkson’s paper:

Chaloner; Did you clean the surface in any way prior to examination?

Clarkson: Yes. I used a Directional Ultrasonic Cleaner, which is a gun device manufactured by the Simms
Group R. & D. Ltd. The specimen is placed under water and ‘blasted’ for two or three seconds by a stream
of high-velocity bubbles from the gun. This suffices to remove all the loose dirt and dust; but if the gun is
operated for longer than a few seconds the surface of the specimen may be damaged by abrasion.

COMBINED TRANSMISSION AND SCANNING
ELECTRON MICROSCOPY OF IN SITU
PALAEOZOIC SPORES

by T. N. TAYLOR

Abstract. This paper discusses pollen and spores isolated from Carboniferous reproductive organs, including
fructifications belonging to lycopods, cordaites, and seed ferns. Microspores of the monosaccate genus Endosporites
were macerated from sporangia of a single cone. Morphological variability ranges from grains still within the tetra-
hedral arrangement to solitary spores. Information is provided concerning saccus-corpus organization and exine
ultrastructure, ornamentation, and stages of saccus ontogeny. Pollen grains included within the genus Florinites
were examined in cordaitalean pollen sacs from different localities and stratigraphic levels. Both proximal and distal
attachments between saccus and corpus are demonstrated. Spores of a new Pennsylvanian reproductive structure are
described as consisting of a complex tectate exine supporting a verrucate ornamentation. Prepollen grains of the
Schopfipollinites-type were isolated from a number of medullosan pteridosperm reproductive structures including
the genera Dolerotheca, Rhetinotheca, Aulacotheca, Whittleseya, and Halletheca. Comparative studies of the spore
exines suggest the occurrence of fundamental ultrastructural differences among the grains. Information is presented
concerning the possible site of gamete emission from Schopfipollinites prepollen grains.

The successful application of transmission electron microscopy to the study of fossil
pollen exines by Ehrlich and Hall (1959) initiated a new era of palaeobotanical inquiry.
With the subsequent availability of the scanning electron microscope and its ease of
application in palaeontological research problems, researchers today are examining
evidences of biological activity extending from the Precambrian to Recent, and includ-
ing all levels of biological organization.

While dispersed spores and pollen grains found in rocks of Palaeozoic age have
received a great deal of attention both for palaeontological and geological purposes,
spores and pollen found in fructifications of known biological affinities have received
little attention. It is the intent of the present paper to discuss the application of com-
bined transmission and scanning electron microscopy to the study of in situ pollen,
prepollen, and spores of Carboniferous (Pennsylvanian) age, and to demonstrate
some of the types of information which have been made available utilizing these
methods.

Studies of in situ pollen grains and spores are of particular importance because in
most instances grains are present in sufficient numbers that ontogenetic differences
may be separated from those which are truly taxonomic. This developmental
approach may be undertaken with certain types of reproductive structures which
mature in a sequential manner. The opportunity of sampling almost pure popula-
tions of spores, differing developmentally, from varying positions within a single cone
provides an ideal means of studying spore wall ontogeny.

Because of the nature of this symposium the number of light photomicrographs
used has been greatly reduced. It must be pointed out, however, that the use of light
microscopy constitutes a valuable and indispensable aspect of any study in which the
maximum number of pollen grain or spore features are to be elucidated.

[Palaeontology, Vol. 16, Part 4, 1973, pp. 765-776, pis. 96-98.]

766

PALAEONTOLOGY, VOLUME 16

Technique. Pollen grains and spores were macerated from sporangia using dilute
hydrochloric acid (2%), soaked in 12% hydrofluoric acid for 12 hours, and subse-
quently divided into three fractions. Grains to be examined by transmission electron
microscopy were dehydrated to propylene oxide and embedded in Spurr low viscosity
media. Sections cut at approximately 20 nm were poststained in a 5% aqueous solu-
tion of uranyl acetate for 20 minutes, followed by lead citrate. Grains to be examined
with the scanning electron microscope were washed in two changes of distilled water,
spread on standard SEM specimen stubs coated, while in suspension, with a thin
film of dried silver conductive paint, and allowed to dry in air. The grains were then
vapour coated with a thin film of gold and examined with a Cambridge Mark IIA
instrument. Grains of the final sample were dehydrated to xylene and embedded in
Harleco Synthetic Resin for examination by light microscopy.

In the case of very large grains, several hundred microns in diameter, such as
specimens of SchopfipoUinites, very satisfactory results have been obtained using the
transmission-scanning operational mode (Swift and Brown, 1970). This technique
enables entire sections to be examined in a single field of view. It consequently
eliminates the necessity of constructing elaborate montages which are otherwise
necessary for large grains because the minimum magnifications available with the
transmission electron microscope are too high to record cell and wall component
interrelationships. Sections ranging from 50 nm-1 were cut and mounted on
single-hole slot grids covered with a thin collodion/carbon support film. Grids were
subsequently stained with aqueous uranyl acetate followed by lead citrate and fitted
into a special device mounted in the instrument stub holder.

The following is a discussion of four different spore types and the kind of informa-
tion which has been assembled on their organization, morphology, and ultrastructure.

ENDOSPORITES

Endosporites Wilson and Coe 1940 is a monosaccate, trilete, microspore now known
to be biologically related to some heterosporous members of the Lycophytina. The
source of Endosporites specimens in previous studies has been from compressed,
structureless cones, or from sporae dispersae assemblages. The material presented
here comes from calcium carbonate petrifactions collected from probable lower
Pennsylvanian sediments in eastern Kentucky (Good and Taylor 1970). Of particular

EXPLANATION OF PLATE 96

Fig. 1. Transmission electron micrograph. Composite reconstruction of three spores of Endosporites
tetrad. Arrows indicate positions of apical papillae, x 510.

Fig. 2. Partial lateral view of monosaccate spore showing thickened corpus wall (left) with external orna-
mentation, and internal saccus reticulations, x 2700.

Fig. 3. Three spores of Endosporites tetrad showing relationship between corpus and saccus wall, x 2100.

Fig. 4. Scanning electron micrograph of Endosporites tetrad showing three spores, with the fourth repre-
sented by the corpus and a remnant of the saccus. x 500.

Fig. 5. Lateral view of Florinites grain showing proximal and distal saccus attachment, and internal saccus
reticulations. xlOOO.

Fig. 6. Limbus and proximal surface of Endosporites spore, x 4750.

Fig. 7. Immature Endosporites spore prior to saccus enlargement. X 3300.

PLATE 96

TAYLOR, Palaeozoic spores

768

PALAEONTOLOGY, VOLUME 16

importance with this material is the fact that within certain sporangia various stages
of spore ontogeny are represented, including spores still within the tetrahedral
arrangement (PI. 96, fig. 1).

One of the most obvious advantages of transmission and scanning electron micro-
scopy in the study of pollen grains and spores lies in the ability to discern more
accurately and interpret correctly, complex ornamentation patterns which may be
present on both the internal and external exine surfaces. PI. 96, fig. 6 shows the distal
surface of a spore ornamented by closely spaced, blunt-tipped spinules which are
basally fused to form a reticulum. Along the equatorial rim of the spore, where the
proximal and distal faces of the saccus become continuous, the spinous projections
are fused, and together with the loosely arranged exinous strands, provide a thickness
to the limbus. The proximal surface of Endosporites is uniformly smooth, interrupted
only by irregularly shaped and randomly disposed pits (PI. 96, fig. 6). Such features
are too small to be resolvable with light microscopy, and are extremely difficult to
characterize by transmission electron microscopy.

PI. 96, fig. 4 shows three complete spores of a tetrad and the fourth represented by
the corpus and a small remnant of the saccus wall. Spores with the saccus fragmented
and torn allow observations to be made of the internal saccus surface and show
irregular reticulations of the wall, as well as ornamentation of the corpus.

One of the conspicuous features of Endosporites microspores from the Kentucky
locality when examined by light microscopy is the occurrence of three apical papillae
(= interradial papillae) that appear as small dark crescent-shaped objects close to the
spore centre; one between each of the laesurae. Transmission and scanning electron
microscopy define the nature and relationship of these structures to the saccus and
corpus walls. The lowest spore in PI. 96, fig. 1 shows two of the apical papillae
separated by a shallow depression and occurring on the inner surface of the corpus
wall; but not arising from the outer saccus wall as has been previously thought.

Ultrastructure and exine stratification of Endosporites are best illustrated in
PI. 96, fig. 1 where the relationship between the wall of the corpus and saccus is
apparent. The exine is constructed of two layers, with units comparable to the topo-
graphic equivalents ‘nexine’ and ‘sexine’. In Endosporites, corpus-saccus attachment
occurs only at the proximal pole (PI. 96, figs. 1, 7). In the region where the corpus and
saccus are fused on the proximal surface the wall is quite thick (3 /xm) and consists
of a series of anastomosing exinous strands. PI. 96, figs. 1, 3 illustrates the proximal
continuity provided between saccus and corpus by these delicate strands as viewed
in both the transmission and scanning modes.

The spores macerated from the sporangia range from 73 to 121 /xm in diameter;
however, some of the microspores are appreciably smaller (21 /xm). A large number
of these smaller spores lack trilete scars and it has been suggested that these may
represent isolated central bodies. Other small spores of a similar diameter, but
possessing a well-defined trilete mark and in some instances apical papillae, are
further distinguished from the more typical Endosporites grains by a more highly
ornamented distal surface and the absence of a clearly defined saccus (PI. 96, fig. 7).
Transmission micrographs of this latter spore type show apical papillae and a
reduced saccus surrounding the central body. These two features suggest a level of
spore ontogeny in which the saccus is just beginning to differentiate through the

TAYLOR: ELECTRON MICROSCOPY OF SPORES

769

separation of the two wall layers. A fully developed saccus in Endosporites appears
like the configuration of the spores in PI. 96, fig. 1.

FLORINITES AND FLO RINITES-TYPE POLLEN

The genus Florinites Schopf, Wilson, and Bentall 1944 was instituted for dispersed
monosaccate pollen grains of presumed cordaitalean affinity. The generic diagnosis has
been difficult to apply, principally because of the confusion regarding grain morpho-
logy and structural organization. With the improved techniques now available, new
information has been obtained on the morphology of this type of grain. Specimens
used were macerated from Cordaianthus pollen sacs found in coal balls collected near
West Mineral, Kansas (middle Pennsylvanian), and in eastern Kentucky.

Florinites grains are monosaccate, and consist of a spherical central body (= corpus)
surrounded laterally by a large, internally reticulate, air bladder (= saccus). In polar
view the saccus appears circular-elliptical, with the corpus typically circular in out-
line. In several morphological studies of cordaite pollen the grains are described as
consisting of a large air sac which completely encircles the internal corpus. In these
studies body-bladder attachment has been described as occurring only at the distal
pole. Combined transmission and scanning electron microscopy demonstrate that the
attachment between the body and the bladder occurs at both the proximal and distal
poles (PI. 96, fig. 5). On the proximal surface body-bladder attachment is approxi-
mately equal to the maximum body diameter, whereas distally this attachment is
typically less than body diameter. Distally the body-bladder attachment may assume
a variety of configurations, varying from small and irregular to large and highly
angular. In some instances attachment may approach a configuration which super-
ficially resembles a sulcus. On the proximal surface of the grain attachment appears
to be more regular and conforms to the symmetry of the central body. Thus in the
case of the monosaccate grain Florinites the structure consists of a central body totally
enclosed by a bladder and fused to it on both the proximal and distal surface.

External bladder ornamentation is best described as laevigate when viewed with the
light microscope. When examined with the SEM, the bladder appears as a series of
irregular depressions corresponding to the outlines of the internal bladder reticula-
tions (PI. 96, fig. 5). The ability to examine a fractured internal or enclosed grain
component, in this instance the external surface of the corpus, provides information
which would be impossible to discern with light microscopy (PI. 97, fig. 5).

Florinites grains demonstrate the same apparent bladder ontogeny characteristic
of extant saccate pollen in which the bladder arises by a separation between the
sexine and nexine. Small ridges which characterize the external surface of the central
body on the lateral walls represent former regions of exine attachment prior to saccus
inflation (PI. 96, fig. 5). The nexine, which is typically lamellated in extant saccate
pollen, shows no observable ultrastructural layering in the Florinites specimens.

Another monosaccate pollen grain resembling Florinites in many morphological
features was also recovered from Cordaianthus pollen sacs collected at the eastern
Kentucky locality. The grains are larger than the Florinites specimens, ranging in
size from 115 to 180 /urn. Both radially symmetrical and bilaterally symmetrical
specimens were found. Grains which show a radial organization are exclusively

770

PALAEONTOLOGY, VOLUME 16

trilete, while the bilateral specimens have a suture organization of the monolete type.
Most grains show varying degrees of suture expression between these two types. In
this grain saccus-corpus attachment also occurs at both the proximal and distal
poles.

Externally the saccus is psilate except in the region of the proximal pole. PI. 97,
fig. 6 indicates that the region of the trilete mark is complex, consisting of a series of
uneven muri which are most prominent closest to the suture, decreasing in size away
from the laesurae. The trilete mark consists of a narrow Y-shaped depression support-
ing a median, elevated ridge. The central ridge appears to arise from the base of the
depression, and is supported by delicate ribs which are uniformly situated at right
angles to the long axis of the laesura.

Internally, the saccus of this grain is ornamented by a network of inwardly project-
ing wall thickenings which form a reticulate pattern, and appear similar to the orna-
mentation of Florinites. The external corpus wall is ornamented by a delicate reti-
culum (PI. 96, fig. 2). In section view the projections of the reticulum appear as finely
spaced, uniformly developed processes which at some levels appear to bifurcate at
their tips (PI. 96, fig. 2). These processes appear continuous with the saccus reticulum
and constitute an ontogenetic feature resulting from the separation of the sexine
early in the formation of the saccus. The corpus wall is quite thick, measuring approxi-
mately 2 |u,m (PI. 96, fig. 2). Ultrastructurally, this layer is composed of a series of
irregularly thickened lamellations which may be up to 0-2 /xm thick (PI. 97, fig. 4).

TECTATE GRAIN

The grain illustrated in PI. 97, fig. 1 was extracted from a reproductive organ
consisting of numerous thick-walled sporangia which are attached to vascularized
bract-like structures by elongate pedicels (Taylor 1972). The spores are trilete
and circular— subcircular in outline. They range from 38 to 55 ;um in diameter and
are characterized by trilete rays which extend approximately three-quarters of
the spore radius. The spore illustrated in PI. 97, fig. 1 shows the elevated and pro-
minent nature of the trilete mark. Ornamentation consists of a series of irregular
verrucae which extend up to 1-4 [xm high (PI. 97, figs. 1, 2). When examined by
light microscopy the surface of the verrucae appear to bear slight depressions.
Ultrathin sections of the spores indicate, however, that this pattern is the result of

EXPLANATION OF PLATE 97

Fig. 1 . Proximal surface of Pennsylvanian spore showing verrucate ornamentation and trilete mark, x 1 200.

Fig. 2. Fractured surface of spore in Fig. 1 showing level of exine organization and relationship of wall
layer components, x 6000.

Fig. 3. Transmission electron micrograph of Fig. 1 spore exine. x 8500.

Fig. 4. Transmission electron micrograph of corpus wall of monosaccate spore in PI. 96, fig. 2, showing
nexine lamellations. x 15000.

Fig. 5. Scanning electron micrograph of Florinites grain showing ruptured saccus and ornamentation of
corpus. X 5900.

Fig. 6. Proximal region of monosaccate grain showing complex organization of trilete mark, x 900.

Fig. 7. Transmission electron micrograph of spore in Fig. 1 showing wall stratification. Interruptions in
the nexine indicate the position of two laesurae of the trilete mark, x 1400.

PLATE 97

TAYLOR, Palaeozoic spores

Ill

PALAEONTOLOGY, VOLUME 16

exine organization rather than surface features of the verrucae (PI. 97, figs. 3, 7).
The section of the spore illustrated in PI. 97, fig. 7 is slightly oblique in the proximal-
distal plane so that two of the laesurae appear as interruptions in the inner component
of the wall.

The exine of the grain consists of four easily delimited layers. Using the terms
commonly applied in pollen exine organization, these may be called tectum, colu-
mellae, pedium; and some level of nexine development (PI. 97, figs. 3, 7). The colu-
mellae extend from the foot layer or pedium, but are not in contact with the tectum
at their distal ends. The tectum appears to be attached along the sides of the columellae,
rather than at the distal end of each structure (P. 97, fig. 1). The broken surface of the
spore wall illustrated in PI. 97, fig. 2 clearly shows the relationships of the exine
components, and further clarifies that the tectum is not fused with the pedium
between the columellae. The nexine, or inner preserved layer of the spore wall, is
uniform in thickness except in the region of the trilete mark where some thickening is
present (PI. 97, fig. 7).

SCHOPFIPOLLINITES

The genus Schopfipollinites Potonie and Kremp 1954 (= Monoletes) has been used
for large (100-500 /Ltm) bilaterally symmetrical pollen grains of the prepollen type,
thought to be produced by medullosan pteridosperms. The grains are typically
characterized by a single proximal suture having a slight angular deflection (PI. 98,
fig. 2). On the distal surface two longitudinal grooves separated by a median ridge
(umbo) are occasionally present; however, the occurrence of this ridge does not
appear to be a constant feature of the taxon. Exine ornamentation as examined by
transmitted light is typically described as minutely granulose-reticulate, or smooth.
PI. 98, fig. 6 is a light photomicrograph of what appears to be the external orna-
mentation pattern of a grain extracted from the microsporangiate organ Halletheca
(Taylor 1971). The scanning electron micrograph of a Halletheca grain (PI. 98,
fig. 3) shows that the surface is highly variable in ornament, and that the apparent
reticulate appearance of the wall in PI. 98, fig. 6 is a feature of the internal organiza-
tion of the wall rather than of surface topography. In some grains, especially in the
region of the suture, exine deposition was not complete at the time of fossilization

EXPLANATION OF PLATE 98

Figs. 1-8. Schopfipollinites grams.

Fig. 1. Distal view of grain macerated from Dolerotheca fructification. x204.

Fig. 2. Proximal view showing suture with median deflection. x216.

Fig. 3. Portion of Halletheca grain showing internal lumina. Compare with organization illustrated
in Fig. 6 obtained with light microscopy. X 4950.

Fig. 4. Internal organization of exine of grain macerated from Schopfitheca reproductive organ, x 5600.
Fig. 5. Transmission scanning micrograph of grain showing region of proximal suture and distal umbo.

Note thickened wall in proximal suture region. x4000.

Fig. 6. Light photomicrograph of grain macerated from Halletheca sporangium, x 2500.

Fig. 7. Proximal surface of grain showing incomplete deposition of exine. x 965.

Fig. 8. Electron micrograph of grain macerated from Schopfitheca reproductive structure showing internal
exine organization. Compare with organization of grain in Fig. 5. x 5500.

PLATE 98

TAYLOR, Palaeozoic spores

774

PALAEONTOLOGY, VOLUME 16

SO that internal exine organization is easily visible without fracturing or sectioning
the spore wall (PI. 98, fig. 7). Grains of this type further point out the necessity of
a combined approach to the elucidation of spore wall features.

To determine the constancy of this form of exine organization and its potential use
as a taxonomic feature in dispersed spores, specimens of Schopfipollinites were
extracted from a number of medullosan pollen organs (Dolewtheca, Rhetinotheca,
Aulacotheca, Whittleseya) and compared with the organization found in Halletheca.
While a number of SchopfipoUinites-contain'mg reproductive organs have been
described, only very few are known from structurally preserved specimens. Conse-
quently, the opportunity accurately to delimit reproductive organs differing in pre-
servational mode by the identification of their spores is of considerable importance to
palaeobotanical systematics.

One such structureless reproductive organ containing Schopfipollinites grains is the
genus Schopfitheca (Delevoryas 1964). The specimen consists of a stalked, clavate-
pyriform microsporangiate structure approximately 20 mm long. Ultrastructurally,
the wall of the prepollen grain consists of a thickened foot layer from which arise
a series of anastomosing baculae (PI. 98, fig. 8). At some levels these units are fused
and demonstrate an organization similar to that in Halletheca or Dolerotheca grains
(PI. 98, fig. 5). The scanning electron micrograph of the fractured surface of one of
these Schopfitheca grains (PI. 98, fig. 4) indicates clearly a distinct difference in
internal exine organization from the Halletheca grain.

One approach to the problem of identifying stages in exine development is the use
of stereomicrography and photogrammetric analysis (Boyde 1970), which enables
one to view and accurately measure structural components of the wall. The capa-
bility of making stereo pictures with the scanning electron microscope not only
provides for an accurate three dimensional representation of the exine organization,
but provides a means of making precise parallax measurements of these three
dimensional structures. It has already been possible to correlate changes in exine
thickness with distinct differences in structural configuration, and to correlate these
in turn with grain maturity as determined by such additional features as level of
sporangium ontogeny. Preliminary information on the internal wall organization
of Schopfipollinites prepollen grains extracted from different reproductive organs,
as well as on those of differing preservational modes, appears to show that structural
differences in exine organization exist. Whether these differences represent stages in
wall ontogeny or are solely taxonomic must await continued investigation.

The internal organization of fossil pollen and spores may also provide important
information on other aspects of the plants that produced them ; for instance, the mode
of gamete emission in prepollen grains of the Schopfipollinites-type. Germinal exit
in such grains has generally been regarded as occurring from the proximal suture (see
Chaloner 1970 for an excellent review of this problem), although Renault (1896)
suggested evidence of distal germination in grains produced by Dolerotheca fertilis.
An examination of the proximal suture of a Dolerotheca grain (PI. 98, fig. 5) indicates
that although the total thickness of the grain wall is reduced by approximately one-
third, the floor and wall adjacent to the suture are distinctly thickened. The suture
would therefore seem an unlikely site for gamete emission. On the distal surface, how-
ever, the exine is appreciably thinner beneath each of the distal grooves, and super-

TAYLOR: ELECTRON MICROSCOPY OF SPORES

775

ficially appears a more probable exit site. Relative exine thickness, when compared
in a large number of Palaeozoic prepollen and pollen types, may provide cumulative
information on germination mode and evolutionary trends associated with the
process of fertilization.

CONCLUSION

The examples included in this paper illustrate the value of combined transmission
and scanning electron microscopy to the study of pollen grains and spores present in
reproductive organs. The SEM thoroughly delineates features of the internal and
external exine surfaces, and also provides a means whereby developmental features
may be critically studied. Transmission electron microscopy provides information
about the ultrastructural organization of the wall, as well as supplementing informa-
tion obtained by the SEM. There is little doubt that basic structural and functional
differences are present in the walls of spores and pollen grains produced by various
types of Palaeozoic vascular plants. Such differences may be correlated to provide
information on the evolutionary relationships between seemingly diverse taxa, as
well as providing a means of determining the biological origin of various dispersed
spores and pollen grains.

Acknowledgements. The author is indebted to Sheila D, Brack, Department of Botany, Ohio University,
and Michael A. Millay, Department of Biological Sciences, University of Illinois at Chicago Circle, and
to the American Philosophical Society and National Science Foundation (GB-35958) for financial
assistance.

REFERENCES

BOYDE, A. 1970. Practical problems and methods in the three-dimensional analysis of scanning electron
microscope images. Proc. 3rd Annual Scanning Electron Microscope Symposium, IITRI, 105-112.

CHALONER, w. G. 1970. The evolution of miospore polarity. Geoscience and Man, 1, 47-56.

DELEVORYAS, T. 1964. A probable pteridosperm microsporangiate fructification from the Pennsylvanian of
Illinois. Palaeontology, 7, 60-63.

EHRLICH, H. G. and HALL, j. w. 1959. The ultrastructure of Eocene pollen. Grana Palynologica, 2, 32-35.

GOOD, c. w. and taylor, t. n. 1970. On the structure of Cordaites felicis Benson from the lower Pennsyl-
vanian of North America. Palaeontology, 13, 29-39.

POTONiE, R. and g. o. w. kremp. 1954. Die Gattungen der palaozoischen Sporae dispersae und ihre Strati-
graphie. Geol. Jb. 69, 111-1 94.

RENAULT, B. 1896. Bassin Houiller d’Autun et d’Epinac. Etudes des Gites Mineraux de la France, fasc. IV,
text 578 pp. (Atlas published 1893).

SCHOPF, j. M., WILSON, L. R. and BENTALL, R. 1944. An annotated synopsis of Paleozoic fossil spores and the
definition of generic groups. III. Geol. Survey Kept. Invest. 91.

SWIFT, J. A. and brown, a. c. 1970. Transmission scanning electron microscopy of biological materials.
Proc. 3rd Annual Scanning Electron Microscopy Symposium, IITRI, 115-120.

TAYLOR, T. N. 1971. Halletheca reticulatus gen. et sp. n.: a synangiate Pennsylvanian pteridosperm pollen
organ. Amer. J. Bot. 58, 300-308.

1972. A new Carboniferous sporangial aggregation. Rev. Palaeo. Palyn. 14, 309-318.

WILSON, L. R. and coe, e. a. 1940. Descriptions of some unassigned plant microfossils from the Des Moines
series of Iowa. Amer. Mid. Nat. 23, 182-196.

T. N. TAYLOR
Department of Botany
Ohio University
Athens, Ohio, 45701
U.S.A.

Typescript received 25 October 1972

776

PALAEONTOLOGY, VOLUME 16

Discussion on Dr. Taylor’s paper:

Kempf: Dr. Taylor has extremely fine material from the Palaeozoic. In some of the spores it was quite
obvious that two layers have been preserved, a thin inner layer and a thicker outer layer, which occasionally
showed zonations. I would regard these layers as exine and perine. The saccus was always derived from the
outer layer — it was not between the two layers, but within the outer layer the perine. Did you also section
Equisetuml What is the fine structure of the wall?

Taylor: The wall of Equisetum is very nondescript with really no fine structure at all.

Kempf : You know the work of Gullvag, who is studying Equisetum. She has shown a tubular fine structure
(which may be artefactual). Perhaps there is some fine structure of this kind?

Taylor: I don’t know. I would like to comment on this perine problem. I’m not too concerned with what
we call the layers of the walls, because in the Palaeozoic we have very little to compare the structures with.
If we could compare with living plants, and show that the perine was present on the living form, then that
would be a different matter. The other point is, the perine is the layer that is put on last. It is a develop-
mental term really, and again, working with Palaeozoic material, we are not really in a position to put
names to the different layers. If one is to make such distinctions, they must be made at the biochemical
level, but at our present state of knowledge we are in no position to do this kind of work.

Kempf: At one time you were describing a wall with four layers; but there should be only two layers, and
three of these layers are zonations of the outer layer. Also the perine is not formed last, it is a primary
layer, and it is formed first, then the exine and then the intine. You can observe this optically, and some-
times you find spores with only the perine and the exine present and no intine.

Taylor: But how can you demonstrate that what you are calling the intine is not just another layer of the
exine without doing biochemical tests? It seems to me rather a matter of semantics.

Skelton: Have you carried your research back to Silurian plants— prespore plants— to see if you can get
any structure from them?

Taylor: No, I have restricted my work to the Upper Palaeozoic.

WALL STRUCTURE OF SOME AGGLUTINATED

FORAMINIFERIDA

by J. W. MURRAY

Abstract. Present knowledge of agglutinated wall structure and composition is briefly reviewed. Examination of
19 recent species shows the existence of three wall types: simple imperforate wall with an organic cement; complex
alveolar imperforate wall with an organic cement ; ‘perforate’ wall with a calcareous cement. It is concluded that more
attention should be paid to wall structure and composition in descriptions of species and in taxonomy. Forms with
a calcareous cement seem to be stenohaline marine or hypersaline and therefore useful indicators of environment.

Williamson (1858, p. xi) recognized and named the three main wall types seen in
recent foraminiferids : agglutinated, porcellaneous, and hyaline. Then followed the
classification by Carpenter, Parker, and Jones (1862) partly based on these characters,
together with the presence or absence of pores. Since that time wall structure has held
an important position in taxonomy.

Although there are many observations on agglutinated wall structure scattered
through the literature, there have been no recent detailed studies (see Lindenberg
1967 for a review). Investigations using electron microscopy have been concentrated
mainly on the porcellaneous and calcareous lamellar wall types. However, Jahn
(1953) and Towe (1967) published micrographs taken with transmission electron
microscopes and Murray (1971) has illustrated the surface texture of twenty-five
species in seventeen plates of scanning electron micrographs.

The purpose of this paper is to present a brief review of the present state of know-
ledge of agglutinated wall structure and to compare with this the results obtained
during the examination of nineteen recent species.

The present state of knowledge of agglutinated wall structure may be summarized
as follows:

1 . The wall consists of detrital particles held together by a cement secreted by the
animal.

2. Many different kinds of detrital particles including organic debris (see Thalmann
1948) are used by different species. Some seem to show no selectivity (e.g.
Reophax curtus. Smith and Kaesler 1970) while others are highly selective (see
Hedley 1964). However, it is not uncommon for the grain size to vary from one
part of the test to another or within the thickness of the wall (Lacroix 1931).

3. The cement may be entirely organic or it may be mineralized. Hedley (1963)
found the organic material to be ‘. . . an acid mucopolysaccharide (protein
linked with carbohydrate), with organically bound iron and, most probably,
organically bound calcium’. Mineralization may involve calcareous or
ferruginous deposits or both.

4. Cements mineralized with ferruginous material are known to contain the iron
as ferric oxide (Hedley 1963 ; Towe 1967) in a fine-grained amorphous condition
(Towe 1967).

[Palaeontology, Vol. 16, Part 4, 1973, pp. 777-786, pis. 99-100.]

1

778

PALAEONTOLOGY, VOLUME 16

5. Cements mineralized with calcareous material contain microgranular calcite 5
to 10 ixm in diameter (Wood 1949).

6. Pores, tubes, and alveolae have been recognized in some walls and these are
lined with a thin organic layer (Moebius 1880; N^rvang 1966). In most described
examples the pore tubes do not penetrate to the outer surface of the wall. The
pore tubes commonly branch (Lacroix 1939).

7. Thin organic membranes around the detrital particles have been recognized by
N^rvang (1966).

8. The ratio of detrital particles to cement is highly variable (Cushman 1929).

9. Some walls contain inter-grain spaces due to incomplete cementation (Barten-
stein 1952, p. 315).

Well-preserved specimens from Recent sediment were selected for study. Some
were examined whole, others broken to reveal internal structures and still others
were sectioned using the following method:

The specimens were placed on a metal stub (for use in the scanning microscope)
together with a small piece of ‘Lakeside’ thermoplastic cement. The stub was then
gently heated in a Bunsen flame to melt the Lakeside and allow it to penetrate into
and around the specimens. After cooling individual specimens were manipulated
into the desired orientation using a hot needle. Then they were carefully ground
away (under a stereoscopic microscope) using a finely ground glass slide lubricated
with water. The sections were then etched in 5% EDTA for periods ranging from
j to 10 minutes.

All specimens were prepared for examination in the scanning electron microscope
by coating them with a 40/60 mixture of gold/palladium in a vacuum coating unit.

The X-ray diffraction traces were made from bulk assemblages of each species.

The presence of ferric iron was inferred from the development of a prussian blue
colour in specimens treated with a solution of potassium ferrocyanide in hydrochloric
acid (2 gm in 100 mis of 1-75% HCl).

METHODS

MATERIAL

Species

Saccammina atlantica (Cushman)
Miliantmina fusca (Brady)

Cribrostomoides columbiense (Cushman)
Cribrostomoides crassimargo (Norman)
Cribrostomoides jejfreysii (Williamson)
Cyclammina cancellata (Brady)
Ammoscalaria pseudospiralis (Williamson)
Textularia earlandi (Parker)

Textularia sagittula (Defrance)

Textularia sp.

Siphotextularia flintii (Cushman)
Trochammina inflata (Montagu)
Trochammina lobata (Cushman)
Jadanunina macrescens (Brady)

Locality

Shelf off Long Island, U.S.A.
Christchurch Harbour, England.

Van Damme Beach, California.

Shelf off Long Island, U.S.A.

Western Approaches to English Channel.
Continental slope W. of English Channel.
Kattegat.

Celtic Sea.

Western Approaches to English Channel.
Shelf ofTTrucial Coast, Persian Gulf.
Celtic Sea.

Christchurch Harbour, England.

East coast, U.S.A.

Christchurch Harbour, England.

MURRAY; AGGLUTINATED FORAMINIFERIDA

779

Species

Gaudryina rudis (Wright)

Eggerella advena (Cushman)
Eggerella scabra (Williamson)
Clavulina pacifica (Cushman)
Martinottiella communis (D’Orbigny)

Locality

Western Approaches to England Channel.
Shelf off Long Island, U.S.A.

Western Approaches to English Channel.
Off Jeddah, Red Sea.

Western Approaches to English Channel.

RESULTS

Space limitations prevent a full description of each of the species studied so only
a few species will be described in detail.

Eorms having an organic cement. Saccammina atlantica has a unilocular test of variable form although it
is commonly pyriform with an aperture at the narrow end. The wall is made up of a detrital quartz grains
of variable size and shape (PI. 99, fig. 1). Much of the wall is built of larger grains which span the entire
thickness. These grains are closely fitted together with a separation of less than 1 jam along most of their
edges (PI. 99, fig. 2). However, complete fitting is not possible and the spaces are filled with a mosaic of
progressively smaller grains (PI. 99, figs. 4, 5). In these areas the wall is several grains thick. The surface of
the wall is rough both on the outside (PI. 99, figs. 1, 6) and on the inside (PI. 99, fig. 5). There are no wall
pores. Specimens placed in 5% EDTA or 1-75% HCl showed no reaction and it is concluded that the cement
contains no CaCOj.

Other species found to have a simple structure include Eggerella advena, E. scabra, Cribrostomoides
columbiense, C. crassimargo, C. jeffreysii, Textularia earlandi, Martinottiella communis, Trochammina
lobata and Miliammina fusca.

In Ammoscalaria pseudospiralis there is a clearly visible organic lining in the chambers (PI. 99, fig. 7).
This can also be seen in Trochammina inflata.

In many of these forms economy of cementation leads to small inter-grain spaces in the wall (first noted
by Bartenstein, 1952) but no evidence has so far been seen to suggest that these are in any way pores.

Cyclammina cancellata has a complex labyrinthic wall structure (see Banner 1970, for the most recent
description). The outer wall (epidermis) is imperforate and smoothly finished. The inner wall (hypodermis)
is coarsely alveolar. It is known from previous studies that the cement is organic, with iron mineralization
(Hedley 1963). In the present study it was found that the outer and inner wall surfaces and the linings of the
alveolae are all completely bound with cement. Within the thickness of the wall the grains are only loosely
cemented at their points of contact. No differences could be observed in specimens treated with acid so it
is concluded that there is no calcite cement. However, some specimens have coccoliths among their detrital
grains and the presence of such calcareous material may account for the calcium recorded in analyses by
Brady (1884), Faure-Fremiet (1911), and Vinogradov (1953). Hedley (1963) published an analysis of care-
fully cleaned specimens in which CaO was absent.

Forms having a calcareous cement. Clavulina pacifica starts with a triserial juvenile portion and then
becomes uniserial (PI. 100, fig. 1). The outer surface of the wall shows the presence of larger detrital grains,
including quartz, amphibole, and sponge spicules, and smoother areas of cement and fine detrital grains
(PI. 100, fig. 2). There are no pores penetrating the outer surface. By contrast the inner surface is smoothly
finished and shows many pores generally 2 to 3 pm in diameter and closed with an organic membrane
(PI. 100, fig. 3). Broken sections reveal that the pores extend almost through the wall but end blindly just
beneath the outer surface (PI. 100, fig. 4). Specimens impregnated with ‘Lakeside’, sectioned and etched
with 5% EDTA reveal the complexity of the pores. The latter, now filled with ‘Lakeside’, are seen to be
cylindrical tubes through much of the wall but they branch just beneath the outer surface (PI. 100, fig. 5).
Moreover, they appear to be lined with an organic layer which also extends between the pores as vertical
partitions (PI. 100, fig. 6). Pores also extend into the septa but do not penetrate to the apertural side.

Apart from the organic material observed in the wall, the cement consists of calcite (confirmed by X-ray
diffraction). It occurs as small units commonly less than 0-5 ^m in size (PI. 100, fig. 7) and sometimes as
elongate rods on the outer surface (PI. 100, fig. 8). Much of the inner part of the wall seems to consist of
calcite cement, the detrital grains being mainly in the outer part.

Brief etching with acid causes removal of some of the cement from the outer surface thus allowing the
ends of the pore-tubes to be seen. A similar perforate appearance of the test results from gentle abrasion.

780

PALAEONTOLOGY, VOLUME 16

Textularia sagittula shows similarities with Clavulina pacifica particularly in having a calcite cement and
blindly ending pore tubes. The detrital grains are mainly quartz and they are only loosely fitted together on
the outer surface, the intervening spaces being occupied with calcite cement (confirmed by X-ray diffrac-
tion). Sections impregnated with ‘Lakeside’ and etched in acid reveal an anastomosing network of pore
tubes which end blindly beneath the outer surface of the wall (PI. 99, fig. 8). The calcite cement occurs as
more or less equigranular grains 0-5 to 0-7 ;ixm in size.

Gaudryina rudis likewise has an agglutinated wall with a calcareous cement. The outer surface is rough,
due to detrital shell debris incorporated on the sides of the test, although the apertural face is smooth
(see Murray, 1971, pi. 14). The inner surface of the chamber side wall is perforated by pores 7 to
8 jum in diameter. These tubular pores end blindly beneath the outer surface of the wall although etched
and abraded specimens give the appearance of being perforate. The septa and apertural face lack pores
and tubes.

Other species having this type of wall with blindly ending pores are Textularia sp. from the Persian Gulf
and Siphotextularia flintii.

Ferric iron. With the exception of Cribrostomoides columbiense, Cyclammina cancellata, and Ammoscalaria
pseudospiralis, all the species were tested for ferric iron and all reacted positively although with different
intensities. In the case of the forms with calcareous cements the test was destroyed by the acid solution
and the colouration was developed in the residual organic framework.

DISCUSSION

The results presented here agree, in general, with those of previous workers, but
there are some differences.

The presence of pores in Textularia was first described by Carpenter, Parker, and
Jones (1862, p. 191). A more complete description by Moebius (1880) recorded the
presence of a ‘chitinous’ lining in the chambers and in the pore tubes. He also noted
that the pores reached the surface only in the younger chambers. Lacroix (1931)
studied the same species as Moebius, T. agglutinans d’Orbigny. He observed that the
pore tubes bifurcate close to the outer surface and stated that they opened on to the
surface through very small pores. He also noticed that the inner organic lining covered
the pores as well as the chamber wall. Reyment (1969) recognized ‘ultrapores’ in
Textilina mexicana (Cushman) but he gave no information about the passage of
these pores through the wall.

In the present study no evidence could be found of pore openings at the outer
surface except where the test had clearly suffered abrasion or etching with acid. In
Clavulina pacifica and Gaudryina rudis the pore-bifurcations beneath the wall surface
have a diameter of approximately 1 p.m and if they opened on the surface they should
be clearly visible. It seems certain that they are closed either by an organic membrane
or by calcite cement.

EXPLANATION OF PLATE 99

Figs. 1-6. Saccammina atlantica (Cushman) 1, General view, x 120. 2, close fit of two quartz grains,
X 1150. 3, The aperture, showing large and small quartz grains, x750. 4, Close fit of small grains
between large detrital grains, x 1500. 5, Detail of inner side of wall showing small grains filling the
gaps between the larger grains, x650. 6, Broken section of wall, X 1270.

Fig. 7. Ammoscalaria pseudospiralis (Williamson) showing organic lining (o) on inside of wall, X 1300.

Fig. 8. Textularia sagittula Defrance. Impregnated and etched section of wall showing anastomosing pore
tubes (p) and quartz grains (q) on the outer side, x 1400.

PLATE 99

MURRAY, agglutinated foraminifera

782

PALAEONTOLOGY, VOLUME 16

N0rvang (1966) studied Textularia sagittula Defrance and found the walls to be
imperforate. Lacroix (1931) studied what he called the same species and found the
pores to be present but small (1 /nm in diameter).

The data on pores in agglutinated walls with a calcareous cement is summarized
below :

1 . They may be tubular with bifurcations near the outer wall surface or they may
form an anastomosing network throughout the wall.

2. They end blindly just beneath the outer wall ; no definite pore openings have been
observed on the outer surface in the present study.

3. The pores are lined with a thin organic membrane.

4. They are closed with an organic membrane on the inside of the chamber and
possibly also at the outer ends.

5. They are normally developed mainly in the chamber walls on the sides of the
test. They are less well developed or are absent in the apertural face and in the
septa.

6. In Clavulina pacifica the pores form 25% of the volume of the wall.

The pores of agglutinated walls were considered by Reiss (1963) to be different
from those of calcareous lamellar foraminiferids because they are curved. Perhaps
the most distinctive features are that they commonly branch and sometimes anasto-
mose. In calcareous lamellar foraminiferids the pores are maintained even during the
addition of further wall layers (see Hansen and Reiss 1971 for illustrations). By
contrast, in agglutinated foraminiferids the wall is non-lamellar and each chamber
wall is secreted as a single event.

All the forms with pores seem to be stenohaline marine or hypersaline species
which are presumably in osmotic equilibrium with their environment. The forms
with an organic cement not mineralized with calcareous material lack pores.
Marszalek, Wright, and Hay (1969) have suggested such a test ‘. . . offers good refuge
to the foraminifer under times of stress, and allows time for osmoregulatory adjust-
ment to the new conditions’. This could account for the presence of such forms in
hyposaline environments.

The wall of Cyclammina cancellata presents a special case. Although the hypo-
dermis contains alveolae these are in complete communication with the chamber
lumen and they are not crossed by an organic membrane at the inner ends. They do
not therefore compare with the pores discussed above and the wall of Cyclammina
must be regarded as imperforate.

EXPLANATION OF PLATE 100

Figs. 1-8. Clavulina pacifica Cushman. 1 , General view, X 70. 2, Detail of wall texture showing detrital
grains (g), x220. 3, Inner side of wall with pores covered by an organic membrane, x4000. 4, Broken
section of wall showing the pore tubes ending blindly at the outer side, x 1350. 5, 6, Impregnated and
etched section of wall showing the pore tubes bifurcating and ending blindly below the outer surface,
X 1400, in 5 and transverse sections of the pore tubes and associated organic membranes, x670, in 6.
7, Broken section of wall showing closely packed pore tubes built mainly of cement, X 3400. 8, Detail
of outer wall surface showing elongate cement crystals between detrital grains, x 3400.

PLATE 100

MURRAY, agglutinated foraminifera

784

PALAEONTOLOGY, VOLUME 16

WALL STRUCTURE AND CLASSIFICATION

In the classification adopted in the Treatise (Loeblich and Tappan 1964) nearly all
agglutinated forms are placed in the suborder Textulariina. This includes the super-
family Ammodiscacea, in which the wall is said to be ‘agglutinated, simple, or
labyrinthic’, and the superfamily Lituolacea, ‘wall siliceous or agglutinated, with
calcareous, siliceous, or ferruginous cement’.

At lower taxonomic levels the information is often lacking in detail or at variance
with the present observations. For the Saccamminidae the wall is not mentioned. The
family Lituolidae has ‘wall agglutinated, with calcareous cement or microgranular
calcite, interior simple to labyrinthic, epidermal layer imperforate’. Genera of this
family which have been found in the present study not to have a calcareous cement
include Cribrostomoides, Cyclammina, and Ammoscalaria. The family Textulariidae
has ‘wall agglutinated’. For Textularia it is said to be ‘simple’. The wall of Sipho-
textularia is not described. The family Trochamminidae just has ‘wall agglutinated’.
The same is true of the Ataxophragmiidae. Gaudryina has no wall description,
Eggerella has ‘wall finely agglutinated on pseudochitinous base, may be of calcareous
particles in calcareous cement’. This disagrees with the results presented here.
Clavulina has ‘wall agglutinated with much calcareous cement’. Clavulina pacifica
agrees with this. Martinottiella has ‘wall finely agglutinated’. Thus, of the four
Ataxophragmiidae examined two have an organic cement and no pores {Eggerella
and Martinottiella) and two have a calcareous cement and pores {Gaudryina and
Clavulina).

This raises the question of the value of the wall structure and cement composition
as taxonomic features. In a general sense the agglutinated wall structure is clearly
useful. However, should more notice be taken of the detailed structure? Since the
cement is secreted by the foraminiferid it must surely be of greater taxonomic value
than the nature of the detrital particles gathered from the sediment. The kind of
cement is a reflection of the physiology of the animal. This must be at least of equi-
valent importance to the nature of coiling or the arrangement of the chambers.
Unfortunately, in the majority of descriptions of agglutinated species a full descrip-
tion of the wall structure and composition is omitted.

ECOLOGICAL SIGNIFICANCE

Foraminiferids with agglutinated walls are known to be particularly common in
the deep sea, in cold shelf seas, and in shallow and intertidal hyposaline waters.
There is now clear evidence that the nature of the cement is of ecological significance.

Pokorny (1958) suggested that forms with an organic cement characterize cold
water. Lindenberg (1966) inferred that in the Dogger (Jurassic) of south west Ger-
many the forms with a calcareous cement lived in more marine waters than those
with an organic cement.

The results of the present study support Lindenberg’s view. All the examined
species having a calcareous cement come from normal marine or hypersaline environ-
ments (Western Approaches to the English Channel, Celtic Sea, Persian Gulf, Red
Sea). Those with a simple wall and an organic cement are found in hyposaline marshes
and lagoons {Trochammina inflata, Jadammina macrescens, Miliammina fu.sca).

MURRAY: AGGLUTINATED FORAMINIFERIDA

785

hyposaline shelf seas {Saccammina atlantica, Cribrostomoides crassimargo, Eggerella
advena), and normal marine shelf seas {Martinottiella communis, Textularia earlandi,
Eggerella scabra). Thus they occur in many different environments. The only form
with a complex alveolar wall studied here is characteristic of the continental slope
with sigma-t values of 27-7 (see Banner 1970, p. 244).

The restriction of forms with a calcareous cement to normal marine and hyper-
saline shelf seas should prove useful in helping to interpret the palaeoecology of
fossil assemblages.

CONCLUSIONS

The nineteen species discussed here are hardly representative of the hundreds of
agglutinated species but the results nevertheless have some interest. Clearly there are
more types of agglutinated wall structure to be discovered. At the taxonomic level
many more species need to be investigated in detail and then it will be possible to
emend generic and family descriptions and groupings. From the ecological point of
view the nature of the cement appears to be important and controlled laboratory
experiments should be carried out to gain further knowledge.

Acknowledgements. I am grateful to N.E.R.C. for a research grant to cover the cost of using the scanning
microscope and to Professor H. Hinton for kindly allowing access to the microscope. Thanks are due to
R. Godwin for printing the photographs from negatives taken by myself, and to G. Day for carrying out
the X-ray analyses. Dr. H. J. Hansen kindly read the manuscript.

REFERENCES

BANNER, F. T. 1970. A synopsis of the Spirocyclinidae. Revista Espanola de Micropaleontologia, 2, 243-290,
pis. 1-14.

BARTENSTEiN, H. 1952. Taxonomischc Bemerkungen zu den Ammobaculites, Haplophragmium, Lituola und
verwandten Gattungen (For.) Senckenbergiana, 33, 313-342, pis. 1-7.

BRADY, H. B. 1884. Report on the foraminifera dredged by HMS ^Challenger during the years 1873-1876.

Challenger Exped., 1873-76, Rept., London, 9, 1-814, pis. 1-115.

CARPENTER, w. B., PARKER, w. K. and JONES, T. R. 1862. Introduction to the study of the Foraminifera. 319 pp.,
22 pis. Ray Soc., London.

CUSHMAN, J. A. 1929. The term ‘arenaceous Foraminifera’ and its meaning. Contr. Cushman Lab. foramin.
Res. 5, 25-27.

FAURE-FREMiET, E. 191 1. La Constitution du test chez les foraminiferes arenaces. Bull. Inst, oceanogr. Monaco,
216, 1-7.

ELANSEN, H. J. and REISS, z. 1971 . Electron microscopy of rotaliacean wall structures. Bull. geol. Soc. Denmark,
20, 329-346, pis. 1-21.

HEDLEY, R. H. 1963. Cement and iron in the arenaceous Foraminifera. Micropaleontology, 9, 433-441, pi. 1.
1964. The biology of Foraminifera. Int. Rev. Gen. Exper. Zool. 1, 1 45.

JAHN, B. 1953. Elektronenmikroskopische Untersuchungen an Foraminiferenschalen. Z. Wiss. Mikrosk.
61, 294-297.

LACROIX, E. 1931. Microstructure du test des Textulariidae. Bull. Inst, oceanogr. Monaco, 582, 1 18.
LINDENBERG, H. G. 1966. Untersuchungen an lituoliden Foraminiferen aus dem SW-deutschen Dogger,
1. Ammopalmula n.g. und Ammobaculites Cushman 1910. Senckenberg. leth. 4,1, 461-78.

1967. Gehause aus Sand bei einzelligen Tieren. Natur Mus., Frankf. 97, 244-258.

LOEBLICH, A. R., JR. and TAPPAN, H. 1964. Sarcodina chiefly ‘Thecamoebians’ and Foraminiferida, in
MOORE, R. c., ed.. Treatise on invertebrate paleontology. New York, Geol. Soc. Amer. and Univ. Kansas
Press, pt. C, Prostista 2, 1, Cl -501a.

MARSZALEK, D. s., WRIGHT, R. c. and HAY, w. w. 1969. Function of the test in Foraminifera. Trans. Gulf
Coast Ass. geol. Socs. 19, 341- 352.

786

PALAEONTOLOGY, VOLUME 16

MOEBius, K. 1880. Foraminiferen von Mauritius. In moebius, k., richter, f. and von martens, e. Beitrdge
zur meeres Fauna der Inseln Mauritius und der Seychellan. 65-112, pis. 1-14. Berlin.

MURRAY, J. w. 1971. An atlas of British recent foraminiferids. 244 pp., 96 pis. Heinemann, London.
N0RVANG, A. 1966. TextUma nov. gen., Textularia Defrance and Spiroplectammina Cushman (Foraminifera).
Biol. Skr. 15, 1-16, pis. 1-2.

POKORNY, v. 1958. Principles of Zoological Micropaleontology. (Transl. allen, k. a., 1963, 652 pp.
Pergamon.)

REISS, z. 1963. Reclassification of perforate Foraminifera. Bull. Israel geol. Surv. 35, 111 pp., 8 pis.
reyment, r. a. 1969. Textilina mexicana (Cushman) from the western Niger Delta. Bull. Geol. Instn Univ.
Uppsala, N.S., 1, 75-81, pis. 1-4.

SMITH, M. A. and KAESLER, R. L. 1970. Selection of adventitious test material by Reophax curtus (Foramini-
ferida). J. Paleont. 44, 953-957.

THALMANN, H. E. 1948. Mitteilungen fiber Foraminiferen VII. 30. Organisches Baumaterial der sands-
chaligen Foraminiferen. Eclog. geol. Helv. 41, 366-368.

TOWE, K. M. 1967. Wall structure and cementation in Haplophragmoides canariensis. Contr. Cushman Fdn
foramin. Res. 18, 147-151.

VINOGRADOV, A. p. 1953. The elementary composition of marine organisms. Mem. Sears Fdn mar. Res.
2, 1-647.

WILLIAMSON, w. c. 1858. On the Recent Foraminifera of Great Britain. 107 pp., 7 pis. Ray. Soc., London.
WOOD, A. 1949. The structure of the wall of the test in the Foraminifera; its value in classification. Q. Jl
geol. Soc. Lond. 104, 229-255, pis. 13-15.

J. W. MURRAY
Department of Geology
The University

Typescript received 25 October 1972 Bristol, BS8 ITR

DisciLssion on Dr. Murray’s paper:

Green; Is there any organic membrane lining the pores?

Murray : Yes. This is something which I omitted to mention in the talk. I have some additional micrographs
which show organic walls running between the pores perpendicular to the chamber surface, so the whole
of the inside of the wall has a mesh-work of organic material, and this presumably replaces the calcitic
cement.

Green: Filling the gaps in?

Murray: Or else the opposite; it is trying to reduce the density of the wall, because the density of calcite is
great with respect to sea water.

Rood : Is the cement of such a nature that it could conduct an ion flow through the cement into the inner
structures?

Murray: I do not think that any of the work I have done so far could prove this, but it seems strange if
the animal felt the need to transport material through the cement, when it has tubular pores in the wall.
The odd thing about these pores is that they only occur in the sideways facing parts of the test— they
don’t occur in the final face of the test, the one that is held down to the sea floor where the animal is living.

Daniels: Perhaps in life the grains over the ends of the pores are loose, and the animal can create some
sort of exit.

Murray: I have taken many pictures of the outer surface, and the cement always comes round the grains
holding them in place very firmly.

Sylvester Bradley: How do you think the cement got there?

Murray: When the animal secretes its skeleton, there would have to be places where the pseudopodia
came out.

Sylvester Bradley; You don’t think that these pores were where the pseudopods came out?

Murray: I must admit that that is a good point. But one has to ask why the pseudopods covered the ends
of the pores with a membrane after withdrawing into the chamber.

TRANSMISSION ELECTRON MICROSCOPY OF

FOSSIL SPORES

by E. K. KEMPF

Abstract. Transmission electron microscopy of fossil spores presents special difficulties. A major problem in
megaspores is obtaining low power pictures of whole spores; this is overcome by using single-hole discs instead of
grids. Care is needed to avoid damaging sporoderm ultrastructure by oxidation during preparation. Results achieved
since 1965 are reviewed, and the sporoderm fine structures of Setosisporites (Carboniferous) as well as Salvinia and
Alms (both Tertiary) are illustrated and interpreted.

Seven years ago, in May 1965, the Department of Geology at the University of
Cologne was provided with a transmission electron microscope, and a program was
started to investigate the stratification and fine structure of fossil sporoderms in
ultra-thin sections.

At that time there were only two papers published on this subject. Ehrlich and
Hall (1959) studied some Eocene pollen grains without knowing, however, to what
genera the material belonged. Pettitt and Chaloner (1964) tried to elucidate the fine
structure of Mesozoic microspores from pollen sacs of Cheirolepidium muensteri,
which are of the Classopollis type. Both publications demonstrated that it is possible
and worth while to study fossil material in this manner.

TECHNICAL PROBLEMS AND SOLUTIONS

Our investigations were carried out step by step. The first task was to ascertain that
the fine structure of sporoderms had been preserved in the fossil state without any
change during fossilization. For this, comparison was necessary between the fine
structural details of Recent and fossil sporoderms of the same species, and selection
of suitable material turned out to be quite difficult.

Most of the papers published since 1952 on transmission electron microscopy of
Recent sporoderms dealt with microspores, and especially with pollen grains. In
preparing the material for such investigations, anthers from living or herbarium
plants were reduced to small pieces, fixed, dehydrated, embedded in epoxy resins, and
ultra-thin sectioned. There are, in comparison, only a few cases where fossil anthers,
or parts of them, are found that may be treated in this way; for example Classopollis
(Pettitt and Chaloner 1964) or Alnus (this paper). Most fossil microspores, however,
occur dispersed and because of their extremely small size are not easy to handle.
While the latter problem can be mastered using a micromanipulator, the main diffi-
culty still remains: to determine the plant species in which the single microspore
originated. In consequence of this uncertainty and as fine structural details are not very
numerous in sporoderms of microspores, we decided to turn towards megaspores.

Cenozoic megaspores, which had been collected for biostratigraphical reasons, as
well as Recent megaspores to serve as comparative material, were at hand. For the
task in question Recent and Pleistocene megaspores of Salvinia natans seemed to be

[Palaeontology, Vol. 16, Part 4, 197.3, pp. 787-797, pis. 101-103.]

788

PALAEONTOLOGY, VOLUME 16

most suitable. They were subjected to the usual sample preparation techniques and
ultra-thin sectioned. The transmission electron micrographs gave a positive answer
to our question. Sporoderm fine structure was preserved, and apart from some dif-
ferences in electron density there was no change observable which could have been
caused by fossilization. Other megaspores from Cenozoic, Mesozoic, and Palaeozoic
strata were subsequently studied and corroborated the result.

The illustration of the findings for publication posed new difficulties. Because of
the formvar film coated copper grids, which were used to support the ultra-thin
sections, up to 50% of the large megaspore sections (about 0-5 mm in diameter) were
hidden. In many of the ultra-thin sections it was possible to reveal the stratification
and fine structure of a megaspore sporoderm, but the photographs are not suitable
for publication. The paper of Pettitt (1966), which had been published in the mean-
time, also suffered from this difficulty. Indeed in most papers that dealt with trans-
mission electron microscopy of sporoderms very tiny areas were figured at large
magnifications, while figures at low magnifications giving a general view were miss-
ing or presented only as non-equivalent photo micrographs.

To give the maximum amount of information, best presentation and interpretation
of sporoderm stratification and fine structure it seemed absolutely necessary to cover
the whole field from very low up to the highest magnifications with transmission
electron micrographs. Therefore we changed from copper grids to copper discs with
one single hole of 400, 800, or even 1000 ju.m in diameter. To support the large ultra-
thin sections it is necessary to reinforce the thin formvar films with which the holes
were covered. It is then possible due to this method to get transmission electron
micrographs at a magnification of about x90 with moderate resolving power, at
magnifications of about x450 and x 1700 with better resolving power, and above
X 6000 with high resolving power. Thus micrographs of single megaspore specimens
in transmission electron microscopy can be produced, which nowadays, of course,
should be supplemented by scanning electron micrographs.

Another difficulty is to find the best embedding medium for fossil spores. Because
of the hardness of fossil sporopollenin and the necessity of good impregnation we
use traditional methacrylate, in the formulation given below.

A further question is whether to stain the specimens or not. Until now we have
avoided staining apart from some trials in order not to complicate our studies. In
the future it will be necessary perhaps to use different stains, especially in micro-
spores. Unstained specimens, should, however, always be studied for comparison.

We consider the demonstration that fossil sporoderms are often preserved in an
excellent state as a main result of our studies. If however, one follows standard micro-
palaeontological or palynological techniques in which oxidizing chemicals are used, a
certain amount of damage may occur. A most careful separation of the fossils from the
embedding sediments is therefore required. In order to clean the specimens selected
for study by the transmission electron microscope cold hydrofluoric acid (40%) is used,
followed by washing with hot hydrochloric acid (25%) and distilled water.

Recapitulating, one can say that transmission electron microscopy of fossil spores
at least requires the following treatment:

1. Careful separation from the embedding sediments, if possible without making
use of oxidizing chemicals, in order to prevent damage.

KEMPF; TRANSMISSION MICROSCOPY OF SPORES

789

2. Cleaning of the selected specimens by using cold hydrofluoric acid (40%),
followed by washing with hot hydrochloric acid (25%) and distilled water.

3. Dehydration in a graded ethanol series followed by embedding in a 1 : 9 mixture
of butyl/methyl methacrylate, containing 1% benzoyl peroxide. If different
resins are used as embedding media, another kind of dehydration might be
necessary.

4. After polymerization, cutting of ultra-thin sections on an ultra-microtome
using glass and diamond knives. For magnifications up to x 10 000 mounting
of ultra-thin sections should be done on formvar-film coated single-hole copper
discs; for larger magnifications copper grids with or without film coating may
result in higher resolving power.

STRATIFICATION AND FINE STRUCTURE OF FOSSIL SPORODERMS

Most of the results have been obtained from fossil megaspores, but some charac-
teristic features of microspores are also presented.

For the interpretation of sporoderm stratification and fine structure it is neces-
sary to make comparative studies of Recent spores in ultra-thin sections. Until now
the resulting transmission electron micrographs never revealed more than three
layers, which in my papers are named intine, exine, and perine, from inside to out-
side. It should be mentioned here that my terms exine and perine do not correspond
with such terms of some other authors.

Cenozoic filicopsid megaspores and microspores. Most of the megaspores which have
been found in Cenozoic strata belong to genera of the heterosporous ferns. Many
species have been studied in ultra-thin sections, such as Azolla (Kempf 1969 a, b) and
Salvinia (Kempf \91\b).

In fossil Azolla megaspores, two layers are preserved. The inner one, the exine,
is quite thick, electron dense, and restricted to the rounded distal half of the mega-
spore. The outer layer, the perine, surrounds the exine. At the proximal pole it forms
a large gula, to which the floats adhere via threads. The number of floats differs from
subgenus to subgenus.

When, at larger magnifications, megaspore sporoderms of different species are
compared with each other, the exines look quite similar. The fine structure of the
perine, however, is heterogeneous and changes from species to species. It is there-
fore very useful for identifications on the species level. Within one species, the perine
fine structure changes according to a zonation. In Miocene Azolla nana the structure
resembles the differentiation into foot layer, bacula, and tectum of certain pollen
grains. Furthermore there are fine structural differences between the distal and the
proximal part, so that around the gula, the perine is quite different from that in the
distal part of the megaspore. In Azolla, the intine is not preserved in the fossil state
because it consists mainly of cellulosic material.

Hitherto, fossil intine has only been found in the megaspores of some species of
Salvinia, e.g. Salvinia natans. The electron density of this material suggests that its
preservation was made possible by a certain content of sporopollenin. The exine
fine structure, as in Azolla, does not vary very much from species to species. The
perine is again characterized by its zonation and great variability of fine structure.

790

PALAEONTOLOGY, VOLUME 16

Sometimes even the remains of the sporangioderm, which enveloped the megaspore,
are preserved.

From Salvinia rhenana megaspores, we learned that the sporoderms can be
damaged where oxidizing chemicals are used in the laboratory. Where H2O2 had
been used to disintegrate the sample, the intine had disappeared and the exine
revealed initial signs of corrosion. New samples, treated without oxidizing chemicals,
released megaspores in which the exine and intine were very well preserved.

The Miocene Salvinia cerehrata was preserved as complete sori, which, after break-
ing open, yielded megaspores and microsporangia. A longitudinal section of a mega-
spore shows the dense exine and the general arrangement of the perine fine structure,
which is most complicated at the proximal pole (PI. 101, fig. 3). There the delicate
gula is hidden by three large germinal valves. The scanning electron micrograph gives
a three-dimensional impression of this region (PI. 101, fig. 4). The perine fine struc-
ture resembles that of Halletheca from the Carboniferous (Taylor 1971).

Ultra-thin sections of complete microsporangia disclosed the characteristics of
the microspores. In Salvinia all microspore exines of a microsporangium share a
common perine, which in its fine structure resembles that of the corresponding mega-
spore. Within the perine mass the exines are arranged in large cavities near the surface,
each of which is provided with a triradiate germination mark. The exines are very
homogeneous, electron dense and therefore poor in fine structure. With the excep-
tion of a slight thickening towards the germinal suture, there is no sculpture on the
exine surface.

Cenozoic angiosperm pollen grains. As an example of fossil angiosperm pollen grains,
pollen sacs of Alnus were embedded and ultra-thin sectioned. The material was col-
lected from a clay lens within the lower-most part of the main seam of the Rhenish
brown coal (Kempf \91\b), where it was found together with Salvinia cerebrata,
amongst other plant fossils. The scanning electron micrograph of a piece of a pollen
sac (PI. 103, fig. 1) clearly demonstrates that these pollen grains represent the genus
Alnus. As a dispersed spore, this type is known as AlnipoUenites metaplasmus (Potonie
1931) Potonie 1960.

The transmission electron micrographs of ultra-thin sections (PI. 103, figs. 2-4)

EXPLANATION OF PLATE 101

Megaspores of Salvinia cerebrata. Lower Miocene, W. Germany. Figs. 1, 2, 4: scanning electron micro-
graphs; fig. 3: transmission electron micrograph.

1. Distal view; spore surface irregularly corrugated (cerebral sculpture), x 100.

2. Proximal view; triradiate gula with germination mark largely hidden by the three germinal valves,
xlOO.

3. Ultra-thin longitudinal section of proximal part showing two germinal valves and one ray of gula
with germinal suture; exine (EX) relatively thin and quite dense; perine (PE) very thick and structurally
subdivided in three zones (inner, middle, and outer zone); outer zone consisting of large, middle zone
of small cavities ; felt-like inner zone somewhat stretched because of laboratory treatment ; x 675 (E 1 0009
10012, B 661, S 36393-2).

4. Proximal part of longitudinal half seen from inside; thin, dense exine showing triradiate germination
mark on inner side; perine presenting two germinal valves and one ray of gula; spongy fine structure
zonally differentiated, x675.

PLATE 101

KEMPF, Miocene megaspores

792

PALAEONTOLOGY, VOLUME 16

resemble those of Recent pollen grains of Alnus (Takeoka and Stix 1963). The sporo-
derm consists of two layers. The inner layer, the exine (= secondary exine, endexine,
or nexine of other authors), is quite thin and electron dense. It is doubled in thickness
over the arci, but looks like a porous membrane below the apertures. The perine
(= primary exine, ektexine, or sexine of other authors) is formed by bacula and
tectum. The bacula arise from the exine at irregular intervals, but are most numerous
in the range of the arci. The coherent tectum is equatorially penetrated by the four
large germinative pores and elsewhere by a great number of minute tubes (perpendi-
cular bright lines). At each pore a vestibulum is formed by an arching upwards of the
tectum. At the base of the bacula, exine and perine are fused together. There seems
to be no foot layer; a condition recently described also for the Upper Cretaceous
pollen grain Wodehouseia spinata (Leffingwell et al. 1970).

Mesozoic lycopsid megaspores and microspores. In Mesozoic nonmarine sediments,
lycopsid megaspores are rather common microfossils. One of the first form species
to be studied by us in ultra-thin sections was ' Horstisporites’’ semireticulatus (Kempf
\91\a). In this megaspore the sporoderm consists of a very heavy outer layer com-
posed of ramifying sporopollenin threads — which was named perine, and a very thin
laminated inner layer— which was named exine. Compared with the information at
that time available about the sporoderm fine structure of Recent Selaginella mega-
spores (Martens 1960; Stainier 1965, 1967), this nomenclature seemed to be wrong.

It was necessary therefore to study some more Recent Selaginella megaspores
(Kempf 1970), and this indicated that the naming of the layers in the fossil mega-
spore had been correct. In Selaginella the perine in fact is a very heavy layer with
a variety of fine structure of different kinds, which mostly is arranged in concentric
zones. The exine on the other hand is represented by a quite thin and laminated layer.
The intine cannot be expected to be found in the fossil state since it is mainly com-
posed of cellulosic substances.

Subsequently other Mesozoic megaspore species were examined in ultra-thin
sections (Kempf 1972). A very thin laminated exine and a quite heavy perine, with

EXPLANATION OF PLATE 102

Figs. 1, 2. Megaspore Setosisporites brevispinosus, Namurian, Poland.

Fig. 3. Megaspore Setosisporites hirsutus, Westphalian B, W. Germany. All transmission electron micro-
graphs.

1. Ultra-thin longitudinal section of compressed specimen; proximal part with one ray of germinal suture
(GS) and very thin exine which has loosened from inner side of perine; distal part with short spines rising
from outer surface; inner third of perine more dense than outer two-thirds, indicating some kind of zonal
differentiation in fine structure, x 340 (E 9329, B 740, S 36398-P 5).

2. Distal part of sporoderm in ultra-thin section; exine very thin and dense; it is doubled in this place as
the exine cavity has collapsed; perine very thick with a fine structure formed by sporopollenin threads;
a certain zonation in fine structure caused by variations in diameter or in main orientation; the spine
regularly arises from the outer zone, x 6300 (E 9348, B 740, S 36398-Q 1 ).

3. Ultra-thin section of sporoderm; exine extremely thin; perine thicker by far with a zonal fine structure
formed by sporopollenin threads; within inner and middle zone the threads are concentrically arranged
but differing in diameter; within the large outer zone the threads form an irregular network the free
space of which has partly been filled with an unknown substance during fossilization or laboratory treat-
ment, X 5780 (E 8857, B 621, S 36397-G 4).

PLATE 102

KEMPF, Carboniferous megaspores

794

PALAEONTOLOGY, VOLUME 16

fine structure formed by ramifying sporo-pollenin threads, were found in nearly all
of these sporoderms.

Microspores were adhering to the surface of some of these megaspores and thus
were sectioned by chance. They also consist of two layers— exine and perine— but
it is not possible to detect any relationships to the most likely corresponding mega-
spore. Pettitt (1966) made the same observation when he compared the fine structures
of Recent Selaginella pulcherrima megaspores and microspores.

In one type of megaspore a single thick layer only was encountered (Jux and Kempf
1971). Because the fine structure, which consists of radial tubes, differs from that of
all other previously known megaspore sporoderms, a new form genus was created.
This type of megaspore may perhaps represent some kind of Calamitacean plant.

Palaeozoic megaspores. The number of megaspore forms known from Palaeozoic
deposits is very large. At the moment, however, it is quite difficult to find megaspores
which are in such a good state of preservation that they are really suitable for ultra-
thin section studies. We have attempted this with megaspores of Setosisporites
hirsutus (PI. 102, fig. 3) and Setosisporites brevispinosus (PI. 102, figs. 1, 2). In fine
structure they are similar to some of the Mesozoic megaspores, but it is obvious that
the exine is extremely thin, while the perine shows an enormous thickness.

GENERAL TRENDS AND FUTURE WORK

It can be demonstrated that transmission electron microscopy of fossil spores
provides us with a large amount of new and valuable information. Details, previously
unknown or misinterpreted because of the limitations of optical microscopy, are
revealed.

Of course most of the new information is relevant only to a single genus or species.
There are, however, also general trends. One of these is the observation that in mega-
spores there is an obvious decrease in the ratio of perine to exine thickness, which is
related to the state of evolutionary development of the megaspore (Kempf 1972).

EXPLANATI06J OF PLATE 103

Pollen grains of Alnus ("AlmpoUenites metaplasmus'). Lower Miocene, W. Germany.

Fig. 1. scanning electron micrograph; figs. 2-4: transmission electron micrographs.

1. Part of pollen sac; all pollen grains with four equatorial apertures, which are connected with each other
by two quite distinct arci; sporoderm surface provided with tiny spines, x2000.

2. Ultra-thin section of pollen grain; sporoderm consists of exine and perine; exine very thin, dense and
doubled in thickness in the area of the arci (AR), but like a porous membrane below the apertures;
perine is formed by bacula and tectum; bacula rising from the exine at irregular intervals, most numerous
in the range of the arci; the otherwise coherent tectum is penetrated only by the four large pores and by
a great number of minute tubes (perpendicular bright lines); at each major pore a vestibulum is formed
by arching upwards of the tectum, x 5500 (E 9899, B 771, S 6238 1-J 5).

3. Ultra-thin section of pollen grain, in which two pores and four arci were met with; x3360 (E 9871,
B77LS62381-J 2).

4. Ultra-thin section of pollen grain; sporoderm consists of exine and perine which at the base of the
baculae are fused together; foot layer seems to be missing; perine is penetrated by minute tubes (per-
pendicular bright lines); x28 000 (E 9897, B 771, S 62381 J 5).

PLATE 103

KEMPF, Miocene megaspores

796

PALAEONTOLOGY, VOLUME 16

One of the most important general results is a better knowledge of sporoderm
stratification. It becomes apparent that all sporoderms comprise three layers : intine,
exine, and perine. In the fossil state normally only two of them are preserved; exine
and perine, with Salvinia as the single known exception. The fossilization of sporo-
derm layers depends on their content of sporopollenin. Where this is lacking the layer
will also be missing in the fossil record.

The knowledge of sporoderm stratification, mainly obtained from megaspores,
also can be applied to pollen grain sporoderms. There the exine has hitherto been
named endexine, nexine, or secondary exine, while the perine was described as
ektexine, sexine, or primary exine. In pollen grains it is sometimes not easy to recog-
nize where the exine ends and the perine begins, as these two layers are often fused
together as in Alnus (PI. 103, fig. 4). There are, however, differences depending on
function, which become most distinct in the range of the germinal openings.
Apparently sporoderm stratification and fine structure are mainly subject to func-
tional requirements.

Biologically, there are great differences between pteridophyte and spermatophyte
megaspores, as well as pteridophyte microspores and spermatophyte pollen grains.
However, if one considers the sporoderms alone these differences are less serious.
Future work therefore should also consider the megaspore membranes of seed
plants, as has been shown for Palaeozoic material by Zimmerman and Taylor (1971).

In general, it is demonstrable that in palynology and in other branches of palaeo-
botany a vast field is opened up by the use of the transmission electron microscope.
Such studies nowadays should be completed by scanning electron microscopy,
although due to its poor resolving power this method is incapable of providing suffi-
cient detail.

Acknowledgements. The transmission electron microscope (Zeiss EM 9 A) and the instruments for pre-
parative work (Leitz ultramicrotome, Leitz micromanipulator, etc.) were provided by the Deutsche
Forschungsgemeinschaft. Mr. W. Mackowiak carried out the preparative work, as well as the analysis of
ultra-thin sections in the electron microscope, with great skill. Megaspores of Setosisporites hirsutus and
Setosisporites brevispinosus were presented by Dr. H. Grebe (Geological Survey, Krefeld) and Dr. A.
Jachowicz (Geological Institute, Sosnowiec). The scanning electron micrographs were taken on a JEOL
U 3 SEM (Kontron, LWU, Munich: operator Mrs. Liebel) by courtesy of Mr. G. Cichy (Kontron,
Diisseldorf).

REFERENCES

EHRLICH, H. G. and HALL, J. w. 1959. The ultrastructure of Eocene pollen. Grana palynologica, 2, 32-35, 2 pis.

jux, u. and kempf, e. k. 1971. Microstructures of the Mesozoic megaspore Tasmanitriletes n. g. Grana, 11,
95-100, 1 pi.

KEMPF, E. K. 1969r/. ElektronenmikroskopiederSporodermis von kiinozoischen Megasporender Wasserfarn-
Gattung Azolla. Paldont. Z. 43, 95-108, 3 pis.

1969/>. Elektronenmikroskopie der Megasporen von Azolla tegeliensis aus dem Altpleistozan der

Niederlande. Palaeontographica, 128 B, 167-179, 8 pis.

1970. Elektronenmikroskopie der Sporodermis von Megasporen der Gattung Selaginella (Pteri-

dophyta). Rev. Palaeobotan. Palynol. 10, 99-116, 3 pis.

1971a. Electron microscopy of the megaspore Horstisporites semireticulatus from Liassic strata of

Germany. Grana, 11, 18-22, 1 pi.

19716. Elektronenmikroskopie der Sporodermis von Mega- und Mikrosporen der Pteridophyten-

Gattung Salvinia aus dem Tertiiir und Quartar Deutschlands. Palaeontographica, 136 B, 47-70, 13 pis.

1972. Electron microscopy of Mesozoic megaspores from Denmark. Grana, 11, 151-163, 5 pis.

KEMPF: TRANSMISSION MICROSCOPY OF SPORFS

797

LEFFINGWELL, H. A. et ol. 1970. A Study of the fossil pollen Wodehouseia spinata. Bull. Canad. Petrol. Geol.
18, 238-262, 7 pis.

MARTENS, p. 1960. Sur une structure microscopique orientee dans la paroi megasporale d’une selaginelle.
Nouvelles observations sur la structure des parois megasporales de Selaginella myosurus (Sow.) Alston.
Compt. Rend., Paris, 250, 1599-1602, 1774-1775, 2 pis.

PETTiTT, j. M. 1966. Exine structure in some fossil and recent spores and pollen as revealed by light and
electron microscopy. Bull. Brit. Mus. (Nat. Hist.), Geol. 13, 221-257, 21 pis.

and CHALONER, w. G. 1964. The ultrastructure of the Mesozoic pollen Classopollis. Pollen et Spores, 6,

611-620, 1 pi.

STAiNiER, F. 1965. Structure et infrastructure des parois sporales chez deux selaginelles (Selaginella myosurus
et S. kraussiana). Cellule, 65, 220-244, 5 pis.

1967. Morphological study of the walls of the mega- and microspores of Selaginella myosurus (Sw.)

Alston. Rev. Palaeobot. Palynol. 3, 47-50, 1 pi.

TAKEOKA, M. and STix, E. 1963. On the fine structure of the pollen walls in some Scandinavian Betulaceae.
Grana palynologica, 4, 161-188, 14 pis.

TAYLOR, T. N. 1971. Halletheca reticulatus gen. et sp. n.: a synangiate Pennsylvanian pteridosperm pollen
organ. Amer. Jl. Bot. 58, 300-308.

ZIMMERMAN, R. p. and TAYLOR, T. N. 1971. The ultrastructure of Paleozoic megaspores membranes. Pollen
et Spores, 12, 451-468, 6 pis.

Revised typescript received 1 February 1973

E. K. KEMPF

Geologisch-Palaontologisches Institut
Universitat Koln
D-5 Koln 1
Zulpicher Strasse 49
W. Germany

Discussion on Dr. Kempf’s paper:

Chaloner: In spores of fossil plants that you cannot immediately attribute to any group of living plants,
e.g. to ferns or lycopods, how do you decide what is perine and what is exine?

Kempf : By analogy with spores of modem plants. There the exine plays no part in the ornamentation of the
sporoderm surface. The ornamentation of the surface of a megaspore, for instance, is always made up by
the perine. If the perine is composed of several zones, each zone follows the ornament observed at the
surface. The exine, however, is unaffected. Further, the perine has a functional morphology; it has to protect
the spore from the environment, and also it can fill its interstices with gas so that it can float on water.

Chaloner: It is a pity that we do not have ‘types’ in our terminology as well as in taxonomy. If we had
a ‘type’ for the perine, it would surely be in the fem family Polypodiaceae. For most people, the term
perine is more or less confined to this family. Other ferns, Osmunda, for example, do not have a perine.
So that in Osmunda spores the exine must be forming the sculpture.

Kempf : I know of no instance where— according to transmission electron micrographs— the exine forms
the sculpture.

Chaloner: How do you define exine then?

Kempf : It is the middle layer of a three-layered sporoderm.

^ . \>/i- ’ •'^■1 ’’■^'*- i'tS4ll:k- "^'K)6?.i.

■m

I'jrv

Pr

''■•■T i ,.'

;f-,

ft.,>a!<}®

'*;■ '

# ‘ !»•

iSsWi .MA'i'li!^'-

;^i

..■miirrst <«■

-ji:;

!V» i

•’ ( 4 ■<#*'■ ■' ■

JSi

OC

o '■:••

,.U' (.'.i'l^v7i .»■*'?

■ . r ■

W=|l

,.: ^ . • ■' .<«y (' <,. un ■

l /i ; ■ '<,i; •>-‘'' f7ii» u':4jfi^>A.ii^FW♦A^PM^51^|fi(^I^^^^W^

, ' : ■ : ' '.«s,s'iM;i •>- '^M^f

.<-. K.V”* ■» /• V'if'i ••'•m:T««».' Jj!-.? M 'r'lt'pi-’* ff''

(.- . • ” V ...-aV^ .■•«’•• ■ ■*-■ ' ■■ • ' ■» T-Mv •'‘I A ,-m -, «.«•* ^ '. •'^'J

. ; . - W)f s.M',..»'. .v *hIi w wiHP‘V' ««>i ^

' y?i;y'y ^7;i;'v>>f^^l^v4%•/!ii^ tc^^" <’-i

" *^'^'. V* ■ •’•''■■■’■'■'

d:

» ' ? U I

J v'ti,'*

'■ I', I '■ ,'C^''til-i%.'' . 'Vr •<

1

ACRITARCHS OF THE MIDDLE SILURIAN
ROCHESTER FORMATION OF SOUTHERN

ONTARIO

by BINDRA THUSU

Abstract. The Rochester Formation yields an acritarch microfiora containing 24 genera and 46 species and varieties.
Genus Hemideun ffia ; Species Domasia canadensis, D. rochesterensis, Hemideunffia trifurcala, Elektoriskos simplex,
Filisphaeridium bifurcatum, Gorgonisphaeridium wenlockium, Lophosphaeridium rugosum and L. microgranulosum',
varieties; Deunjfia furcata var. niagarensis, D. monospinosa var. tonowandensis, and D. ramuscidosa var. rochesterensis
are proposed as new.

The Rochester netromorph assemblage, in particular the Deunjfia and Domasia complex, shows a close resemblance
to those from the Ilion Shale in Utica, New York, the Rochester Formation in West Virginia, and the Buildwas Beds
in Shropshire, Britain. The stratigraphically restricted genus Deunjfia, in particular D. ramusculosa and D. furcata
which probably made their first appearance during the Lower Wenlockian times, is present in all four areas.

However, the absence of Deunjfia and the presence of Domasia complex in the Wenlockian assemblages from Visby
and Belgium suggest some connections with the Rochester assemblage ; but the absence of Deunjfia- Domasia complex
in the Wenlockian assemblages from Spain and Sahara suggests fewer connections with the Rochester assemblage.

This paper contains results of acritarch studies on the Rochester Formation in
southern Ontario (text-fig. 1). This region lies south-east of the Canadian shield and
forms the marginal areas of the Michigan Basin and the Allegheny Trough.

The Rochester Formation is Tonawandian (Middle Silurian) in age and consists
of calcareous shale with interbeds of argillaceous micrite or biomicrite. It is exposed
at numerous localities along the Niagara Escarpment in New York and southern
Ontario. It has a maximum thickness of 150 feet east of Walcott, New York. To the
east the formation grades into the Herkimer Sandstone. To the west it thins succes-
sively to 60 feet near Niagara Gorge, 56 to 35 feet in the St. Catherines and Grimsby
region, 14 feet at Hamilton, and 2-4 feet in the most northerly exposure at Clappison
road cut, near Hamilton, Ontario (text-fig. 1). Well log data (Caley, 1940) indicate
a general thinning of the formation from south-east to north-west Ontario without
any lithological change.

South-easterly the Rochester Formation extends into Pennsylvania, Maryland,
and West Virginia and varies from 20 to 40 feet in thickness, and maintains the same
lithic character (Folk 1962)-.

Investigation by Thusu (1972) shows that the Rochester deposition in southern
Ontario took place in a calm, subtidal, inner neritic, low-energy environment with
good circulation, interrupted by occasional storms, which extended agitation into the
subtidal zone.

However, conditions were not uniform within the Rochester sea and the several
resulting environments undoubtedly influenced the distribution of fossils (Thusu
1972). For example, in southern Ontario and western New York bryozoans are
present in large numbers, but to the east their number and diversity are greatly
reduced. In Pennsylvania and West Virginia a normal marine fauna is absent. Folk
(1962) postulates that parts of the Rochester Sea were protected by barrier-bars

[Palaeontology, Vol. 16, Part 4, 197.1, pp. 799-826, pis. 104-106.]

Niagara falls access road ;

Sir Adam Beck- Niagara generating station.

mUKOT

IHE

FffgWrol —

2-Decew falls Generating station. St. Catherines.

3. Rockway , Fifteen mile Creek .

4. Glendridge road-cut.

5.6.7. Grimsby road, Beach road.

8.9- Grimsby Gorge. Grimsby

10. Grimsby road-cut west

11. Fruitland Road cliff.

12.13 Stoney Creek road-cut and gorge.

14 Highway 20..5toney creek

15. Albion Falls.

16,17 Wentworth Street-Sherman Avenue. Hamilton,
18 Ancaster highway 2.

19,20, Dundas quarry. Sydenham road-cut.

21 Clappison Corners road-cut.

o

m

o

00

THXT-MG. 1. Columnar sections of the Rochester and adjacent formations exposed along the Niagara

Escarpment in southern Ontario.

THUSU: SILURIAN ACRITARCHS

801

which cut off wave action and resulted in lagoonal eonditions in eastern West Virginia,
while the presence of rich marine fauna in other parts of West Virginia and Pennsyl-
vania suggest the temporary removal of barriers and hence open marine conditions.

PREVIOUS WORK ON PALAEOZOIC ACRITARCHS OF SOUTH-WESTERN
ONTARIO AND WESTERN NEW YORK

White (1862, p. 385) reported a variety of acritarchs from Silurian nodules in
eentral and western New York. He identified these as Zanthidia, the sporangia of
desmids and gemmules of sponges. However, they were neglected and forgotten in
North Ameriea for 73 years, until Laird (1935, p. 256) again discovered Xanthidia in
the Loekport Formation (Middle Silurian) of Ontario. Basehnagel (1942, p. 1)
reported acritarchs from the Onondaga Chert (Devonian) of eentral New York. He
assigned them to genera and families of freshwater algae. Fisher (1953, p. 13) reported
a number of acritarchs from thin-sections of the Neagha and Maplewood Shale in
New York. He regarded them to be zygospores of brown algae and compared some
of his forms with those reported by Deflandre (1946) from the Silurian of France.

The first major contribution was made by the French palynologist Deunff (1954u,
1955, 1957). He made the first taxonomic studies on some of the Devonian acritarehs
of Canada. Recent studies by Staplin el al. (1965) and Loeblich (1970) are mainly of
taxonomie importance. But Cramer (1968, 1970, 19706) made the first attempt to use
acritarchs in Lower Palaeozoic stratigraphy and correlation.

MATERIAL AND METHODS

Channel samples from the Rochester Formation used in the present investigation
consist of dark grey to black fissile calcareous shale, in part dolomitic, and argillaceous
grey limestone.

Localities are designated according to the Silurian section numbers (see Map I)
given by Bolton (1957). Maximum thickness of the Rochester Formation is given at
the end of each locality name.

Locality 1.
ROC.l
ROC.2692

ROC.2

ROC.2691

ROC.7

ROC.2690

Niagara Falls, access road. Sir Adam Beck-Niagara generating station, 60 feet.

Basal part of the section, Calcareous Shale, 20 feet.

Basal part of the section. Calcareous Shale on the American side of the Niagara generating
station (sample provided by Geology Department, Sheffield University).

Middle part of the section. Calcareous Shale, 20 feet.

Middle part of the section, Calcareous Shale, location same as ROC.2692.

Upper part of the section. Calcareous Shale, 20 feet.

Upper part of the section. Calcareous Shale, location same as ROC.2692.

Locality 2.
ROC. 15
ROC. 16
ROC. 17

Locality 14.
ROC.8
ROC.9

Locality 16.
ROC. 11
ROC. 10

DeCew Falls Generating station, St. Catherines, 56 feet (locality 1 of Bolton, 1957).
Basal part of the section, Calcareous Shale thin limestone, 20 feet.

Middle part of the section, thin bedded limestone and Calcareous Shale, 20 feet.
Upper part, Calcareous Shale, massive limestone, 16 feet.

Highway 20, Stoney Creek, 24 feet.

Basal part limestone. Calcareous Shale, 12 feet.

Upper part, Calcareous Shale with gypsum, 12 feet.

Jolly Cut, Hamilton, 13 feet.

Basal part, Calcareous Shale, 7 feet.

Upper part. Calcareous Shale, 6 feet.

802

PALAEONTOLOGY, VOLUME 16

Locality 18. Ancaster, Highway 2, 9 feet.

ROC. 5 Basal part, Calcareous Shale, 4-5 feet.

ROC.4 Upper part. Calcareous Shale, 4-5 feet.

Locality 19, 20. Dundas Quarry, Sydenham road-cut, 5 feet.

ROC.43007 Calcareous Shale 4 to 4-5 feet.

Locality 21. Clappison Comers road-cut, 2-4 feet.

ROC.6 Calcareous Shale, 2-4 feet.

ROC. 12 Calcareous Shale, 2 feet (250 feet east of sample ROC.6).

Apart from those from the Rochester Formation, additional samples studied and
reported in the present study are as follows:

(i) Ilion Shale (Wenlockian) Utica, New York. Three samples collected from the
type area are from south branch of Moyer Creek, Utica, New York. The reader is
referred to Zenger (1965, p. 191, section 166) for stratigraphic details of this section.
All shale samples yield an abundant microplankton flora characterized by an excel-
lent state of preservation.

(ii) Hogklint Shale (Wenlockian) Snackgardsbaden, Visby, Sweden. One pro-
ductive sample used is from a Korpklint Bay on the north side of Snackgardsbaden,
five kilometres along the coast north of Visby. The sample comes from behind a small
marine swimming pool (material and stratigraphic information provided by Dr.
D. Owen of the Manchester University Museum, England).

Standard chemical techniques were employed, similar to those described by
Cramer (1970) for the extraction of organic-walled micro-fossils. Most of the speci-
mens were examined at x400 magnification. All acritarch figures are housed in the
micropalaeontological collection of the Geological Survey of Canada in Ottawa.
Each specimen is recorded with east-west and north-south mechanical-stage read-
ings of Vickers Photomicroscope M-15.

Acknowledgements. For help and encouragement during the course of this work, the writer is grateful to
Professor D. L. Dineley, Dr. O. Dixon, Dr. M. J. Copeland, Dr. David Owen, Dr. C. Downie, Fritz Cramer
and Carmina Cramer, Dr. D. Fisher, and Dr. D. Zenger. Dr. D. Broad, and Dr. J. Murray helped during
the preparation of the manuscript and made many linguistic corrections, Mr. R. Godwin and Mr. E. Seavill
photographed and reproduced plates, and Mr. P. Shepherd helped in laboratory techniques.

SYSTEMATIC PALAEONTOLOGY

The following list of all acritarch species identified follows the morphographic
classification of Downie et al. (1963). Most of the species are illustrated in the plates
but only those asterisked are described below. All the new taxa are based on a
minimum of 25 specimens. Table 1 shows the number of specimens counted in each
sample. All slides are deposited with the Palaeontological Division of the Geological
Survey of Canada.

Group ACRiTARCHA Evitt 1963

Subgroup NETROMORPHITAE Downie, Evitt, and Sarjeant 1963

Deunffia furcata (pi. 104, fig. 20).

* D. furcata var. niagarensis new var. (pi. 104, fig. 14).

D. monospiuosa Downie 1960 (pi. 104, fig. 8).

THUSU: SILURIAN ACRITARCHS

803

*D. monospinosa var. tonawandensis new var. (pi. 104, figs. 4, 24).

D. ramusculosa Downie, 1960 (pi. 104, figs. 7, 15, 22).

*D. ramusculosa var. rochesterensis new var. (pi. 104, fig. 18).

Domasia amphora Martin 1969 (pi. 104, figs. 9, 19).

D. bispinosa Downie 1960 (pi. 104, fig. 11).

*D. canadensis sp. nov. (pi. 104, figs. 6, 12, 21).

D. elongata Downie 1960 (pi. 104, figs. 23, 25).

*D. rochesterensis sp. nov. (pi. 104, figs. 2, 5).

D. trispinosa Downie, 1960 (pi. 104, fig. 1).

* Hemideunffia trifurcata sp. nov. (pi. 104, figs. 10, 16).

Leiofusa algerensis Cramer 1970 (pi. 104, fig. 17).

L. fusiformis Eisenack 1938 (pi. 104, figs. 3, 13).

Eupoikilofusa stratifera Cramer 1964 (pi. 105, fig. 1).

Subgroup ACANTHOMORPHiTAE Downie, Evitt, and Sarjeant 1963

Ammonidium cf. A. microladum (Downie) Lister 1970 (pi. 105, fig. 5).
Baltisphaeridium longispinosum (Eisenack) Downie 1963 (pi. 105, fig. 19).

*B. pilaris Cramer, 1964 (pi. 105, figs. 6, 18).

* Diexallophasis denticulata (Stockmans and Williere) Loeblich 1970 (pi. 105, fig. 3).
Elektoriskos pogonius Loeblich 1970 (pi. 105, fig. 10).

*E. simplex sp. nov. (pi. 105, fig. 9).

* Eilisphaeridium bifureatum sp. nov. (pi. 105, figs. 8, 12).

*Gorgonisphaeridium wenloekium sp. nov. (pi. 105, fig. 11).

Helosphaeridium latispinosum Lister 1970 (pi. 105, fig. 16).

Micrystridium stellatum Deflandre 1945 (pi. 105, fig. 7).

Multiplicisphaeridium arbuseuliferum Downie 1963 (pi. 105, figs. 4, 15).

M. eoplanktonieum (Eisenack) Lister 1970 (pi. 105, figs. 2, 14).

M.fisherii (Cramer) Lister 1970 (pi. 105, fig. 17).

Quadraditum fantastieum Cramer 1964 (pi. 105, fig. 13).

Tunisphaeridium tentaeuliferum (Martin) Cramer 1970 (pi. 105, fig. 20).
Visbysphaera dilatispinosa (Downie) Lister 1970 (pi. 106, fig. 1).

Subgroup POLYGONOMORPHiTAE Downic, Evitt, and Sarjeant 1963

*Evittia monterrosa (Cramer) new comb. (pi. 106, figs. 2, 7).

E. remota (Deunlf) Lister 1970 (pi. 106, fig. 4).

*Veryhachium lairdi (Deflandare) ex-Deunlf 1954 (pi. 106, figs. 5, 6).

*V. limaciforme Stockmans and Williere 1963 (pi. 106, figs. 8, 10).

V. trispinosum Eisenack 1931 (pi. 106, fig. 9).

V. wenloekium (Downie) Downie and Sarjeant 1964 (pi. 106, fig. 3).

Subgroup HERKOMORPHiTAE Downie, Evitt, and Sarjeant 1963

Cymatiosphaera octoplana Downie 1959 (pi. 106, figs. 11, 14).

C. wenlockia Downie 1959 (pi. 106, fig. 19).

Dictyotidium dictyotum (Eisenack) 1955 (pi. 106, fig. 12).

804

PALAEONTOLOGY, VOLUME 16

Subgroup PTEROSPERMORPHiTAE Dowuie, Evitt, and Sarjeant 1963

Pterospermopsis cf. P. martinii Cramer and Cramer 1968 (pi. 106, fig. 15).
P. onondagaensis Deunff 1955 (pi. 106, figs. 16, 20).

Duvemaysphaera aranaides (Cramer) 1970 (pi. 106, fig. 18).

Subgroup SPHAEOMORPHiTAE Downie, Evitt, and Sarjeant 1963

* Loplwsphaeridium rugosum sp. nov. (pi. 106, fig. 13).

*L. microgrcmuJosum sp. nov. (pi. 106, fig. 17).

Leiospheridia spp.

Spores Pimctatisporites sp. rare.

TABLE 1. Numbers of specimens recorded in sample from localities 1, 14, 18, and 21 out of approximately

250 specimens counted in each sample.

Loc. 1

Loc. 14

Loc. 18

Loc. 21

ROC.

2692

ROC.

2691

ROC.

2690

ROC.8 ROC.9

ROC.5 ROC.4

ROC.6

Deunffia furcata

2

6

4

6

6

5

D. furcata var. niagarensis

1

2

3

4

3

2

D. monospinosa

5

5

5

8

6

2

D. monospinosa var. tonawandensis

1

2

3

4

1

4

6

D. ramusculosa

30

35

30

28

4

18

14

10

D. ramusculosa var. rochesterensis

11

3

9

12

4

4

10

2

Domasia amphora

10

7

3

2

2

14

D. bispinosa

6

4

2

6

5

6

4

D. canadensis

23

15

10

3

8

14

18

D. elongata

15

16

22

22

26

10

12

25

D. rochesterensis

12

7

5

6

4

4

D. trispinosa

21

18

16

16

17

12

8

12

Hemideunffia trifurcata

3

1

1

4

3

Leiofusa algerensis

9

5

3

2

4

6

5

L. fusiformis

6

6

4

2

7

4

4

2

E. stratifera

1

2

1

Ammonidium cf. A. microcladum

4

5

3

6

2

2

6

14

Baltisphaeridium longispinosum

15

7

6

4

8

6

6

16

B. pilaris

7

3

3

2

4

10

2

15

Diexallophasis denticulata

12

12

11

6

14

14

8

20

Elektoriskos pogonius

2

1

2

4

4

3

E. simplex

4

4

6

3

4

4

8

Filisphaeridium hifurcatum

1

2

2

2

6

4

Gorgonisphaeridium wenlockium

2

1

2

1

2

Hel osphaeridium lat isp i nosum

12

2

18

10

13

10

10

15

Micrhvstridium Stella turn

11

3

9

6

15

6

4

10

M ultiplicisphaeridiiim arhusculiferum

4

5

4

2

3

8

4

8

M. eoplanktonicum

5

2

2

8

1

4

4

3

M. fisher a

6

3

4

2

727

6

2

6

Quadraditum fantasticum

1

1

Tunisphaeridium tentaculiferum

1

2

4

Vishysphaera dilutispinosa

1

1

THUSU; SILURIAN ACRITARCHS

805

Loc. 1

Loc.

14

Loc. 18

Loc. 21

ROC,

2692

ROC.

2691

ROC.

2690

ROC.8

ROC.9

ROC.5 ROC.4

ROC.6

Evittia monterrosa

5

2

2

6

2

Evittia remota

2

2

2

4

2

5

V. lairdi

4

4

7

4

2

4

6

3

V. limaciforme

6

1

7

2

21

4

2

3

V. trispinosum

11

12

14

6

11

2

6

8

V. wenlockiwn

10

3

2

6

12

2

6

12

Cymatiosphaera octoplana

1

5

3

6

2

2

2

C. Wenlockia

3

3

1

4

4

2

Dictyotidium dictyotum

2

2

11

12

11

8

4

7

Pterospermopsis cf. P. martinii

2

1

2

P. onodagaensis
Duvernaysphaera aranaides

2

1

2

1

2

1

2

2

2

Lophosphaeridium rugosum

2

2

5

6

4

18

16

L. microgranulosum

2

1

2

1

Subgroup NETROMORPHiTAE Dowuie, Evitt, and Sarjeant 1963

Diagnosis. Acritarchs having an elongate to fusiform test without an inner body.
Surface smooth, granular, striate, spines or with large ornament. One or more spines
closed distally may be present at one or both poles. Questionable openings observed
in a few species.

Remarks. Well-known genera include: Deunffia Downie 1960; Domasia Downie
1960: Leiofusa Eisenack 1938: Dactylofusa Brito and Santos 1965 and Poikilofusa
Staplin, Jansonius and Pocock 1965. These have a wide geographical distribution
and are characteristic of the Silurian strata. Other genera include Anthatractus Deunff
1954: Baiomeniscus Loeblich 1970: Disparifusa Leoblich 1970: Limulidia Eisenack
1958: Pseudolwmlidia Brito and Santos 1965; Leiovalia Eisenack 1965: Lunulidia,
Leiovalia and Pseudolwmlidia are forms with close morphological relationship with
Leiofusa. In fact the genotypes of Lunulidia and Leiovalia were originally described
under Leiofusa (Eisenack 1938, 1951). These are relatively lesser known genera and
range from the Upper Ordovician to the Lower Devonian.

Some of the netromorphs, e.g. Domasia, show a close morphological similarity
with the polygonomorphs, for example Veryhachium. Downie (1963, p. 637) observed
V. elongatum and found forms with intermediate morphology between Veryhachium
of the V. trispinosum — V. trisulcum group, and Domasia of the D. trispinosa —
D. elongata group. However, Downie (1963, 633-634) recognized close relationship
of Leiofusa, Deunffia, and Domasia.

Deflandre and Deflandre (1964) removed Deunffia and Domasia from Netro-
morphitae, considering them simplified Polygonomorphitae.

However, Brito (1967) rejected the views of Deflandre and Deflandre and suggested
a close relationship of Deunffia, Domasia, and Leiofusa.

806

PALAEONTOLOGY, VOLUME 16

Leiofusa tumido
(small forms)

Subgroup Nelromorphitae

Subgroup Poly-
gonomorphitae

V. Hmaciforme

Veryhachium

e/ongotum

TEXT-FIG. 2. Hypothetical relationship between Leiofusa, Eupoikilofusa, Deunffia, and Domasia of sub-
group Netromorphitae and distinction between Domasia and Veryhachium (subgroup Polygonomorphitae).

This writer shares the views of Downie (1963) and Brito (1967) on possible relation-
ship of some Netromorphs. Some of the Rochester acritarchs seem critical to demon-
strate a relationship between Leiofusa, Deunffia, Domasia and Eupoikilofusa
(text-fig. 2). Further, this writer believes that Lunulidia, Leiovalia are forms with
close morphological relationship with Leiofusa.

Undoubtedly some of the Netromorphs, in particular species of Domasia are
intermediate in morphology with Polygonomorphs, for example Veryhachium
elongatum. However, examination of hundreds of specimens show two fundamental
characters on which species of Domasia and Veryhachium can be separated.

1. Veryhachium spp. are distinctly polygonal, and processes always leave the
vesicle from the corners. Domasia spp. have an elongate to fusiform vesicle and
processes are generally located near the poles of the vesicle.

2. Domasia spp. are always with septate processes; in Veryhachium this feature is
extremely rare.

Genus deunfeia Downie 1960 emend.

Type species. Deunffia monospinosa Downie 1960. Middle Silurian, England.

Original diagnosis. Vesicle hollow, elongate ellipsoidal, more or less smooth, less
than 100 |u in length. Body composed of thin, pale yellow to brown, organic mem-
brane. Ornament consisting of a single hollow spine situated at one end. Spine
terminates in a point or branches in various ways.

THUSU: SILURIAN ACRITARCHS

807

The above-mentioned diagnosis is emended to include:

1 . Ornament consisting of one or two hollow processes situated at opposite ends.
The long process terminates in a point or branches in various ways, other process is
short and terminates in a point and does not exceed more than a few microns.

2. The 100 |U, length limit is removed to include larger forms.

Deunffia furcata var. niagarensis new var.

Plate 104, fig. 14

Holotype. GSC No. 31615, locality 14, sample ROC.9, Slide No. 9/C.

Remarks. D. furcata var. niagarensis differs from D. furcata by the presence of a small
spine at the opposite end of the main shaft.

Dimensions. Size of the vesicle 18 p. (range, 16-18 (u); width of the vesicle 12 /x (range,
10-13 ju); length of the main shaft 1 6 /x (range, 14-16 /u.); length of the small process 1 fu.
(range, 1-2 ^x).

Deunffia monospinosa var. tonowadensis new var.

Plate 104, figs. 4, 24

Type specimens. Holotype GSC No. 31609, locality 21, sample ROC. 12, Slide No. 12/a; paratype GSC
No. 31609a, locality 18, sample ROC.4, Slide No. 4/A.

Remarks. This variety differs from D. monospinosa by the presence of a small spine
at the opposite end of the main shaft, and from L. alegerensis by a much smaller
spine.

Dimensions. Size of the vesicle 18 fx (range, 16-24 jx); width of the vesicle 9 fx (range,
7-11 j(x); length of the main shaft 60 p (range, 50-75 p); length of the small process 8 /x
(range, 6-8 p).

Deunffia ramusculosa var. rochesterensis new var.

Plate 104. fig. 18

Holotype. GSC No. 31618, locality 1, sample ROC.l, Slide No. 1/B.

Remarks. This species differs from D. ramusculosa by the presence of a small process
(3-4 p) opposite to the end of the long shaft. Where the process is broken, a small
hole or an opening is observed. However, where the vesicle is squashed or folded,
D. ramusculosa may be confined with D. ramusculosa var. rochesterensis.

Genus domasia Downie 1960

Type species. Domasia trispinosa Downie 1960. Middle Silurian, England.

Diagnosis. Vesicle hollow, elongate, ellipsoidal more or less smooth, about 20 p in
length. Body composed of pale yellow to brown organic membrane. Ornament
consisting of two relatively long hollow spines arising near one pole and a single
spine of variable length at the opposite end.

808

PALAEONTOLOGY, VOLUME 16

Remarks. The two processes arising near one pole may bifurcate at the tips, or a thin
hairy process may arise from the middle of the spine. D. elongata, D. trispinosa,
D. hispinosa, and D. amphora are morphologically very closely related species. How-
ever, the nature of the entrance of the processes into the vesicle seems to differentiate
these species. D. elongata and D. amphora have processes merging before entering
the vesicle, while D. trispinosa and D. hispinosa have processes entering the vesicle
separately. The Rochester acritarchs show a wide variety of transional forms between
D. elongata and D. amphora: Downie (1963, p. 637) records a transition between
D. elongata and D. trispinosa.

Domasia canadensis sp. nov.

Plate 104, figs. 6, 12, 21

Type specimens. Holotype GSC No. 31610, locality 1, sample ROC.l, Slide No. 1/B; Paratypes GSC
No. 31610a-b, locality 1, sample ROC.l, Slide No. 1/A, 1/E.

Description. Vesicle fusiform, pale, smooth or faintly granulose. One pole tapering
at first into a stretched, unseptate neck, which is drawn out into two processes. Other
pole tapers and terminates into one short process.

EXPLANATION OF PLATE 104

All figures x 500 and from unretouched negatives.

Fig. 1. Domasia trispinosa Downie 1960, GSC No. 31606, locality 14, sample ROC. 8, slide number 8/E.

Figs. 2, 5. D. rochesterensis sp. nov. 2, holotype GSC No. 31607, locality 1, sample ROC.l, slide number
1/C. 5, paratype GSC No. 31607a, locality 1, sample ROC.l, slide number 1/B.

Figs. 3, 13. Leiofusa fusiformis Eisenack 1938, GSC Nos. 31608 a-b, 3, locality 1, sample ROC.l^slide
number 1/A. 13, locality 21, sample ROC. 12, slide number 12/A.

Figs. 4, 24. Deiinffia monospinosa var. tonawandensis new var. 4, holotype GSC No. 31609, locality 21,
sample ROC. 12, slide number 12/A. 24, paratype GSC No. 31609a, locality 18, sample ROC.4, slide
number 4/A.

Figs. 6, 12, 21. D. canadensis sp. nov. 6, holotype GSC No. 31610, locality 1, sample ROC.l, slide number
1/B. 12, 21, paratypes GSC No. 31610 a-b, locality 1, sample ROC.l, slide numbers 1/A, 1/E.

Figs. 7, 15, 22. Deunffia ramusculosa Downie 1960, GSC Nos. 31611 a-c. 7, locality 21, sample ROC.12,
slide number 12/A. 15, locality 14, sample ROC. 8, slide number 8/A. 22, locality 1, sample ROC.l,
slide number 1 /A.

Fig. 8. D. monospinosa Downie 1960, GSC No. 31612, locality 1, sample ROC.l, slide number 1/C.

Figs. 9, 19. Domasia amphora Martin 1969, GSC No. 31613. 9, locality 18, sample ROC.5, slide number 5/A.
19, locality 1, sample number ROC.l, slide number 1/B.

Figs. 10, 16. Hemideunffia trifurcata sp. nov. 10, holotype GSC No. 31614, locality 1, sample ROC.l,
slide number 1/C. 16, paratype GSC No. 31614a, locality 1, sample number ROC.2691, slide number
2691 /A.

Fig. 1 1. Domasia hispinosa Downie 1960, holotype 31615, locality 14, sample ROC. 8, slide number 8/E.

Fig. 14. Deunffia furcata var. niagarensis new var., holotype GSC No. 31615, locality 14, sample ROC. 9,
slide number 9/C.

Fig. 17. Leiofma algerensis Cramer 1970, GSC No. 31617, locality 1, sample ROC.l, slide number 1/F.

Fig. 18. Deunffia ramusculo.sa var. rochesterensis new var. Holotype GSC No. 31618, locality 1, sample
ROC.l, slide number 1/C.

Fig. 20. D. furcata Downie 1960, GSC No. 31619, locality 1, sample ROC.l, slide number 1/E.

Figs. 23, 25. D. elongata Downie 1960, GSC No. 31620 a-b, locality 1, sample ROC.l, slide number 1/B.

PLATE 104

THUSU, Silurian acritarchs

810

PALAEONTOLOGY, VOLUME 16

Remarks. D. candensis sp. nov. differs from D. amphora by the presence of a distinct
unseptate and stretched neck. In D. amphora, the vesicle terminates into a small neck,
but a sharp break is present between the neck and the vesicle.

Dimensions. Length of the vesicle 36 fx (range, 32-40 /x); width of the vesicle 7 fj.
(range, 6-10 /u); length of the long processes 41 fj, (range, 39-45 /x); length of the short
process 13 jj, (range, 9-15 |Lx); length of the neck 3-4 /x (range, 3-7 fx).

Domasia roehesterensis sp. nov.

Plate 104, figs. 2, 5

Type specimens. Holotype GSC No. 31607, locality 1, sample ROC.l, Slide No. 1/C, Paratype GSC
No. 31607a, locality 1, sample ROC.l, Slide No. 1/8.

Description. Vesicle hollow, fusiform, pale yellow, smooth. Two long processes arise
near one pole. Each process gives rise to a short secondary process at about the mid-
length. A single short process present at the opposite end. Processes septate.

Remarks. The presence of a short secondary process arising from the long processes
differentiates this species from Domasia elongata.

Dimensions. Size of the vesicle 24-34 ix (range, 20-38 /x); width of the vesicle 8-10 ju,
(range, 7-12 fx); length of the long processes 33 fx (range, 30-45 jx); length of the
secondary processes 5-7 jx (range, 4-7 fx); length of the short process 15-17 ix (range,

12-20 /x).

Genus hemideunffia gen. nov.

Type species. Hemideunffia trifurcata sp. nov.

Diagnosis. Vesicle highly elongate, tapering, pale yellow, thin, smooth walled; one
pole simple; opposite pole terminates into three processes, processes may be closed
or open at the distal end, septate processes may be present.

Remarks. This genus differs from Leiofusa by the presence of a trifurcate process at
one pole, and from Deun ffia by the presence of a long, tapering vesicle. Indeed, this
genus is intermediate between Leiofusa and Deunffia.

Hemideunffia trifurcata sp. nov.

Plate 104, figs. 10, 16

Type specimens. Holotype GSC No. 31614, locality 1, sample ROC.l, Slide No. 1/C; Paratype GSC
No. 31614a, locality 1, sample ROC.2691, Slide No. 2691 /A.

Description. Visicle highly elongate, fusiform, tapering, smooth thin walled, pale
transparent, fragile often compressed, twisted or folded; one pole simple; opposite
pole terminates into a trifurcate process, processes generally closed at the tips,
non-septate.

Dimensions. Length of the vesicle 70 p. (range, 65-85 pi); width of the vesicle 4-8 p,
(range, 4-10 /x); length of the processes 9 p. (range, 6-10 p.).

THUSU: SILURIAN ACRITARCHS

811

Genus baltisphaeridium (Eisenack 1958), emend, Downie and Sarjeant 1963,

emend.

Type species. Ovum hispidium longispinosum Eisenack 1931, Baltic, Ordovician.

Remarks. In this work Baltisphaeridium is accepted in a broad sense as amended by
Downie and Sarjeant (1963) but is restricted to include forms with predominantly
homomorphic, unbranched processes, closed or open to the vesicle cavity. Forms
with hetromorphic nature of the process termination, and free communication of the
process with vesicle, are transferred to Multiplicisphaeridium. Processes with equi-
furcate termination at the tips is transferred to Ammonidium. Forms in which pro-
cesses communicate freely with the vesicle and without any tendency for the processes
to close off at its junction with the central body are transferred to Diexallophasis.
Forms in which the central body is ornamented by numerous prominent pila, rather
being smooth or with minor ornamentation are considered to belong to Pilifero-
sphaera, and forms in which processes are very short and terminate in a point or in
short bifurcations with a feather or rosette of small spines, just below their distal
end are considered to belong to Tylotopalla.

Baltisphaeridium pilaris Cramer 1964

Plate 104, figs. 6, 18

1970 Baltisphaeridium pilaris var. typicum Cramer, p. 166. pi. 18. figs. 55d-h (gives detailed
synonymy).

1970 Cymhosphaeridium pilar (Cramer) Lister, p. 63, figs. 256-266.

Remarks. The branching of B. pilaris is very complex and considerable variation
exists within a species. However, in B. pilaris var. typicum, the cauliflorate termina-
tions of the processes is seen in most of the specimens studied. B. pilaris was trans-
ferred by Fister (1970, p. 64) to a new genus Cymbosphaeridium. This genus is primarily
based on the ‘reflected plate formula’ proposed by Fister (1970, p. 63-65). This
writer does not believe in creating new form-genera on hypothetical characteristics,
drawn from the reconstruction of hypothetical thecae of acritarchs. Indeed, such
study is important to understand the phylogenetic relationship of acritarchs, but not
for creating new genera in a morphographic system of classification.

Diexallophasis denticulata (Stockmans and Williere) Foeblich 1970

Plate 105, fig. 3

1963 Diexallophasia denticulata (Stockmans and Williere) Loeblich, p. 715, figs. 8 a-e, 9 a-c.

1963 Baltisphaeridium denticulatum Stockmans and Williere, p. 458, pi. 1, fig. 4.

1970 Baltisphaeridium denticulatum Cramer, pp. 136-138.

1965 Baltisphaeridium denticulatum Martin, p. 5, pi. 1, figs. 5, 6, text-figs. 5, 6.

1966 Baltisphaeridium denticulatum Martin, p. 309.

1968 Baltisphaeridium denticulatum Jardine and Yapoudjian, pi. 3, fig. 26.

1963 Baltisphaeridium granulatispinosum Downie, p. 640, pi. 9, figs. 1, 7.

1966 Baltisphaeridium granulatispinosum Martin, p. 310, pi. 1, fig. 24.

1968 Baltisphaeridium granulatispinosum Martin, p. 48, pi. 3, fig. 127, pi. 4, fig. 186, pi. 7, fig. 310,
pi. 8, figs. 360, 362.

1970 Evittia granulatispinosa Lister, p. 67, pi. 4, figs. 2, 3, 5-9, 12, pi. 5, fig. 2: text-figs. 170, 20b.

812

PALAEONTOLOGY, VOLUME 16

Remarks. Cramer (1970, pp. 138-140) describes five varieties of D. denticulata
{B. dent iculat urn). However, Lister (1970, p. 68) and the present writer recognize
extreme variability in this group. The character of spines and ornament is con-
tinuously variable, thus encouraging the use of this species in a broad sense.

Diexallophasis granulatispinosum {= Baltisphaeridiurn granulatispinosum) is
regarded as a junior synonym of D. denit culata (= Baltisphaeridiurn denticulatum).
However, Lister (1970, p. 68) recognized Evittia granulatispinosa (B. granulati-
spinosum) as a valid taxon, and states ‘in view of the fact that the process tips of the
holotype of denticulatum are missing and the character of the process tips is essential
to the diagnosis of any species of Evittia, the specific name denticulatum is considered
to be nomen dubiurn. But the present author rejects the transfer of denticulatum to
genus Evittia for the reasons stated on p. 815, and recognizes the transfer of denti-
culatum in Diexallophasis. The diagnosis of Diexallophasis broadly defines the
character of tips, and hence would not warrant the rejection of the holotype of
denticulatum because of the missing tips.

Genus elektoriskos Loeblich 1970

Type species. Elektoriskos auora Loeblich 1970, Middle Silurian Maplewood Shale, New York.

Diagnosis. Circular to subcircular central body, wall apparently single layered.

EXPLANATION OF PLATE 105

All hgures x 500 and from unretouched negatives.

Fig. 1. Eupoikilofusa stratifera Cramer 1964, GSC No. 31621, locality 14, sample ROC. 8, slide number 8/B.

Figs. 2, 14. Multiplicisphaeridium eoplanktonicum Downie 1963, GSC Nos. 31622 a-b. 2, locality 1, sample
ROC. 1, slide number 1/A. 14, locality 18, sample ROC.5, slide number 5/A.

Fig. 3. Diexallophasis denticulata (Stockmans and Williere) Loeblich 1970. GSC No. 31623, locality 1,
sample ROC.l, slide number 1/A.

Figs. 4, 15. M. arbusculiferum Downie 1963, GSC Nos. 31624 a-b. 4, locality 16, sample ROC.l 1, slide
number 11/B. 15, locality 14, sample ROC.8, slide number 8/A.

Fig. 5. Ammonidium cf. A. rnicrocladum (Downie) Lister 1970, GSC No. 31626.

Figs. 6, 18. Baltisphaeridiurn pilaris Cramer 1964, GSC Nos. 31626. 6, locality 1, sample ROC.2690, slide
number 2690/A. 18, locality 21, sample ROC. 12, slide number 12/A.

Fig. 7. Micrhystridium stellatum Deflandre 1945, GSC No. 31627, locality 14, sample ROC.8, slide
number 8/B.

Figs. 8, 12. Eilisphaeridium hifurcatum sp. nov. 8, holotype GSC No. 31628, locality 1, sample ROC.l,
slide number 1/C. 12, paratype GSC No. 31628a, locality 14, sample ROC.8, slide number 8/A.

Fig. 9. Elektoriskos simplex sp. nov., holotype GSC No. 31629, locality 18, sample ROC.5, slide
number 5/A.

Fig. 10. E. pogonius Leoblich 1970, GSC No. 31630, locality 1, sample ROC.l, slide number 1/A.

Fig. 11. Gorgonisphaeridium wenlockium sp. nov., holotype GSC No. 31631, locality 14, sample ROC.8,
slide number 8/A.

Fig. 13. Quadraditum fantisticum Cramer 1964, GSC No. 31632, locality 1, sample ROC. 2691, slide
number 2691 /A.

Fig. 16. Helosphaeridium latispinosum Lister 1970, GSC No. 31633, locality 21, sample ROC. 12, slide
number 12/A.

Fig. 17. M.fisherii (Cramer) Lister 1970, GSC No. 31634, locality 14, sample ROC.8, slide number 8/A.

Fig. 19. B. longispinosum Downie 1963, GSC No. 31635, locality 1, sample ROC.l, slide number 1/A.

Fig. 20. Tunisphaeridium tenlaculiferum (Martin) Cramer 1970, GSC No. 31636, locality 6, sample ROC. 12,
slide number 12/A.

PLATE 105

V?', •*>■

THUSU, Silurian acritarchs

814

PALAEONTOLOGY, VOLUME 16

psilate, chagrenate to granulate with numerous slender, flexible but solid processes
which do not communicate with the interior of the central body.

Remarks. This genus diflfers from Comasphaeridium in lacking the densely crowded
hair-like processes and from Filispfiaeridium in lacking the distal differentiation of
the processes.

Elektoriskos simplex sp. nov.

Plate 105, fig. 9

1970 Elektoriskos sp. Loeblich, p. 719, fig. 13c.

Holotype. GSC No. 31629, locality 18, sample ROC. 5, Slide No. 5/A.

Description. Vesicle rounded to spherical, thin, smooth, numerous (30-40) short,
solid processes, without communicating with vesicle.

Remarks. E. simplex sp. nov. differs from E. sequestratus Loeblich by the absence of
small grana on the vesicle.

Genus filisphaeridium Staplin, Jansonius and Pocock 1965

Type species. Micrhystridium setasessitante Jansonius 1962, Lower Triassic, Alberta, Canada.

Eilisphaeridium bifurcatum sp. nov.

Plate 105, figs. 8, 12

Type specimens. Holotype GSC No. 31628, locality 1, sample ROC.l, Slide No. 1/C; Paratype GSC
No. 31628a, locality 14, sample ROC. 8, Slide No. 8/A.

Description. Vesicle ellipsoidal to subspherical, smooth, processes 30 or more short
tapering about 10% of the vesicle, bifurcating at the tips, process communicate freely
with the vesicle.

Remarks. E. bifurcatum sp. nov. differs from E. brevispinosum Lister by the presence
of bifurcating tips.

Dimensions. Size of the vesicle 27-36 /n (range, 25-38 /ti); length of the process 3-5 p.
(range, 3-5 p).

Genus gorgonisphaeridium Staplin, Jansonius and Pocock 1965

Type species. Gorgonisphaeridium winslowii Staplin, Jansonius and Pocock 1965, Lower Carboniferous,
Southern Alberta, Canada.

Gorgonisphaeridium wenlockium sp. nov.

Plate 105, fig. 11

Holotype. GSC No. 31632, locality 14, sample ROC. 8, Slide No. 8/A.

Description. Vesical spherical, wall firm and thick, brownish, numerous short spines
(30-40) with bulbous base and pointed tip, processes communicate freely with
vesicle.

THUSU: SILURIAN ACRITARCHS

815

Remarks. This species differs from G. spicatum (Staplin) by its larger size, and pointed
tips.

Dimensions. Size of the vesicle 80 fx (range, 70-82 /x); length of the spines 6-16 jx
(range, 5-lS fx); width of the spines at the base 5-6 fx (range, 5-6 fx); distance between
the spines 9-13 /u (range, 9-13 fx).

Genus evittia Brito 1967

Type species. Evittia sommeri Brito 1967. Lower Devonian, Brazil.

Restricted diagnosis. Vesicle triangular to polygonal like Veryhachium but with the
processes typically ramified, vesicle wall and processes sculptured.

Remarks. Cramer (1970, p. 47) considered Evittia a partial junior synonym of
Baltisphaeridium on the grounds that the basal portion of the processes and their
number strongly influence the shape of the vesicle. This is possible if Evitta possessed
many processes. But Brito’s generic diagnosis mentions ‘. . . having a general struc-
ture of Veryhachium . . .’ which in addition to a triangular or polygonal vesicle shape
also puts a limit on the number of the processes. In fact the Subgroup Polygono-
morphitae to which Evittia belongs is characterized by a low-number of processes.

Lister (1970, p. 66) emended Evittia, and broadened the genus to include sub-
spherical acanthomorph species of Baltisphaeridium thus greatly departing from
Brito’s original intentions. It should be noted that Evittia like Veryhachium is defined
on subgroup basis by a combination of shell shape and low spine number. Thus
introducing within this genus subspherical species regardless of spine number is
rejected as contrary to the original intention of Brito.

Evittia monterrosa (Cramer) nov. comb.

Plate 106, figs. 2, 7

1969 Baltisphaeridium monterrosae Cramer, p. 490, pi. 1, figs. 5-7.

1970 Baltisphaeridium monterrosae Cramer, p. 129, pi. 8, figs. 127-135.

Remarks. The basis for the transfer of this species to Evittia is the presence of a poly-
gonal vesicle with a small number of processes (2-5), bifurcating or trifurcating at
the distal end.

Genus veryhachium Deunff 1954, emend. Downie and Sarjeant 1963
Veryhachium lairdi (Deflandre) Deunff 1954

Plate 106, figs. 5, 6

1946 Hystrichosphaeridium lairdi Deflandre, card 1112.

1954 Veryhachium lairdi Deunff, p. 306.

1963 Veryhachium lairdi Stockmans and Williere, p. 454, pi. 4, fig. 5.

1964 Veryhachium lairdi Cramer, p. 309, pi. 11, fig. 16.

1965 Veryhachium lairdi Martin, pp. 13-14, figs. 14-15.

1969 Veryhachium lairdi Martin, p. 95, pi. 2, figs. 75-83.

1970 Veryhachium lairdi Loeblich, p. 741.

816

PALAEONTOLOGY, VOLUME 16

Remarks. V. lairdi has a maximum of five processes. However a single specimen
shown on Plate 106, figs. 5-6 contains six processes and is provisionally placed with
the specimens of V. lairdi.

Veryhachium limaciforme Stockmans and Williere 1963

Plate 106, figs. 8, 10

1963 Veryhachium limaciforme Stockmans and Williere, p. 433, pi. 1, figs. 12, 14, 15, 19.

1965 Veryhachium limaciforme Martin, p. 22, fig. 21.

1969 Veryhachium limaciforme Martin, p. 96, pis. 7, 8, figs. 354, 402.

1963 Veryhachium elongatum Downie, p. 637, pi. 92, fig. 10.

1963 Veryhachium delmeri Stockmans and Williere, p. 453, pi. 1, fig. 17.

1965 Veryhachium delmeri Martin, p. 22, fig. 22.

1966 Veryhachium delmeri Martin, p. 316.

1969 Veryhachium delmeri Martin, p. 90, pis. 4, 6, figs. 176, 346-347.

1970 Domasia limaciforme (Stockmans and Williere) Cramer, p. 68, pi. 1, figs. 16, 27, 28.

Remarks. This species is retained in the genus Veryhachium on three counts: 1. A
highly elongate, triangular vesicle. 2. The attachment of the processes to the corners
of the vesicle. 3. Absence or very rare septate processes, unlike the species of Domasia.

EXPLANATION OF PLATE 106

All figures x 500 and from unretouched negatives.

Fig. 1. Visbysphaera dilatispinosa (Downie) Lister 1970, holotype GSC No. 31637, locality 14, sample
ROC. 8, slide number 8/A.

Figs. 2, 7. Evittia monterrosa (Cramer) new comb. GSC Nos. 31638 a-b. 2, locality 1, sample ROC.l, slide
number 1/A. 7, locality 14, sample ROC. 8, slide number 8/A.

Fig. 3. Veryhachium wenlockium (Downie) Downie and Sarjeant 1964, GSC No. 31639, locality 1, sample
ROC.l, slide number 1/B.

Fig. 4. E. romota (Deunff) Lister 1970, GSC No. 31640, locality 18, sample ROC.5, slide number 5/A.

Figs. 5, 6. V. lairdi (Deflandare) ex-Deunlf 1954, GSC Nos. 31641 a-b. 5, locality 14, sample ROC.9,
slide number 9/A. 6, locality 1, sample ROC.l, slide number 1/A.

Figs. 8, 10. V. limaciforme Stockmans and Williers 1963, GSC Nos. 31642 a-b. 8, locality 14, sample
ROC.8, slide number 8/E. 10, slide number 8/A.

Fig. 9. V. trispinosum Eisenack 1931, GSC No. 31643, locality 1, sample ROC.l, slide number 1/A.

Figs. 11, 14. Cymatiosphaera octoplana Downie 1959, GSC Nos. 31644 a-b. 1 1, locality 14, sample ROC.8,
slide number 8/B. 14, slide number 8/A.

Fig. 12. Dictyotidium dictyotum (Eisenack) 1955, GSC No. 31645, locality 1, sample ROC.l, slide
number 1/G.

Fig. 13. Lophosphaeridium rugosum sp. nov., holotype GSC No. 31646, locality 18, sample ROC.5, slide
number 5/A.

Fig. 15. Pterospermopsis cf. P. martinii Cramer & Cramer 1968, GSC No. 31647, locality 1, sample ROC.l,
slide number 1/C.

Figs. 16, 20. P. onondagaensis DeunfiT 1955, GSC Nos. 31648 a-b. 16, locality 8, sample ROC.8, slide
number 8/A. 20, locality 1, sample ROC.l, slide number 1/A.

Fig. 17. L. microgranulo.mm sp. nov., holotype GSC No. 31649, locality 14, sample ROC.8, slide
number 8/A.

Fig. 18. Duvernaysphaera aranaides (Cramer) 1970, GSC No. 31650, locality 1, sample ROC.l, slide
number 1 /C.

Fig. 19. C. wenlockia Downie 1969, GSC No. 31651, locality 1, sample ROC.l, slide number 1/E.

PLATE 106

THUSU, Silurian acritarchs

818

PALAEONTOLOGY, VOLUME 16

Genus lophosphaeridium Timofeyev 1959

Type species. Lophosphaeridium rarum Timofeyev, designated by Downie (1963), Ordovician, Russia.

Lophosphaeridium rugosum sp. nov.

Plate 106, fig. 13

Holotype. GSC No. 31646, locality 18, sample ROC. 5, Slide No. 1/G.

Description. Vesicle rounded, orange yellow, thick walled, verrucose ornament, often
in crescentric pattern.

Remarks. This species could be an alete spore.

Dimension. 120-130 ju.

Lophosphaeridium microgranulosum sp. nov.

Plate 106, fig. 17

Holotype. GSC No. 31649, locality 14, Sample ROC.8, Slide No. 8/A.

Description. Vesicle rounded, orange yellow, wall thick, finely granulose, granules
closely packed, giving the appearance of a finely meshed network.

Dimensions. 160 fx.

DISCUSSION OF THE MICROFLORA

Age of the Rochester Microflora. The presence of highly evolved netromorphs of the
Deunjfia complex {Deunffia ramusculosa, D. furcata) in the Rochester Formation
allows correlation with the Buildwas Beds in Shropshire (Downie 1963). On this
basis the Rochester microflora is assigned a Lower Wenlockian age.

Intracontinental comparisons

Power Glen Formation {Lower Llandovery), Niagara Gorge, southern Ontario and
New York State. The Rochester acritarchs reported from the Power Glen Formation
(Cramer and Cramer 1970) belong to the Diexallophasis denticulata and Very-
hachium trispinosum complex (Table 2). These are forms ranging from the Upper
Llandovery to the Emsian.

Neagha and Maplewood Shales {Upper Llandovery) of New York State. The Rochester
acritarchs present in these shales (Table 2) are dominantly long-ranging species
(Llandovery/Ludlow) also. However, Baltisphaeridium neagha, Dactylofusa neagha,
Neoveryhachium carminae, and Carminella maplewoodensis reported by Cramer and
Cramer (1970, p. 713) and Elektoriskos aurora, Holothuriadeigma heterakanium, and
Multiplicisphaeridium mergaeferum reported by Loeblich (1970) in the Neagha Shale
were not found in the Rochester Formation.

Loeblich (1970) reported a number of species from the Maplewood Shale, in parti-
cular Baimeniscus granulatus, Diexallophasis caperoradiola, Estiastra stellata, and
species of Elektoriskos and Leiofusa. Furthermore, both Cramer and Cramer (1970)
and Loeblich (1970) record the presence of Neoveryhachium carminae and Carminella

THUSU: SILURIAN ACRITARCHS

819

TABLE 2. Selected acritarch species in the Rochester Formation and their reported occurrence in the Middle

Silurian strata in North America.

Lower

Upper

Llandovery

Age

Llandovery

Llandovery

Wenlock

Wenlock

Location

1 2

3

4 5 6

7

8

9

10

Deunffia furcata

X

X

D. monospinosa

* 9

D. ramusculosa

X

X

D. ramusculosa var. rochesterensis

Domasia amphora

X ? ?

D. bispinosa

X

X

D. canadensis

D. canadensis var. A.

D. elongata

X

X

X

X

X

D. rochesterensis

X

D. trispinosa

X

X

Leiofusa algerensis

X

Eupokilofusa striatifera

X

X

X

X

X

X

X

Ammonidium microcladum

X

X

X

Baltisphaeridium pilaris

X

X

Diexallophasis denticulata

X X

X

X

X

X

X

X

Elektroiskos pogonius

X

X

X

X

X

X

X

Helosphaerisphaeridium latispinosum

?x

X

Multiplicisphaeridium arbusculiferum

X

X

M. eoplanktonicum

X

M. fisherii

X

X

X

X

X

X

Quadraditum fantasticum

X

Tunisphaeridium tentaculiferum

X

X

X

X

X

Visbysphaera dilatispinosa

X

Evittia monterrosa

X

X

X

Cymatiosphaera wenlockia

X

X

X

X

X

Dictyotidium dictyotum

X

X

Duvernaysphaera aranaides

X

X

X

*? D. monocantha.

Legends for locations:

1. Power Glen Fm., S. Ontario (Cramer & Cramer 1970).

2. Maplewood Shale, New York (Cramer 1968, 19706).

3. Neagha Shale, New York, S. Ontario (Cramer 1970, Loeblich 1970).

4. Ross Brook Fm., Nova Scotia (Cramer 19706, Loeblich 1970).

5. Gun River Fm., Anticosti Is. (Cramer 1970).

6. Jupiter Fm., Anticosti Is. (Cramer 1970).

7. Rose Hill Fm., Pennsylvania (Cramer 1969).

8. Tuscarora Fm., Pennsylvania (Cramer 1969).

9. Rochester Fm.. Maryland-W. Virginia. Reaugh (pers. comm.).

10. Ilion Shale, New York. This work.

maplewoodensis. These species were not found in the Rochester Formation with the
exception of Elektoriskos pogonius.

Deunffia furcata, D. ramusculosa, D. ramuscolusa var. rochesterensis and Domasia
spp. are present in the Rochester assemblage and their absence in the Maplewood
and Neagha assemblages is the most distinguishing feature.

According to Cramer and Cramer (1970, p. 1080) these differences in the acritarch
assemblage are due to the existence of two distinct acritarch biofaces. In other words,
Maplewood and Neagha assemblage at one end and the Rochester assemblage at the
other, contain two distinct ‘biofacies’, the Neoveryhachium carminae facies and
Deunffia furcata and Domasia facies. Furthermore, Cramer and Cramer do not

820

PALAEONTOLOGY, VOLUME 16

believe that this difference between the biofacies was caused by the difference in age.
However, although in NW. Spain the N. carminae biofacies ranges from Llandovery
to Ludlow (Cramer 1970), in eastern North America the N. carminae facies is
restricted to the Zygobolba excavata ostracode Zone, of Late Llandovery age; while
the Deunffia furcata and Domasia facies appear in the Paraechmina spinosa ostracode
Zone of Basal Wenlock age.

Upper Member of Ross Brook Formation of Nova Scotia. Many Rochester Formation
acritarchs occur in the Ross Brook assemblage. The stratigraphically restricted
(Upper Llandovery to Wenlock) but geographically well-distributed netromorphs
Domasia amphora, D. bispinosa, D. elongata, Leiofusa algerensis, Duvernaysphaera
aranaides are common to both formations (Table 2). However, Deunffia spp. present
in the Rochester assemblage are absent in the Ross Brook assemblage (Cramer
1970Z), p. 747), indicating the younger age of the Rochester assemblage. Cramer
(19706) regards this difference as a function of palaeolatitudes.

Gun River and Jupiter Formations {Upper Llandovery), Anticosti Island, Quebec.
There is little in common between the acritarchs of the Rochester Formation and the
Gun River and Jupiter Formations (Table 2). However, a number of species abundant
in the Gun River and Jupiter assemblages (Cramer 19706, p. 749) are present in the
Wenlock strata of England and Visby, Baltic (see Downie 1963, pp. 646-647).

The Ilion Shale {Wenlock), Utica, New York. The Rochester acritarchs are similar to
the Ilion acritarchs (Table 2). The stratigraphically restricted taxa like Deunffia
ramusculosa, D. furcata, Domasia elongata, D. rochesterensis are common to both the
formations, although quantitative differences occur. For example, Deunffia and
Domasia spp, are abundant in the Rochester Formation but rare in the Ilion Shale.

The Rose Hill and Tuscarora Formations {Upper Llandovery), Pennsylvania. In
Pennsylvania, the Rose Hill and Tuscarora formations underlie the Rochester
Formation. Many long-ranging taxa in the Rochester assemblage in southern
Ontario are well represented in Rose Hill and Tuscarora assemblages in Pennsylvania
(Cramer 1969, p. 486) (Table 2). However, the presence of Domasia and the absence
of Deunffia spp. in the Rose Hill and Tuscarora assemblages supports the present
writer’s contention that Deunffia ramusculosa appears first in the Rochester assem-
blage (Lower Wenlock) in the Appalachian region.

The Rochester Formation in West Virginia. The Rochester acritarch assemblage in
southern Ontario has many netromorph species in common with the Rochester
assemblage in West Virginia (Reaugh, personal communication; Table 2). Domasia
elongata, D. bispinosa and D. trispinosa are especially common in the two areas.
However, only a few Deunffia ramusculosa occur in the Rochester assemblage in
West Virginia. This is attributed to its lagoonal conditions (Reaugh, personal com-
munication).

Intercontinental comparison

The Buildwas Beds {Based Wenlock) in Shropshire, England. The Rochester netro-
morphs show their greatest similarity with the Downie’s assemblage type 1, which is
restricted in the Buildwas Beds (Downie 1963, pp. 646-648). Deunffia and Domasia
are locally abundant in both the Rochester and the Buildwas assemblage (Table 3).

TABLE 3. Selected acritarch species in the Rochester Formation and their reported occurrence in the Middle
Silurian strata (Wenlock) in Britain, Europe and North Africa.

TABLE 3. Selected acritarch species in the Rochester Formation and their reported occurrence in the Middle
Silurian strata (Wenlock) in Britain, Europe and North Africa.

822

PALAEONTOLOGY, VOLUME 16

However, the Buildwas acanthomorphs, in particular the Visbysphaera meson
complex (V. oliogofurcatum, V. meson, V. brevifurcata), were not recorded in the
Rochester assemblage. These taxa plus V. dilatispinosa (present in both areas) are
now known to range from Upper Llandovery to Upper Ludlow (Lister, 1970, pp. 98-
100) and their absence or scarcity in the Rochester Assemblage may be attributed to
local environmental conditions.

The Hogklint Group ( Wenlock), Visby, Baltic. The Rochester acanthomorphs show
some similarity with those reported by Eisenack (1954, 1955, 1959) in the Middle
Silurian of the Baltic region. Of note is the absence in the Rochester assemblage of
Visbysphaera meson complex with the exception of single specimens of Baltisphaeridium
digitatum and B. corallanium. However, Cramer (1970) reports many Middle Silurian
acanthomorph species in the Rochester Formation.

Eisenack obtained acritarchs by hand-picking from aqueous slurry, thus exclud-
ing netromorphs from his assemblages. The writer studied an excellently preserved
acritarch assemblage from the Hogklint Group, Snackgardsbaden, Visby, and
recorded the netromorphs Domasia amphora, D. bispinosa, D. elongata, D. trispinosa,
and Eupoikilofusa stratifera. Cramer (1970, p. 67) recorded D. elongata in the upper-
most portion of the upper Visby Marl, in Gotland. In addition, numerous species
of Baltisphaeridium, Micrhystridium, Veryhachium, Lophosphaeridium, and Leio-
sphaeridia were recorded.

The Hogklint netromorph assemblage is comparable with that of the Rochester
assemblage (Table 3) with the exception of the absence of Deunffia spp. in the Hogklint
assemblage. Further detailed investigation of the netromorph distribution in Hogklint
group is needed for a more precise correlation.

The Wenlock Assemblage from the Montague Noire, France. The Rochester acritarchs
have little in common with the Wenlock assemblage from France. Micrystridium,
Veryhachium, and Baltisphaeridium species common to two areas are long ranging
(Llandovery to Emsian). Of the 16 species in France (Deflandre, 1942), 10 were
identified in Coalbrookdale Beds (i.e. post-Buildwas Beds) in England (Downie 1963,
p. 646). All these taxa are now known to be long-ranging.

The Wenlock Assemblage from Belgium. Of the 21 species in the Rochester Formation
9 occur in the Wenlock of Belgium (Table 3). The Belgian assemblage like the Hogklint
assemblage shows the presence of Domasia and the absence of Deunffia (except the
rare occurrence of D. monocantha in pre-Wenlock strata). The absence of Deunffia
complex from the Wenlockian of Baltic and Belgium, and its abundance in the
Buildwas Beds and the Rochester Formation is worthy of note. However, Neovery-
hachium carminae present in the Belgian assemblage does not occur in either the
Buildwas or the Rochester assemblage.

The San Pedro and Furada Formations in North-West Spain. A small number of
Rochester taxa present in the Spanish assemblage range in age from Upper Llando-
very to Gedinnian (Table 3). There seems to be a provincial differentiation between
the two assemblages. The Spanish are dominated by Neoveryhachium carminae and
numerous local species of Veryhachium and Baltisphaeridium. In contrast the
Rochester is generally dominated by the Deunffia and Domasia complex, micrhystrids
and leiospherids. This striking difference suggests the existence of different micro-

THUSU: SILURIAN ACRITARCHS

823

plankton provinces. Cramer (1970) believes such differences to be climatically
controlled.

Wenlock Assemblages from the Sahara, Algeria. Little similarity exists between the
Rochester taxa and the Wenlockian taxa in the Sahara (Table 3). Long-ranging
(Upper Llandovery/Ludlow) species common to both areas are Diexallophasis
dentieulata, Multiplicisphaeridium fisherri {Baltisphaeridium sp. 5 of Jardine and
Yapaudjian 1968), Evittia remota (Veryhachium sp. 2 of Jardine and Yapaudjian
1968), and Duvernaysphaera aranaides [Pterospermopsis cf. Helios of Jardine and
Yapaudjian 1968). Netromorphs (Deimffia and Domasia) are not reported from the
Sahara. The closer link of Saharan and Spanish assemblages is notable, in particular
the presence of Neoveryhachium carminae.

Palaeobiogeographic and palaeoelimatic considerations

Several palaeontologists, especially Bassler (1906, p. 8, 1911) and recently Owen
(1969, p. 621), have showed interest in possible links of the Wenlockian Appalachian,
Welsh Borderland, and Baltic faunas. It is therefore appropriate to discuss the
relationship of the microflora in these regions.

Comparative Microflora. In order to do this, only stratigraphically restricted netro-
morphs are considered. The species common to more than one locality are as follows :

Species

America

Britain

Gotland

(Rochester Fm.)

(Buildwas Beds)

(Hogklint Gp.)

Deunffia furcata

Present

Present

Missing

D. monospinosa

Present

Present

Missing

D. ramusculosa

Present

Present

Missing

Domasia amphora

Present

?

Present

D. bispinosa

Present

Present

Present

D. elongata

Present

Present

Present

D. trispinosa

Present

Present

Present

This comparison suggests that the Rochester microflora had more links with the
British than with the Baltic area.

Cramer (1970) suggested that acritarch ‘biofaces’ are due mainly to climatic factors.
He proposed three biofacies : ( 1) Neoveryhachium carminae, (ii) Domasia and Deunffia,
and (iii) Baltisphaeridium corallinum. The last biofacies partially coincides with the
Domasia and Deunffia biofacies.

In North America and western Europe, Cramer (1970) found the distribution of
these biofacies roughly paralleled palaeolatititudes (Cramer 1970, text-fig. 6). The
N. carminae biofacies occurred in south-east U.S.A. and the Iberian Peninsula and,
the Domasia and Deunffia biofacies in the Central Appalachian region and the Welsh
Borderland.

The Rochester acritarch assemblage belongs to the Domasia and Deunffia biofaces
and is similar to the Buildwas assemblage in England. However, the Baltic assemblage
(Hogklint Beds), although similar, shows the absence of the Deunffia complex. The
Deunffia complex in the Rochester and Buildwas assemblages may represent a local
climatic zone, separated by a narrower Atlantic ocean. This independently supports
Owen (1969), who on the basis of bryozoa study, suggests similar climatic zones and
a narrower Atlantic ocean for the similarity of Wenlockian bryozoa faunas of the

824

PALAEONTOLOGY, VOLUME 16

Appalachian and Welsh Borderland. A marked difference between these and the
Baltic faunas is attributed to a different climate. However, these suggestions are at
variance with those of other workers. Cramer (1970) considers the Domasia and
Deun ffia biofacies in the Rochester and Buildwas strata and the Domasia biofacies in
Hogklint strata to be indicative of a similar climate. St^rmer (1967, p. 209) considered
the Silurian fossiliferous reefs of Gotland to be facies of the eastern European Plat-
form, having little connection with the Caledonian geosyncline. This would explain
the similarity of the Appalachian and Welsh Borderland biota and their differences
from those of Gotland. But Paul’s (1967) investigations of Silurian cystids suggest
that Echinoencrinitidae and Callocystitidae present in Britain originated from the
Baltic and North America respectively. This would call for connections between these
areas, a suggestion made earlier by Owen (1969).

REFERENCES

BASCHNAGEL, R. A. 1942. Some microfossils from the Onondaga Chert of central New York. Bull. Bujfalo
Soc. Nat. Sci. 17, 1-8.

BASSLER, R. s. 1906. The bryozoan fauna of the Rochester Shale. Bull. U.S. Geol. Surv. 292, 1-137.
BOLTON, T. E. 1957. Silurian stratigraphy and Paleontology of the Niagara escarpment in Ontario. Mem.
Geol. Survey. Can. 289.

BRITO, I. M. 1967. Silurian and Devonian acritarcha from the Maranhao Basin, Brazil. Micropaleontology,
13, 373-482.

BRITO, I. and SANTOS, A. 1965. Contribuicao ao conheciments dos microfosseis Silurianos e devonianos da
Bacia do Maranhao. Notas Prelim. Estudos, Divisdd Geol. Mineral, Rio de Janeiro, 129, 3-22.

CALEY, J. F. 1940. Paleozoic geology of Toronto-Hamilton area, Ontario. Mem. Geol. Survey Can. 224.
CRAMER, F. H. 1964. Microplankton from three paleozoic formations in the province of Leon (N.W. Spain).
Leidse. Geol. Meded. 30, 253-361.

1968. Palynolgic microfossils of the Middle Silurian Maplewood Shale in northwestern New York.

Rev. Micropaleont, 11, 61-70.

1969. Possible implications for Silurian paleogeography from phytoplancton assemblages of the Rose

Hill and Tuscarora Formation of Pennsylvania. J. Paleont. 43, 485-491.

1970. Distribution of selected Silurian acritarchs. Rev. Espan. de micropaleont. Spec. Pub. 202 pp.

19706. Acritarchs and chitinozoa from the Silurian Ross Brook Formation Nova Scotia. Jour.

Geology, 78, 745-749.

DiEZ, M. and CRAMER DEL CARMEN, R. 1968. Coiisideracioiies taxonomicas sobre las acritaras del

Silurico Medio y Superior del Norte de Espana las acritarcas acantomortificas. Boletin. I.G.M. Espana,
79, 541-574.

1 970. Acritarchs from the lower Silurian Neahga Formation, Niagara Peninsula, North America.

Canadian Jour, of Earth Sc. 1077 1085.

DEFLANDRE, G. 1942. Sur les hystrichospheres des calcaires siluriens de la Montague Noire. C.R. Acad.
Sci. Paris, 215, 475-476.

1945. Microfissiles des calcaires Siluriens de la Montague Noire. Ann. Paleont. 31, 41-76.

1946. Hystrichosphaerides 111. Especes du Primaire. Fichier micropaleont., ser. 8. Arch. Orig. Serv.

Docum. C.N.R.S., no. 257, parts I V, cards 1096-1185.

and DECLANDRE-RiGAUD, M. 1964. Fichier micropaleontologique generale-serie 12. Acritarches 1

Polygonomorphitae-Netromorphitae pro parte. Centre National de la Recherche Scientifique generale-
serie 13. Acritarches.

and DEFLANDRE, M. 1964. Notes sur les Acritarches. Rev. Micropal. 7, 111-114.

DEUNFF, J. 1954. Veryhachium genre nouveau d’hystrichospheres du Primaire. C.R. Somm., Soc. Geol.
Fr. 13, 305-306.

1954a. Sur un microplancton du Devonien du Canada recelant des types nouveaux d’hystricho-

sphaerides. C.R. Acad. Sci. Paris, 239, 1064-1066.

THUSU; SILURIAN ACRITARCHS

825

DEUNFF, J. 1955. Un microplancton fossile devonien a hystrichospheres du continent Nord-Americain.
Bull. Microsc. Applique, 5, 138-147.

1957. Micro-organismes nouveaux (hystrichospheres) du Devonien de I’Amerique du Nord. Bull.

Soc. Geol. Min. Bretagne, new ser., 2, 5-15.

1958. Micro-organisms planctoniques du Primaire amoricain 1. Ordovicien du Veryhac’h (Presqulile

de Crozen). Bull. Soc. Geol. Min. Bretagne, new ser., 2, 1-41.

DOWNIE, c. 1959. Hystrichospheres from the Silurian Wenlock Shale of England. Palaeontology, 2, 56-71.

1960. Deunffia and Domasia, new genera of hystrichospheres. Micropalaeontology, 6, 197-202.

1963. ‘Hystrichospheres’ (acritarchs) and spores of the Wenlock Shales (Silurian) of Wenlock,

England. Palaeontology, 6, 625-652.

EViTT, w. R. and sarjeant, w. a. s. 1963. Dinoflagellates, hystrichospheres, and the classification of

acritarchs. Stanford Univ. Publ. (Geol. Sci.), 7 (3), 1-16.

and SARJEANT, w. A. s. 1963. On the interpretation and status of some hystrichosphere genera.

Palaeontology, 6, 83-96.

1964. Bibliography and index of fossil dinoflagellates and acritarchs. Geol. Soc. Amer. Mem.

94, 1-180.

EiSENACK, A. 1931. Neue Mikrofossilien des baltischen Silurs. 1. Paleontographica, A. 13, 74-118.

1938. Hystrichosphaerideen und verwandte Formen im baltischem Silur. Z. Geschiebeforschung,

14, 1-30.

1951. Uber Hystrichosphaerideen und andere Kleinformen aus dem baltischen Silur und Kambrium.

Senckenbergiana, 32, 187-204.

1954. Hystrichospharen aus dem baltischen Gotlandium. Ibid., 34, 205-211.

1955. Chitinozoan, Hystrichospharen und andere Mikrofossilien aus dem Bevrichia-Kalk. Ibid.,

36, 157-188, pis. 1-5.

1958. Mikrofossilien aus dem Ordovizium des Baltikums. 1. Markasitschicht, Dictyonema-Schiefer,

Glaukonitsand, Glaukonitkalk. Senck. Leth. 39, 389-405.

1959. Neotypen baltischen Silurhystrichospharen und neue Arten. Paleontographica, A. 112, 5-6,

193-211.

FISHER, D. w. 1953. A microflora in the Maplewood and Neagha Shales. Bull. Bujfalo Soc. Nat. Sci. 21,
13-18.

FOLK, R. L. 1962. Petrography and origin of the Silurian Rochester and McKenzie Shales, Morgan County,
West Virginia. J. Sedim. Petrol. 32, 539-587.

JANSONius, J. 1962. Palynology of Permian and Triassic sediments. Peace River area. Western Canada.
Paleontographica, B. 110, 35-98.

JARDINE, s. and YAPAUDJiAN, L. 1968. Lithostratigraphic et palynologie du Devonien-Gothlandien greseux
du Bassin de Polignac (Sahara). Rev. Inst. Fr. Petrole. 23, 439-468.

LAIRD, H. c. 1935. The nature and origin of Chert in the Lockport and Onondega Formations of Ontario.
Trans. Royal Canad. Inst. 20, 231-304.

LISTER, T. R. 1970. The acritarchs and Chitinozoa from the Wenlock and Ludlow series of the Ludlow and
Millichope areas, Shropshire. Paleontogr. Soc. London (Monograph), 1-100.

LOEBLiCH, A. R. 1970. Moiphology, ultrastructures and distribution of Paleozoic acritarchs. Proc. North.
Am. Paleont. Conven. 1969, 705-788.

MARTIN, F. 1965. Les acritarches du sondage de la Brasserie Lust, a Kortrizk (Courtrai) (Silurian Beige).
Bull. Soc. belge-Geol. 74, 1-47.

1969 (68). Les acritarches de I’Ordovician et du Silurian Beiges. Determination et valeur strati-

graphique. Mem. Inst. Sci. Nat. Belg. 166, 175 pp.

OWEN, D. E. 1969. Wenlockian bryozoa from Dudley, Niagara and Gotland and their paleogeographic
implications. Paleontology, 12, 621-636.

PAUL, c. R. c. 1967. The British Silurian cystids. Br. Mus. Nat. Hist. Geol. Bull. 13, 297-355.

STAPLiN, G. L., JANSONIUS, J. and POCOCK, s. 1965. Evaluation of some acritarch and hystrichosphere genera.
N. Jb. Geol. Palaon. Abh. 123, 167-201.

STOCKMANS, F. and wiLLiERE, Y. 1963. Lcs hystrichospheres ou mieux les acritarches du Silurien beige.

Sondage de la Brasserie Lust a Courtrai (Kortrijk). Bull. Soc. belge-Geol. 71, 450-481.

ST0RMER, L. 1967. Some aspects of the Caledonian geosyncline and foreland west of the Baltic Shield.
Quart. Jl. geol. Soc. London, 123, 182-204.

M

826

PALAEONTOLOGY vqlUME 16

TIMOFEYEV, B. V. 1959. The ancient flora of the Baltic regions and its stratigraphic significance. Trudy Vses.

neft. nauchno-issled. Geol. razv. Inst. (VNIGRI). 129, 350 pp. (In Russian.)

THUSU, B. 1972. Depositional environments of the Rochester Formation in southern Ontario, Canada.
Jour. Sedim. Petrol. 42, 930-934.

WHITE, M. F. 1862. Discovery of microscopic organisms in the Silurian nodules of the Paleozoic rocks of
New York. Amer. Jour. Sci. ser. 2, 33, 385-386.

ZENGER, D. M. 1965. Stratigraphy of the Lockport Formation (Middle Silurian) in New York State. New
York State Museum and Sci. Service Bull. 404, 1-2106.

Bindra Thusu
Institutt for Geologi
Universitetet i Oslo

Revised typescript received 24 January 1973 Oslo 3, Norway

TRILOBITE CUTICLE MICROSTRUCTURE AND

COMPOSITION

by J. E. DALINGWATER

Abstract, Previous literature on the microstructure and composition of the trilobite cuticle is reviewed. The micro-
structure of the cuticle of Asaphus raniceps Dalman sensu Angelin (1854) is described in detail, and a table outlining
the major features of the cuticles of fourteen other trilobite species is included. In Asaphus raniceps and some other
species, two main regions of the cuticle are consistently present: an outer layer characterized by perpendicular
prisms, representing about one-fifteenth of the total thickness of the cuticle and an inner area forming the bulk of
the cuticle. In the inner area of many cuticles, characteristic primary microstructures are fine perpendicular canals,
a variety of wider canals, and horizontal laminae. Three types of tubercle are distinguished from thin-sections.
Inorganic analyses of Asaphus raniceps show that the cuticle consists largely of calcite. Decalcification of this cuticle
with E.D.T.A. left organic residues.

A LARGE number of incidental observations has been made on the trilobite cuticle,
usually during the course of systematic descriptions. However, no author has
studied the cuticle of a wide range of trilobite species in the thorough way that
Lindstrom (1901) and latterly Clarkson (1967, 1969) have examined visual organs.
Attempts at comparing trilobite cuticles with arachnid or crustacean cuticles have
therefore often been based on inadequate evidence, and in some cases recent work on
extant arthropod cuticles has been totally disregarded.

PREVIOUS LITERATURE

Subdivisions of the trilobite cuticle. The terms ‘chitinous’ and ‘integument’ have
frequently been used in descriptions of trilobite exoskeletons. Although chitin is the
most characteristic organic component of arthropod cuticles, it comprises only a
small fraction of the dry weight of calcified cuticles. The use of the term chitinous to
describe dark hard fossil material is particularly inappropriate: pure chitin is colour-
less, soft, and flexible. The term integument includes the cuticle, the underlying
epidermal cells which secrete it, and the basement membrane. In trilobite
exoskeletons, therefore, only the cuticle is found preserved.

Harley (1861) was probably the first to examine the trilobite cuticle critically, while
attempting to establish the systematic position of various ‘Astacoderma’ from the
Ludlow Bone Bed. Thin-sections through the cuticle of Calymene (L. Ludlow)
showed that it consisted of an outer fibrous layer and an inner prismatic layer, both
pierced by canals. However, Harley doubted whether this was the original structural
appearance. In his monograph on the visual organs of trilobites, Lindstrom (1901)
figured sections of the cuticles of many species but made very few comments on their
structure. Cayeux (1916) compared the structure of trilobite and crustacean cuticles,
and St^rmer (1930), who gave a detailed account of the ‘shell’ structure of some
Trinucleidae, recognized four main layers in the cuticle of Tretaspis and compared
these with the four layers in the cuticle of Homarus. The microstructure of the
cuticle of Phacops accipitrinus maretiolensis R. and E. Richter was described by Rome

[Palaeontology, Vol. 16, Part 4, 197.^, pp. 827-8.59, pis. 107-109.]

828

PALAEONTOLOGY, VOLUME 16

(1936) who distinguished three main regions of the cuticle besides observing many
other features. Hupe (1953) considered that three layers in the cuticle might corre-
spond to the epicuticle, pre- and post-exuvial layers of modern cuticles. Harrington
(1959) reviewed the work of St<3rmer (1930), Hupe (1953), and Kielan (1954) and
interpreted the three main layers of St^rmer and Kielan as corresponding to those
of the modern arthropod endocuticle. However, Rolfe (1962) considered that many
of the subdivisions apparent in fossil cuticles were due to diagenetic replacement.
More recently, Majewske (1969) and Horowitz and Potter (1971) reviewed some of
the previous literature on trilobite cuticles, figured some photomicrographs of thin-
sections, but made few observations. Bathurst (1971) has given a more comprehensive
account of previous work.

Primary microstructures. The term ‘primary microstructures’ was introduced by Rolfe
(1962) to describe all structures found in fossil cuticles comparable with those in
modern arthropod cuticles. A brief review of characteristic structures in extant
cuticles has been given by Dalingwater (1973).

Pore-canals: Rolfe (1962) reviewed descriptions of structures from fossil cuticles
resembling the pore-canals of extant cuticles, and concluded that the various small
canals described by St0rmer (1930), Rome (1936), and Evitt and Whittington (1953)
were perhaps too large and too sparsely distributed to be true pore-canals. However,
Harley (1861) had observed that the cuticle of Calymene was traversed by many
straight tubes 0-5-1-8 ij. in diameter, Lindstrom (1901) figured many sections of
cuticles perforated by numerous minute canals, and Balashova (1948) clearly dis-
tinguished very fine ‘canalicules’ from larger canals perforating the carapace of the
Asaphidae.

Larger canals: Rolfe (1962) suggested that many of the larger canals in fossil
cuticles should be termed gland or setal ducts. In extant arthropods such ducts are
generally larger, are more irregularly placed, and occur less frequently than pore-
canals. Larger apertures through trilobite cuticles have been described by Lindstrom
(1901), Richter (1914), Cayeux (1916), Raymond (1920), St^rmer (1930, 1931),
Whittington (1941, 1956, 1962), Ross (1951), Evitt and Whittington 0953), Hupe
(1953), Whittington and Evitt (1954), Richter and Richter (1954), and Harrington
(1959).

Laminae: Nearly all extant arthropod cuticles are horizontally laminated and
individual laminae are 0-2-10 fi thick. Rolfe (1962) reviewed some of the earlier litera-
ture on laminae in trilobite cuticles, commenting on the work of Zittel ( 1 900), Cayeux
(1916), St^rmer (1930), and Rome (1936). However, Harley (1861) had observed
obscure indications of a finely laminated structure in the cuticle of Calymene, Sorby’s
(1879) ‘lines of growth’ are presumably laminae, and Lindstrom (1901) also figured
many finely laminated cuticles.

Tubercles: Tubercles are defined by Harrington, Moore, and Stubblefield (1959)
as small knob-like prominences on any part of the exoskeleton, whereas smaller
structures are termed granules. However, the term tubercle has been used imprecisely
to describe :

(i) Structures involving thinning and doming of the cuticle (Raymond, 1920;

St^rmer, 1930; Kielan, 1954).

DALINGWATER: TRILOBITE CUTICLE

829

(ii) Domed thickenings differentiated to some extent from the main part of the
cuticle (Rome, 1936).

(hi) Discrete structures embedded in cuticle surfaces (Lindstrom, 1901; Walcott,
1921).

Prisms: The surfaces of many arthropod cuticles are covered by a close polygonal
network. Each polygon possibly represents the area of activity of an underlying
epidermal cell and is formed by inward extensions of epicuticlar material at the inter-
faces between adjacent areas of influence (Dennell 1960). Larger, less regular poly-
gonal areas seen in perpendicular sections of calcified cuticles may represent areas of
activity of crystallization centres. Cayeux (1916) figured some relatively large poly-
gonal areas from the cuticles of ' Trinucleus goldfussf and Hupe (1953) mentioned ‘un
fin reseau a mailles polygonales’ in his account of the trilobite cuticle.

Inorganic chemistry and mineralogy. It has often been stated, without critical examina-
tion of the cuticle itself or of the results of previous research, that the trilobite cuticle
contains significant amounts of phosphate. For example, Zittel (1887) described
alternate thin layers of calcium carbonate and phosphate, but Cayeux (1916) found
no trace of these layers and neither did B^ggild (1930) who clearly thought Zittel was
mistaken. Richter (1933) estimated that the trilobite exoskeleton contained up to
30% phosphate, whereas Cayeux’s (1933) detailed analyses indicated that phosphates
were present only in limited quantities. Only Raw (1952) considered that the cuticle
was originally aragonitic, subsequently becoming silicified or coarsely crystallized
to calcite. Probably the most significant contribution to a study of the original
inorganic composition of the trilobite cuticle was made by Stehli (1955). Although he
analysed only one trilobite pygidium, the specimen came from the Middle Permian
Buckhorn asphalt deposit in which original shell mineralogies seem to have been
preserved. X-ray analysis showed that this pygidium was composed entirely of
calcite.

Organic chemistry. Abelson (1954) referred to the presence of three amino-acids in
an Ordovician Calymene, whereas Fujiwara (1963) found no amino-acids in another
species of the same genus.

MATERIAL AND METHODS

Thin-sections provided most of the information for this study but material was
also prepared by E.D.T.A. decalcification and acetate-peel preparation. A pre-
liminary study, using the Scanning Electron Microscope (S.E.M.), was made on
surfaces as well as on broken and etched material. A wide range of trilobites in vary-
ing modes of preservation was examined. The most satisfactory specimens for thin-
sectioning were those preserved in fine-grained limestones, with part of the exo-
skeleton enclosed in matrix. The identity and orientation of these specimens could
thus be determined, while the matrix formed a most satisfactory natural embedding
medium. Where possible, serial sections were made from a single specimen. Random
section series were also made from samples crowded with trilobite remains. Sections
were finished by hand using 1200 mesh carborundum and then left relatively thick
(20-30 fx). Such thick slices offered much detail, but were unfortunately somewhat

830

PALAEONTOLOGY, VOLUME 16

difficult to photograph satisfactorily. All preparations are stored in the Department
of Zoology, University of Manchester, England.

The cuticle of Asaphus raniceps was in addition analysed in a Phillips X-ray
diffractometer, and by X-ray powder photography.

THE CUTICLE OF ASAPHUS RANICEPS

Specimens of Asaphus raniceps Dalman sensu Angelin (1854) were collected from
the lower ‘Raniceps’ Limestone, Haget, Oland, Sweden. About 100 longitudinal,
transverse, and tangential perpendicular sections were made across all the calcified
areas of the exoskeleton. In addition, a preliminary study of the cuticle of this species
was made using the S.E.M. The cuticle of this species is the best documented here and
serves as a model for detailed description.

The cuticle is relatively thick compared with that of modern arthropods of com-
parable size but varies considerably from one area of exoskeleton to another. In
a ten-slide series made from block 01.0.1 the following ranges of thickness were
measured; cephalon 180-450 fx, hypostome 120-560 fx, thorax 100-300 fx, pygidium
160-300 ,x.

Surfaces of the cuticle examined with the S.E.M. show a matrix of calcite prisms
with morphological indications of c-axes orientated perpendicular to the cuticle
surface (PI. 107, fig. 1). In thin-sections an outer layer comprising one-tenth to one-
fifteenth of the total cuticle thickness is seen, composed of fairly regular perpendicular
prisms (PI. 107, fig. 5). However, under crossed-nicols this layer does not extinguish
uniformly (PI. 107, fig. 6) suggesting that although prisms may be orientated with their
c-axes normal to the surface, this orientation is not totally uniform. Broken sections
of cuticle examined with the S.E.M. confirm light microscope observations that the
outer layer is distinct (PI. 107, fig. 2) from the main part of the cuticle (hereafter
termed the inner area). No individual prisms can be seen in the inner area even when it
is examined at a magnification of about x 1,000 with the light microscope. Again,
total extinction does not occur when particular areas of the cuticle become aligned
with the planes of polarization. Under the S.E.M., broken sections of the inner area
suggest fibrous crystallites running tangentially to the surface (PI. 107, fig. 3) but this
impression is not confirmed by etched sections (PI. 107, fig. 4).

The inner area of the cuticle includes fine perpendicular canals approximately
OT ju. in diameter, and occasionally fine parallel horizontal laminae approximately 1 ^
apart can be seen. The cuticle is thicker at the anterior margin of the cephalon, where
fine vertical canals are often emphasized by impregnation with pyrite. Wider canals

EXPLANATION OF PLATE 107

Figs. 1-4. Scanning electron micrographs of librigenal cuticle of Asaphus raniceps Dalman. 1 Upper
surface of cuticle, showing fairly regular calcite prisms, x 1950. 2 Broken perpendicular section of
cuticle, showing discrete outer layer, x 1730. 3 Broken perpendicular section, showing apparent fibrous
crystallites tangential to surface. x420. 4 Etched perpendicular section of cuticle, x 545.

Figs. 5-6. Longitudinal perpendicular section (L.P.S.) of glabella of Asaphus raniceps Dalman showing
outer layer and inner area of cuticle. Slice 01.98.1. 5 Plane-polarized light. 6 Crossed nicols. Xl25.

PLATE 107

DALINGWATER, trilobite cuticle

832

PALAEONTOLOGY, VOLUME 16

approx. 4 ^ in diameter extend to thorn-like projections representing sections through
ridges on the anterior part of the cephalon (PI. 108, fig. 1). This sculpture is even more
pronounced on the cephalic doublure, the forward projecting ‘thorns’ (= sections
of terrace lines) becoming flattened and scale-like with canals opening at their bases.
The terrace lines on the pygidium are seen in thin-section as thickened ridges of
cuticle, but the outer edge of the pygidial doublure is distinctly corrugated. Sculp-
tured ridges on the hypostome are seen in section to be penetrated by canals approxi-
mately 4 /X in diameter.

Several sections were made through the hypostomal maculae where the cuticle is
slightly thicker. Here, in contrast to other areas of the hypostomal cuticle, fine per-
pendicular structures are accentuated by pyrite.

The outer surface of the thoracic cuticle is generally smooth but cuticular sculp-
turing on the pleural doublure is reflected in thin-sections. Where the cuticle deflects
to form the narrow posterior ventral doublure of the thoracic axial rings, the angle is
characterized by an area of darker cuticle. This area may represent differentiation of
the cuticle to permit flexibility between segments additional to that provided by the
soft intersegmental membrane.

Well-preserved material, particularly when freshly broken from the matrix,
exhibits a glossy lustre. This possibly represents an extremely thin outermost layer
not visible in thin-sections. Isolation of this layer proved difficult but decalcification
with a 5% solution of E.D.T.A. (disodium salt) left a residue of brown material with
a clear surface layer. In several specimens this surface layer is regularly prismatic in
some areas, each prism being 6-7 jx across (PI. 108, fig. 2). The brown residue is usually
amorphous but sometimes contains small twisted tubules about 1 jx in diameter.

X-ray analyses of cuticle from the cephalon, thorax, and pygidium indicated a
composition entirely of calcite.

EXPLANATION OF PLATE 108

Figs. 1-2. Asaphus raniceps Dalman. 1 L.P.S. anterior of cephalon. Slice 01.96.1. x27. 2 E.D.T.A.
preparation, showing prismatic surface layer. E.D.T.A. prep. 19. x360.

Figs. 3-4. lUaenus aduncus Jaanusson. Slice Ol.I.la.2. 3 Tangential perpendicular section (Tan. P.S.)
librigena, perforated by large perpendicular canals. These canals are regularly spaced (50-150 p.) and
were observed in all areas of the exoskeleton examined, x 100. 4 Tan. P.S. edge of thoracic pleura,
cuticular sculpturing perforated by ‘hair-like’ structures, x 90.

Fig. 5. Bumastus barriensis (Murchison). L.P.S. pygidial doublure, showing tightly helically coiled fine
perpendicular canals accentuated by pyrite. Slice Dy.B. lb.2. x 100.

Fig. 6. Paladin eichwaldi shunnerensis (King). Transverse perpendicular section (T.P.S.) pygidial margin,
showing larger perpendicular canals which are particularly prominent in this region of the cuticle.
Slice W.s.la.l. x90.

Fig. 7. Cyrtometopus sp. L.P.S. glabella lobe, showing irregular undulating laminae which give the cuticle
of this genus its characteristic ‘cloudy’ appearance. Slice C.(FMB)L2. x90.

Fig. 8. Encrinurus pimctatus (Wahlenberg). L.P.S. glabella, showing large tubercles~the cuticle domes
and thins. Tubercles of this type are also present in axial regions of the thorax. Small discrete tubercles
are present on lateral areas of the cephalon. Slice Dy.E.7.1. x24.

PLATE 108

DALINGWATER, trilobite cuticle

834

PALAEONTOLOGY, VOLUME 16

THE CUTICLES OF OTHER TRILOBITE SPECIES

Brief descriptions of the cuticles of fourteen trilobite species are deposited with the
British Library, Lending Division, Boston Spa, Yorkshire, reference number
SUP 14001. These are illustrated in Plates 108 and 109, and Table 1 summarizes
their characteristic features.

TABLE I . Synopsis of the characteristic features of the cuticles of fourteen trilobite species.

Material

examined

Cuticle

thickness

(»*)

Outer

layer

Larger

perpendicular

canals

Fine

perpendicular

canals

Laminae

Special

feacures

Agnostus pisiformis (Linnaeus)
Middle Cambrian, Gland.

c.,p.

7-30

-

+

-

+

-

Nileus armadillo Dalman
*Raniceps' Limestone, Gland.

c./s.

100-200

-

-

+

.

+

Illaenus aduncus Jaanusson
'Raniceps' Limestone, Gland.

c./s.

180-700

+

+*

+

+

+*

Bumastus barriensis (Murchison)
Wenlock Limestone, Dudley.

c.,p.

300-500

-

+

+

+

-

Cyrtosymbole pusilla (Giirich)
Famennian Limestone, Poland.

s.

30-100

+

+

-

.

+

Paladin eichwaldi shunnerensis (King)
Shunner Fell Limestone, Yorkshire.

p-

150-200

+

*

+

+

+

-

Ampvx nasutus Dalman

Expansus 6. 'Raniceps* Limestones, Gland

. c.

100-200

+

-

+

+

+

Cyrtometopus clavifrons (Dalman)
'Raniceps* Limestone, Gland.

c.

200-340

-

+

-

+

+

Encrinurus punctatus (Wahlenberg)
Wenlock Limestone, Dudley.

c./s.

o

o

o

+

+

+

+

*

+

Calymene blumenbachii Brongniart
Wenlock Limestone, Dudley.

c./s.

200-400

+

+

+

+

-

Phacops granulatus (Munster)
Famennian Limestone, Poland.

s .

100-600

-

+

-

.

*

+

Trimerocephalus caecus (Giirich)
Famennian Limestone, Poland.

s.

120-270

-

+

_

.

*

+

Acaste downingiae (Murchison)
Wenlock Limestone, Dudley.

c .

160-300

+

+

-

-

+

Boedaspis ensifer Whittington & Bohlin
Expansus Limestone, Gland.

p-

200-250

+

-

+

-

*

+

KEY c. = cephalon,

+ = present.

p. = pygidium, c.

- = not observed.

./s.

« =

complete spec
f Igured.

imens,

s . =

slides,

EXPLANATION OF PLATE 109

Fig. 1. Calymene blumenbachii Brongniart. L.P.S. glabella lobe, showing a range of perpendicular canals
accentuated by pyrite, some of which are similar to the fine canals of other trilobite cuticles. Slice
Dy.C.1.2. xl60.

Fig. 2. Phacops granulatus (Munster). T.P.S. edge of pygidial axis, showing large tubercle; the cuticle
domes and thickens. Slice P.g.(0)l. x92.

Fig. 3. Trimerocephalus caecus (Gunch). T.P.S. large tubercle on edge of librigena. Slice T.c.(0)3. x92.

Fig. 4. Acaste downingiae (Murchison). L.P.S. discrete tubercles on cephalon. Note spine-like structure
arising from one tubercle, also canals which may serve the tubercles. Slice Dy.A.d.1.3. X 135.

Figs. 5-8. Boedaspis ensifer Whittington and Bohlin.

Figs. 5-7. Tan. P.S. edge of pygidium, showing three types (or aspects?) of tubercle involving doming
and thinning of the cuticle. Slice Ol.B.e.lb.2. All x85.

Fig. 8. Tan. P.S. edge of pygidium, showing discrete tubercle. Slice Ol.B.e. la. 1. xl35.

PLATE 109

DALINGWATER, trilobite cuticle

836

PALAEONTOLOGY, VOLUME 16

DISCUSSION

Subdivisions. The calcitic composition of A. raniceps cuticle agrees with the observa-
tions of B^ggild (1930), the detailed analyses of Cayeux (1933), and Stehli’s (1956)
analysis of a pygidium composed entirely of primary calcite. In most trilobite cuticles,
replacement and infiltration of the original inorganic material has probably taken
place ; this may have been partly inhibited by the organic component. Diagenesis of the
calcite in the eye-lenses of A. raniceps has been described in detail by Clarkson (1973).

Many previous authors including Sorby (1879), Cayeux (1916), and Majewske
(1969) have stated that thin-sections of trilobite cuticle extinguish uniformly when
viewed with crossed-nicols, but here it is shown that in A. raniceps cuticle (and in
most other cuticles examined) extinction is not totally uniform. Moreover, examina-
tion of thin-sections of cuticle with the light microscope, and broken sections with
the S.E.M. indicates that calcite prisms with any definite orientation occur only in
the thin outer layer.

St0rmer (1930) suggested that the four cuticular layers of Tretaspis were directly
comparable with the major structural subdivisions in Homarus. However, it is diffi-
cult to envisage how an inner uncalcified area of endocuticle (equivalent to the inner
layer in Homarus) might be preserved when the ventral trilobite cuticle is so rarely
encountered. Moreover, an outer epicuticle, if present, is unlikely to be seen in thin-
sections of trilobite cuticles prepared by conventional petrological methods. Further-
more, Styirmer’s ‘pigmented layer’ is a dense micritic envelope (J. Miller, pers.
comm.). Although Richards (1951) established that the subdivision of any arthropod
cuticle must be based mainly on histochemical distinctions, Hupe (1953) and Harring-
ton (1959) compared the various layers described by previous authors directly with
those of extant arthropod cuticles. Harrington’s statement that: ‘the exoskeleton of
trilobites consists of a thin integument that is directly comparable to the chitinous
cuticle of other Arthropoda’, seems particularly inappropriate. In contrast, Rolfe’s
(1962) comment that many subdivisions in fossil cuticles are the result of replacement,
may be pessimistic. In the present study the only consistent divisions of the trilobite
cuticle seen in thin-section are a thin outer prismatic layer and an inner area. The outer
layer, comprising one-tenth to one-thirtieth of the total thickness, was not observed
in all cuticles studied, possibly because it is easily eroded and the boundary between
it and the inner area seems to be a natural plane of weakness. In broken sections of
cuticle examined with the S.E.M., the outer layer appears superficially similar to the
tanned calcified exocuticle of Carcinus niaenas, where the prominent perpendicular
elements are pore-canals. Clarkson (1973) has described how the outer layer of the
trilobite cuticle corresponds with the cornea ; in some modern decapods, the tanned
calcified exocuticle also laterally merges with the cornea. The appearance of the
prismatic network of thin transparent material obtained on decalcification of A.
raniceps cuticle resembles that of the epicuticle of some extant arthropods. In the
latter, the walls of the regular prisms probably reflect the boundaries of the hypo-
dermal cells responsible for secretion of the cuticle (Dennell 1960).

Thus, although there is some evidence for correlating the subdivisions seen in
some trilobite cuticles with those of extant decapod cuticles, no precise comparisons
can be made in the absence of histochemical data.

DALINGWATER; TRILOBITE CUTICLE

837

The cuticles of some trilobites (e.g. Agnostus) are thin, but most species studied
have thick cuticles compared with modern arthropods. Although the general cuticle
thickness is rarely greater than 500 the cephalic cuticle of some large trilobites
sometimes exceeds 1 mm.

Canals. Fine perpendicular canals, usually less than 1 /x in diameter, occur in the
majority of cuticles studied. They are a characteristic feature of the cuticle and occur
in large numbers closely packed together. In some cuticles, they appear to be helically
coiled. Previous workers, notably Cayeux (1916), have compared these fine canals in
the trilobite cuticle with those in modern arthropod cuticles. They are more evident
in some species or only locally prominent. Larger canals appear to be more prone to
pyrite impregnation and are thus accentuated. The prominence of fine canals in
certain areas may also be due to preferential pyrite impregnation.

However, it is more difficult to understand why canal-like structures are accentuated
in specialized areas of cuticle such as the hypostomal maculae, dark spots, the
cuticle surrounding the eyes, and the anterior margin of the cephalon. Raymond
(1920) suggested that the maculae and dark spots represent muscle-attachment sites.
In contrast, Lindstrom (1901) thought that the hypostomal maculae were rudimentary
optic organs because in many species the marginal areas of the eyes and the maculae
have a similar aspect. Furthermore, he observed that small, closely spaced ‘lenses’
were present on the maculae of some trilobites. Balashova (1948) also noted the
similarity of the eye-margins and maculae in the Asaphidae, but disagreed with
Lindstrom’s conclusions and suggested instead that both areas were characterized in
life by dense concentrations of sensory setae. Harrington (1959) also thought it
unlikely that the hypostomal maculae represented areas of muscle-attachment,
partly because he supposed that the ‘mineralized integument’ was thinner at the
maculae. In fact, the cuticle thickens at the maculae in the Asaphidae, where per-
pendicular elements are most prominent in this area. It is suggested here that all these
areas were sites of muscle-attachment, and that some of the perpendicular elements
may represent tonofibrillae (cuticularized muscle fibres). Their frequent occurrence
in multiple units may explain the rudimentary lens-like structure inferred by Lind-
strom (1901). In recent arthropod cuticles it is often difficult to distinguish tono-
fibrillae from pore-canals (Dr. J. H. Kennaugh, pers. comm.).

A variety of wider canals was observed in the trilobite cuticles studied, some of
which are remarkably similar to canals in modern cuticles. However, because con-
clusive evidence such as the remains of sensory hairs is rare, it is difficult to assess
whether these wider canals represent tegumentary or setal ducts. Wide canals are
often prominent at the extremities of the exoskeleton. In extant decapods similar
regions are plentifully supplied with tegumental glands (Dennell 1960), but in trilo-
bites these canals were probably sensory rather than associated with extensive
phenolic tanning. Contrary to Evitt and Whittington (1953), wide canals are present
even in ‘smooth-shelled trilobites’.

Laminae. The fine parallel laminae described from a few trilobite cuticles are possibly
comparable with those in extant decapod cuticles. However, microfibrils were not
seen even in the best-preserved material, and it seems more likely that examination

838

PALAEONTOLOGY, VOLUME 16

of well-preserved eurypterid cuticles will extend studies of cuticle architecture back
into the Palaeozoic.

Tubercles. The various types of tubercles described by previous authors have been
recognized in the cuticles studied. Although the term ‘tubercle’ has been used indis-
criminately to describe various structures in modern arthropod cuticles, several
authors, notably Kennaugh (1968), have used the term more precisely to describe
discrete structures embedded in the cuticle of arachnids. However, as it is difficult to
determine which type of tubercle is present without sectioning the cuticle, it seems
premature to introduce terminology to distinguish these various structures.

Rome (1936) indicated that Phacops accipitrinus maretiolensis was immediately
identifiable in thin-section but doubted whether other trilobites could be distinguished
so satisfactorily. However, certain species have characteristic, perhaps unique,
cuticles which might enable specific identification to be made from a small fragment.
When the cuticles of many trilobite species have been described in detail (including
variations in structure due to different modes of preservation) this information may
prove useful, particularly in borehole work.

Acknowledgements. I wish to thank Professor R. Dennell for originally suggesting this area of research;
Mr. J. Miller and Mr. M. Downes for help during the later stages of the work; Drs. F. Broadhurst, E. Clark-
son, and H. Osmolska for kindly donating material; and Drs. E. Clarkson and B. Taylor for helpful
criticism of the manuscript. My thanks are also due to the staff of the S.E.M. Unit, Dept, of Textile Techno-
logy, U.M.I.S.T., and to Messrs. L. Lockey and B. Atherton for photographic work. Some of this work was
completed during the tenure of an S.R.C. Studentship, which is gratefully acknowledged.

REFERENCES

ABELSON, p. H. 1954. Amino acids in fossils. Science, 119, 576.

BALASHOVA, E. A. 1948. On the tactile organs of trilobites. Dokl. Akad. Nauk SSSR, 61, 509-11. (In Russian)
BATHURST, R. G. c. 1971. Carbonate Sediments and Their Diagenesis, 620 pp. Elsevier.

B0GGILD, o. B. 1930. The shell structure of the mollusks. K. danske Vidensk. Selsk. Skr. 9, 231-326.
CAYEUX, L. 1916. Introduction a Tetude petrographique des roches sedimentaires, 524 pp. Imprimerie
nationale: Paris.

1933. Role des trilobites dans la genese des gisements de phosphate de chaux paleozoiques. C.r. hehd.

Seanc. Acad. Sci., Paris, 196, 1179-1182.

CLARKSON, E. N. K. 1967. Fine structure of the eye in two species of Phacops (Trilobita). Palaeontology, 10,
603-16.

1969. On the schizochroal eyes of three species of Reedops (Trilobita: Phacopidae) from the Lower

Devonian of Bohemia. Trans. R. Soc. Edinh. 68, 183-205.

1973. The eyes oi' Asaphus raniceps Dalman (Trilobita). Palaeontology, 16, 425-444.

DALiNGWATER, J. E. 1973. The culiclc of a eurypterid. Lethaia, 6, 179-186.

DENNELL, R. 1960. Integument and exoskeleton. In waterman, t. h. (ed.). The Physiology of Crustacea, 1,
449-472. Academic Press.

EViTT, w. R. and Whittington, h. b. 1953. The exoskeleton of Flexicalymene (Trilobita). J. Paleont. 27,
49-55.

FUJiWARA, T. 1963. Palaeochemical studies on the organic substance remaining in various sorts of fossils.

A/ Ac. Rep. Res. Inst. nat. Resourc., Tokyo, 58-59, 139-149.

HARLEY, J. 1861. On the Ludlow Bone-Bed and its crustacean remains. Q. Jl. geol. Soc. Fond. 17, 542-552.
HARRINGTON, H. J. 1959. Microstructure of exoskeleton. In moore, r. c. (ed.), Treatise on Invertebrate
Paleontology O, Arthropoda, 1, 085-7. University of Kansas Press.

DALINGWATER: TRILOBITE CUTICLE

839

HARRINGTON, H. j., MOORE, R. c. and STUBBLEFIELD, c. J. 1959, Morphological terms applied to Trilobita.
In MOORE, R. c. (ed.), Treatise on Invertebrate Paleontology O, Arthropoda 7, 0117-126. University of
Kansas Press.

HOROWITZ, A. s. and potter, p. e. 1971. Introductory Petrography of Fossils, 302 pp. Springer-Verlag.
HUPE, p. 1953. Classes des Trilobites. In piveteau, j. (ed.), Traite de Paleontologie, 3, 44-246. Masson:
Paris.

KENNAUGH, J. H. 1968. An examination of the cuticle of three species of Ricinulei (Arachnida). J. ZooL,
London, 156, 393-404.

KiELAN, z. 1954. Les Trilobites mesodevoniens des Montes de Sainte-Croix. Palaeont. pol. 6, 1-50.
LiNDSTROM, G. 1901. Researches on the visual organs of trilobites. K. svenska Vetensk-Akad. Handl. 34, 1 -89.
MAJEWSKE, o. p. 1969. Recognition of Invertebrate Fossil Fragments in Rocks and Thin Sections, 101 pp.
E. J. Brill; Leiden.

RAW, R. 1952. A note on Ross. ‘Ontogenies of three Garden City Trilobites’. J. Paleont. 26, 854-857.
RAYMOND, p. E. 1920, The appendages, anatomy, and relationships of trilobites. Mem. Conn. Acad. Arts
Sci. 7, 1-170.

RICHARDS, A. G. 1951. The Integument of Arthropods, 41 1 pp. University of Minnesota Press; Minneapolis.
RICHTER, R. 1914. Neue Beobachtungen iiber den Bau der Trilobitengattung Harpes. Zool. Anz. 45, 146-152.

1933. Crustacea (Palaontologie). In dittler, r. et al. (eds.), Handworterbuch der Naturwissen-schaften,

2nd edn., 2, 840-864. Gustav Fischer: Jena.

and RICHTER, E. 1954. Die Trilobiten des Ebbe-Sattels und zu vergleischende Arten. Abh. senckenb.

naturforsch. Ges. 488, 1-76.

ROLFE, w. D. I. 1962. The cuticle of some Middle Silurian Ceratiocaridid Crustacea from Scotland.
Palaeontology, 5, 30-51.

ROME, D. R. 1936. Note sur la microstructure de Fappareil tegumentaire de Phacops (Ph.) accipitrinus
maretiolensis R. and E. Richter. Bull. Mus. r. Hist. nat. Belg. 12, 1-7.

ROSS, R. J. 1951. Stratigraphy of the Garden City formation in North East Utah, and its trilobite fauna.

Peabody Mus. Nat. Hist., Yale Univ., Bulletin, 6, 1-161.

SORBY, H. c. 1879. Anniversary address of the President. Q. Jl geol. Soc. Land. 35, 56-93.

STEHLi, F. G. 1956. Shell mineralogy in Palaeozoic invertebrates. Science, 123, 1031-1032.

ST0RMER, L. 1930. Scandinavian Trinucleidae with special reference to Norwegian species and varieties.
Skr. norske Vidensk Akad. Mat.-naturv. Kl. 4, 1-111.

1931. Boring organisms in trilobite shells. Norske geol. Tidsskr. 12, 533-539.

WALCOTT, c. D. 1921. Notes on the structure of Neolenus. Smithson, misc. Colins. 67, 365-456.
WHITTINGTON, H. B. 1941. Silicified Trenton trilobites. J. Paleont. 15, 492-522.

1950. Sixteen Ordovician genotype trilobites. Ibid. 24, 531-565.

1956. Silicified Middle Ordovician trilobites; the Odontopleuridae. Bull. Mus. comp. Zool. Harv.

114, 155-288.

1962. A natural history of trilobites. Rep. Smithson. Instn. for 1961, 405-414.

and EViTT, w. r. 1954. Silicified Middle Ordovician trilobites. Mem. geol. Soc. Am. 59, 1-137.

ziTTEL, K. A. 1887. Traite de Paleontologie (transl. C. Barrois). 2, 897. Doin: Paris.

1900. A Textbook of Palaeontology (transl. and ed. C. R. Eastman). 1, 706. Macmillan: London.

JOHN E. DALINGWATER
Department of Zoology
The University
Manchester M13 9PL

Revised typescript received 2 March 1973

THE PALAEONTOLOGICAL ASSOCIATION

Annual Report of the Council for 1972

Membership. Over the past two years the ordinary membership has shown a significant rise to c. 760 members
from the previous plateau of c. 710 members reached in 1966. This improvement may in part be due to the
enlarged circulars with their more informative contents. On 31 December 1972 there were 1321 members
(771 Ordinary, 138 Student, and 412 Institutional), a net increase of 85 members during the year.

Finance. During 1972 the Association published Volume 15 of Palaeontology at a cost of £13,221 and
Special Papers 10 and 11 which are expected to cost £4,758. This printing bill of £17,979 is once again the
highest in the Association’s history, a reflection of continual price increases in the printing trade.

Total expenditure was £19,038, a decrease over the previous year since we reprinted no back numbers
of Palaeontology or an Index. An excess provision of £895 for 1971 artificially decreased this year’s figure.
General income was satisfactory, but the high level of subscription income is partly the effect of the postal
strike in early 1971, and represents subscriptions held over for a year. Sales of Palaeontology were encourag-
ing at £7,452, an increase of over £3,000 from 1971, although the latter figure was also adversely affected
by the postal strike. The Association is very grateful to those universities and other institutions who gave
funds to support individual papers, and in particular to the Royal Society for providing a gift of £400 and
a loan of £600 in support of Special Paper 11. The more direct grants that papers receive the more the
Association can publish.

Although there was a welcome excess of income over expenditure during 1972, our publication reserves
were still lower than two years previously. Indeed the reserve of £ 1 2,47 1 , although at first sight a high figure,
is still well below our aim of being one year’s printing costs of Palaeontology alone. In particular more
subscribers, both institutional and private, are needed to Special Papers before that series can become self-
financing.

Publications. Four parts of Palaeontology were published during 1972; they contained 47 papers and
consisted of 693 pages and 132 plates. The cover of Palaeontology was redesigned ruring the year at Oxford
University Press, and was issued from Volume 16 part 1. Special Papers in Palaeontology 11 (for 1972)
was published, and Special Paper 12 (for 1973), the Cambridge symposium volume ’Organisms and con-
tinents through time’, early in 1973. Advance publicity matter for this Special Paper, widely distributed in
N. America through the good offices of Dr. Ellis Yochelson, the S.E.P.M., and the Paleontological Society,
resulted in many orders. A sales campaign to promote the Special Papers generally was mounted during the
year, most of the work being undertaken by Dr. W. J. Kennedy. It is hoped that members will encourage
subscription to this series.

The List of British Palaeontologists referred to in the last annual report was issued to Ordinary and
Student Members during the year. All the work entailed in the compilation of the volume, and the costs of
production, were borne by Robertson Research International Limited, and the Association is greatly
indebted to Dr. R. H. Cummings and his staff for their efforts so generously given.

Meetings. Six meetings were held during 1972-3. The Association is greatly indebted to Professor P. C.
Sylvester-Bradley (Leicester University), Dr. R. H. Cummings (Robertson Research International Limited),
and Professor E. A. Vincent (Oxford University) for granting facilities for meetings, to Dr. C. T. Scrutton
for leading the field meeting and to the local secretaries for their efficient services.

a. The Fifteenth Annual General Meeting was held in the rooms of The Geological Society of London
on Wednesday, 1 March 1972. The Annual Report of the Council for 1971-2 was accepted and the
Council for 1972-3 elected. Dr. C. Downie of Sheffield University delivered the Fifteenth Annual
Address on The Palaeozoic acritarchs’.

h. A Field Demonstration Meeting on ‘Devonian corals and stromatoporoids of the Torbay area’ was
led by Dr. C. T. Scrutton on 6 May 1972.

c. A most successful innovation during the year was a Teach-in ‘Analysis of processes in carbonate
environments’, organized by Dr. Julia A. E. B. Hubbard and led by Dr. R. G. C. Bathurst. Numbers
N

842 THE PALAEONTOLOGICAL ASSOCIATION

were restricted to 40 (including the 1 1 speakers), but were drawn from a wide sector of interests, and
from a wide geographic area. This cross-cutting meeting gained many new friends for the Association
from neighbouring disciplines.

d. A Symposium on The application of electron microscopy to palaeontology’ was held at EMCON 72—
the 5th European Congress on Electron Microscopy, at Manchester University on 1 1 September 1972.
About 60 people attended, 12 papers were presented, some of which will appear in ‘Palaeontology’.
The meeting was organized by Dr. Marjorie D. Muir.

e. A Join! Demonstration Meeting on ‘The electron microscope in micropalaeontology’ was held with the
British Micropalaeontological Group at Eeicester University on 25 October 1972. This was a sequel
to the above meeting, and 22 demonstrations were provided. The local secretary was Mr. R. Clements.

f. An Open Discussion Meeting was held at Oxford University on 17-20 December 1972. Over 120
people attended to hear 24 papers, to see the remarkable film of echinoid behaviour brought along by
Dr. Porter Kier (arrangements are in hand to obtain a hire copy for the British Eilm Institute), and to
view the 20 demonstrations. This was the first of a new style of open meeting organized by the Associa-
tion; its undoubted success will encourage the adoption of this pattern for future annual meetings.
A welcome feature of this year’s meeting was a choice of two field excursions, run jointly with the
British Sedimentological Research Group. The local secretary was Dr. W. J. Kennedy.

Council. The following were elected members of Council of the Association for 1972-3 at the Annual
General Meeting on 1 March 1972: President: Professor M. R. House. Vice-Presidents: Dr. Gwyn Thomas,
Mr. N. E. Hughes. Treasurer: Dr. J. M. Hancock (Deputy Treasurer during Treasurer’s absence abroad:
Dr. E. R. M. Cocks). Membership Treasurer: Dr. A. J. Lloyd. Secretary: Dr. W. D. I. Rolfe. Editors:
Dr. I. Strachan, Dr. R. Goldring, Dr. J. D. Hudson, Dr. D. J. Gobbett, Dr. L. R. M. Cocks. Other members:
Dr. M. G. Bassett, Dr. E. N. K. Clarkson, Dr. R. H. Cummings, Professor D. L. Dineley, Dr. Julia A. E. B.
Hubbard (Circular Reporter), Dr. J. K. Ingham, Mr. M. Mitchell, Dr. Marjorie D. Muir, Dr. B. Owens,
Dr. W. H. C. Ramsbottom, Dr. P. Rawson, Dr. P. L. Robinson, Dr. A. D. Wright.

Circulars. These continued in the enlarged form instituted last year, and there was evidence that the
fuller information provided was welcomed; as a result of the Circular Reporter’s initiative, copies were now
sent to 99 Institutional Members and interested bodies. Pour Circulars (nos. 69-72) were distributed
during the year.

Council activities. Besides continuing the planning of the meetings throughout the year. Council has
given thought to the following new ventures. On the suggestion of the Stimulus Committee it was agreed
that an International Symposium on ‘The Ordovician System’ should be held at Birmingham in 1974.
Plans for the meeting and publications have been made by the sub-committee appointed for this task
(convener Dr. A. D. Wright) and advance publicity circulated. It was decided to institute a series of hand-
books of fossils of various formations, under the general editorship of Dr. J. D. Hudson. The first volume
will be ‘Fossils of Wren’s Nest, Dudley’ by Dr. I. Strachan.

Besides being represented on the British National Committee for Geology, the Association is now repre-
sented on the Botany and Zoology subcommittees of the National Committee for Biology by Professor
W. G. Chaloner and Professor M. R. House respectively. During the year, the Association applied for and
was recognized by The Geological Society of London as a society, ten years’ membership of which entitles
new Fellows to exemption from the usual Admission Fee.

BALANCE SHEET AND ACCOUNTS
EOR THE YEAR ENDING 31 DECEMBER 1972

o

o

rn

'O r-1

ob

o

o

o

o o

C 4 'sO o o

wb rC o rC

ON ro O ^

^ ON

^ <N rn

m NO o

ON OO o

rb NO O

^

o — ^

n

NO

m

o

o

NO

r-

rC

00

o

NO

rb

ir^

ir{
r j

O-

KTl

H

W

OD

00

<

s

:3 <L>

cr b
W S

(U (1)

O ,5-

e ^

o

Os
• ON

. cS^

' <73 0^

£>

5.^1

! o Si

: t. H

) <U '■

il§

^ r4

pi— I ,

u :

c

c3

_C

■ i£

S: S ^

Os

^ o E

I

U

(J d
(J D "rt

'-' X)

O Xi 3
X) i- -a

►S

r-'

ON o
^ U
ra

xi t;

5=! o

S

§ S

-1 ^ I;
o
o
n

5 c/3 '-«'

§ .2^ u

OXJ

o

o • H

W)?S

ON

P. O

U O Os

^ 'Vm

, • a

P--C
o O

^ s

o> Ci

' c ^
£
c

c

0£)

a

o

o

W

o

o

U

<4^

-4 b f_:

c/D ^

c 2^ p^

O 4-* c/5 >

OX) ’S c
do —
-P ^

s -

o c
y o

iC 'X

> w

2

"S =*

Ui

o H

0)

.£ *o

l-H c

S20 Si

u- X

O c3

X)

^ c/3

H

C (U

— o c
^:S o

^ 3 c
§<^
C/D

G

cd

CQ

cd

U

NO ON
ON ON
cn

X

00

o

o

X

O O cn

o o o

rn r^> ^

O

o

<

m

O ON
r<“j ro

*G-

OO*" rn

O O
ri r4

rn r4“j
NO X

00 CTi

^ o

r-- r<)
ON

00

w

CQ

<

Q s
8

OC a

^ n
o (L> ^
•rl O c/3
b C c/3

S i2 8

cd X
QQ PJ
CL,

CT3

CJ) ^

o

G ^

ii'S
< t:

. Si 3
3 .S

’3 pd*

S ^

X dJ
CL u

w ^

c/3

1)

.2 c

■G

Cl >

"S

O G
c/5 _

.S

G

00

o S

c/3 k5

O Cl,

t3 1-
C 3
G .b
0> 'O

s g

O D,

►S w

C<3
Cn
Ci;

<! -a

dj

<\ Cl

^ d G

^ I C

<o G • —

2:* ^

Ci, CX 73
^ ^ 03
oj:j UJ o
.£ xj
■G G ^
G G 0>
'C <L> b
Cl G d

Q O ^

— G (U

•2 b ^

•g aW

C c/3 ^

P 3 2

•-1

C >>
-ti
o K
►J g

i"-

o >5

'o I—'
O a>
C/!) 73

'g ^

OJ

O TO

Qi <

(L>

u

G rv
Cl cd (i>

<U 0- ,<L)

2

G c3

d| g 'I

I < c^<

G

C/5

rO

o r-

c/^ ON

p

O

"G-

-G*

NO

rb

‘T-)

ITN

ri

131 Deposit Account . . 4,372-54

2,591 Current Account . . . 5,81519

Income and Expenditure Account for the Year Ended 31 December 1972

m

^

as

^ r-i (N
m

os r--"

00

00

so

rn

^ O OS OO OS
so o OS so
^ VO " o
so os (N so
oo (N
^ rn

(N

^ O
ro

<n OS

os^ r-;
r-T ^

o

ob

m o i/D o o

r-i O m O ^

^ ub ob lib

in i/^ oo Os

ri

m O O O O O rn

os in r- o O O

rb O O O in

r^i so m m nj os •— '

04

04

ob

oo

W

s

o

u

Z

x>

c/5

>.

CQ

cs O

Cl

fc

^ rr\

5-«

T3 2
<U

> ^
ID folrt

- u

c/5 r '

2i S;

B J

c c •

C3 • ■2

*= ^ -rl

Ph -o §
^ C 3
^ Sm

O -n -r:

QJ >

b£) C

£P c

1 s

b£)

b o

^ Ui

^ H
•c ^

Oil <D

< z

H

c g

h— I C3

. . o

"o O >— 1

^ Q.^S;
;□ L> ""

1^.

Os o

OS Xi

U
P-i

o -i= z u

U. Pii (J -P

CP PC

OS

- fn CJ

■

cd O •♦-»
o-^
o

^ . cd a

O D

u.

03

o

cd

o

Ofi

2

o

a.

p

cd

c/5 Q_

S ^

.2

5 cd
C ■ 5^

Q S-
m

'o S5

o

O. (U

O c
o cd
o:;
o <D
o B

cd

C/^ ^

ro ro so O)

so 04 in

O^ oo^ in

O- '=ct

o

04

mm I 0^4

^ o I ^

m r^

m

os

O

OO 04
SO OS

O

04

m

m

so

oo

m os
m m
00 os

m

m

pj

qc

D

H

5

z

UP

a.

Z

PP

SO

m

0

0

0

0

0

0

SO

0

0

0

—1

so

0

00

0

0

qs

0

0

m

Os

04

0

00

0

m

00

m

SO

m

qs

04

so

0

0

ob

rb

ob

0

ob

m

X

rb

0

sb

C3S

lib

ob

ob

m

lib

ds

bt-

m

0

0

04

os

m

0

m

os

00

0-

m

00

O'

00

0

00

m

m

04^

m^

m

m

0^4^

m^

0-^

00

»— 1

0-

04

m

m

mT

m*'

m

m"

X

3

a

CN

G3

CP CP CP

O

H

cd

Ct

O

>

"O

'O

<

XJ X)
c D

Oh
■ O Sh^

C O
cd ^

O

>

o —

.b _o

.b 'X Qj • •

b ^ Cl. o o

£- S Z z Z

Q- 1—

O O

H H

X3

C

cd

>A ^

2

to -2 ^

o ■•-' "O o
fv 3 3 ,l3 PP O
Q C/5 <C U S X

H

s

3

czi c/5 *:;■

‘ 3

C i-M

a o

(U c

o

cd ^

c/5 C
2

K E

b ID

^ T3
<

S-
2 •

o

^ o

n ^
to O

' $5

CJ -o

S ^

X) o

'o ^
w o

<D C
cjj Cd
cd i-

S 3
^ 2

2 c
o U
o .2

C/5 ^

5- 0“

CD S
2

o ^

a c
o

.2

b ^ 2
-o 2 o
rb (u

toO

Q 2 Q

XJ

c

D

a

X

uo

o

X

PJ

oo oo

so O

mj m

04 04"

o

m

so

Tf

m

o-

Os^

m

so

04

00

os^

04*^

04

H

b

a

a

o

U

ai

J

m

VO

<N

00

Os^

04"^

04

'O 04

^ 5^

Cd os

^ T

S D

O -c
b E

2 D

o

"O D

<D n

X

<D -M
b m

c ^

cd Cd

55

c 3

Cd c/5

Ui

ID 'b
D ^

•S 3

C5

u

oi

u

u

<

CQ

-3 °
(U u

^ b

rtl

> o

cd .

c: x3

=3 b
^ cd
O (U

- cd XJ

i; Z“ 3
ic

^ .b

2 3 o

S- D ^

0 CL-O

^ “

:l> W) Cd

S b ^

.2 u

? 2"

1

S 3 i>

III

c X "c
~ CJ <u

3 X

~ ^ s

C "O

3 3 C

u c

X X

S O

■ g-'B

PP 3

Sheffield Chartered Accountants, Auditors

INDEX

Pages 1-222 are contained in Part 1 ; pages 223-424 in Part 2; pages 425-650 in Part 3; pages 651-849 in Part 4.
Figures in Bold Type indicate plate numbers.

A

Acanthodiacrodiwn sp., 24, 25.

Acaste downingiae, 109.

Achomosphaera cf. hyperacantha, 693, 87.

Acritarchs : Campanian of Alberta, 665 ; observations on
their nature, 239; Precambrian Chuaria, 535; Silurian
of Ontario, 799.

Adams, C. G., Knight, R. H. and Hodgkinson, R. L.
An unusual agglutinating foraminifer from the Upper
Cretaceous of England, 637.

Admollia, 418; amphidoxa, 420, 45-47.

Alberta: Campanian dinoflagellate cysts and acritarchs,
665,

Albida ohesa, 299, 31.

Algae: solenoporoid from Miocene, 223.

Allen, P., Keith, M, L., Tan, F. C. and Deines, P.
Isotopic ratios and Wealden environments, 607.

A Inns, 103.

Ammonidium \ cf. microcladum, 105; sp., 27.

Ammonoidea: aptychi of Bajocian Sonninia, 195;
buoyancy control and siphuncle function, 623.

Ammoscalaria pseudospiralis, 99.

Amphibian: Carboniferous of Scotland, 179.

Anderson, E. J. and Makurath, J. H. Palaeoecology of
Appalachian gypiduhd brachiopods, 381.

Anodonta cygnea, 60.

Apteodinium sp. A, 669, 84.

Aptilechinus, 652; caledonensis, 654, 80-83.

Archegonus (IVeania) feltrimensis, 391.

Aremoricanimi sp.. 26.

Arthropoda : Carboniferous trilobites from Ireland, 39 1 ;
Carboniferous trilobites from Somerset, 551 ; eyes of
Asaphus rcmiceps, 425 ; Jurassic hermit crabs symbiotic
with ectoprocts, 563; morphology and evolution of
the eye in Upper Cambrian Olenidae, 735; new
Eocene crab from Libya, 283; Ordovician trilobites
from New South Wales, 445; trilobite cuticle micro-
structure, 827,

Asaphus raniceps, cuticle, 830, 107, 108; eyes, 425.

Australasia : taxonomy and evolution of Isograptus, 45.

B

Bacisphaeridium sp., 26.

Bactrognathus cf. distortus, 495, 58.

Ballisphaeridiunv, longispinosum, 105; pdiaris, 81 1, 105.

Batten, D. J. Use of palynologic assemblage-types in
Wealden correlation, I ; Palynology ofearlyCretaceous
soil beds and associated strata, 399.

Belodella ; devonica, 3 1 2, 32 ; resima, 3 1 2, 32 ; triangularis,
313, 32.

Bifida, 121 ; lepida, 123, 4-7.

Bird: Cretaceous humerus, 721.

Bivalvia: structural evolution of the shell, 519.

Boedaspis ensifer, 109.

Brachiopoda : Bifida and Kayseria and their affinity, 117;
palaeoecology of Appalachian gypidulids, 381.

Brachyodus africanus, 211 , 28.

Brachypotherium snowi. 111 , 28.

Brauckmann, C. See Hahn, G.

Breed, W. J. See Ford, T. D.

Brett, D. W. See Fowler, K.

Brookfield, M. E. The palaeoenvironment of the Abbots-
bury Ironstone (Upper Jurassic) of Dorset, 261.

Bryozoa: Lias of Northamptonshire, 219.

Bumastus harriensis, 108.

Buoyancy: in ammonoids, 623.

Butler, M. Lower Carboniferous conodont faunas from
the eastern Mendips, England, 477.

C

Calymene hlumenhachii, 108.

Cambrian: eye in Olenidae, 735; Lapworthellids from
Comley, Shropshire, 139.

Canada: Campanian dinoflagellate cysts and acritarchs,
665; Silurian acritarchs, 799.

Canningia senonica, 679, 85.

Carboniferous : amphibian from Scotland, 1 79 ; conodont
faunas from the eastern Mendips, 477; conodont
faunas from south-west Ireland, 335; Visean trilobites
from Ireland, 391; Visean trilobites from Somerset,
551.

Celyphus rallus, 35, 2.

Chaloner, W. G. and Gay, M. M. Scanning electron
microscopy of latex casts of fossil plant impressions,
645.

Chuaria circularis, 539, 61-63.

Cicatricosisporites 568, 66-77.

Clarkson, E. N. K. The eyes of Asaphus raniceps Dalman
(Trilobita), 425.

— Morphology and evolution of the eye in Upper
Cambrian Olenidae (Trilobita), 735.

Clavulina pacifica, 100.

Cleistosphaeridhmr, diversispinosum, 684, 86; sp. A, 684,

85.

Clonograptiis aureus, 707.

Collins, J. S. H. and Morris, S. F. A new crab from the
Middle Eocene of Libya, 283.

Conodonts: Carboniferous of the eastern Mendips, 477;
Carboniferous of south-west Ireland, 335; Devonian
of New South Wales, 307.

Converrucosisporites venitus, 406, 41, 42.

Cooper, R. A. Taxonomy and evolution of Isograptus
Moberg in Australasia, 45.

846

INDEX

Copper, P. Bifida and Kayseria (Brachiopoda) and their
affinity, 1 17.

ICoronifera oceanica, 684, 85.

Correlation: Wealden, England, 1 ; Dorset, 567.

Crab : Eocene of Libya, 283 ; hermit crabs symbiotic with
ectoprocts, 563, 65.

Crassigyrinus scoiicus, 179, 16.

Creodonta indet., 275, 28.

Cretaceous : agglutinating foraminifer, 637 ; bird humerus
from England, 721; Campanian dinoflagellate cysts
and acritarchs from Alberta, 665; isotopic ratios and
Wealden environments, 607; otoliths from England,
293; palynologic correlation of the Dorset ‘Wealden’,
567 ; palynology of soil beds and associated strata, 399 ;
Upper, staminate heads with pollen grains, 41 ;
Wealden palynologic assemblage-types, 1 .

Croxton, C. A. See Hughes, N. F.

Ctenopyge (Eoctenopyge) modesta, 751, 94.

Ctenopvge (Mesoctenopvge); similis, 752, 93; tumida,
752, 94.

Cummingella jonesi jonesi, 559, 64.

Cuticle; plant, 1; trilobite, 827.

Cyclonephelium distinctiim, 689, 87.

Cyinatiogalea sp., 24.

Cvinatiosphaera; octoplana, 106; wenhckia, 106; sp. A.
'696, 88.

Cyrtometopus sp., 108.

D

Dalingwater, J. E. Trilobite cuticle microstructure and
composition, 827.

Dalradian: Protospongia from Ireland, 231.

Deflandrea echinoidea, 694, 87; korojensis, 694, 87;

macrocysta, 695, 87; spectahilis, 693, 87.

Deines. P. See Allen, P.

Deunffia, 806; elongate, \Q4; furcata, 104; /! niagarensis,

807, 104; monospinosa, 104; m. tonawandensis, 807,
104; ramiiscidosa, 104; r. rochesterensis, 807, 104.

Devonian: conodonts from New South Wales, 307.
Diconodinium; arclictim, 672, ^4; firmimi, 669, 84.
Dictyotidium dictyolum, 106.

Diexallophasis', denticulata, 811, 105; sp., 27.
dinoflagellates: cysts from the Campanian of Alberta,
665; Tertiary of Virginia and Maryland, 729.
Dinogynmiunr, cf. albert ii, 676, 85; longicornis, 678, 85.
Diver. W. L, See Muir, M. D.

Doliognatlnis sp., 495, 58.

Dollymae', bouckaerti, 496, 58; sp., 496.

Domasur, amphora, 104; hispinosa, 104; canadensis,

808. 104; rochesterensis, 810, 104; trispinosa, 104;
sp., 27.

Dorset: palaeoenvironment of the Abbotsbury iron-
stone. 261 ; palynologic correlation of the ‘Wealden’,
567.

Downie, C. Observations on the nature of the Acritarchs,
239.

Duvernaysphaera aranaides, 106.

E

Echinodermata : Silurian echinoid from Scotland, 651.
Echinoid: Silurian of Scotland, 651.

Ectoprocts: symbiotic with Jurassic gastropods and
hermit crabs, 563, 65.

Edwards, N. See Fowler, K.

Egypt: Miocene mammalian fauna, 275.

Elektoriskos; pogonius, 105; simplex, 814, 105.

Elictognathus laceratus, 351, 37; 496, 59.

Elliott, G. F. A Miocene solenoporoid alga showing
reproductive structure, 223.

'Elops' neocomiensis, 296, 31.

Encrinurus punctatus, 108.

Endosporites, 766, 96.

England: conodonts from the eastern Mendips, 477;
Cretaceous bird humerus, 721; Eocene taxodiaceous
conifer, 205; otoliths from the Cretaceous, 293;
Wealden palynologic assemblage-types, 1 .

Eocene : new crab from Libya, 283 ; taxodiaceous conifer
remains from southern England, 205.

Eupoikilofnsa stratifera, 105.

Evittia', monterrosa, 815, 106; romota, 106.

Evolution: Isograptus in Australasia, 45; structure of
the bivalve shell, 519.

Exochosphaeridiiinv, cf. phragmites, 687, 86; pseudo-
hystrichodinium, 688, 86; sp. A, 688, 86.

Eyes; Asaphus raniceps, 425; Olenidae, 735.

F

Eilisphaeridium bifurcatum, 814, 105.

Fishes; Cretaceous otoliths from England, 293; holo-
stean, Jurassic of India, 149.

Florinites, 769, 97.

Foraminifer : agglutinated wall structure. 111 ; an unusual
agglutinating form from the Upper Cretaceous, 637.

Ford, T. D. and Breed, W. J. The problematical Pre-
cambrian fossil Chuaria, 535.

Fowler, K., Edwards, N. and Brett, D. W. In situ
coniferous (taxodiaceous) tree remains in the Upper
Eocene of Southern England, 205.

France: Jurassic ectoprocts symbiotic with gastropods
and hermit crabs, 563.

G

Gastropoda : symbiotic with Jurassic ectoprocts, 563, 65.

Gay, M. M. See Chaloner, W. G.

Geragnostus sp., 477, 51.

IGeron sp., 27.

Glyptostroboxylon, 19.

Gnathodus; bulhosus, 497, 56; delicatus, 354, 35; 497,
56; punctatus, 354, .35; 498, 56; semiglaber, 356, 35;
texanus pseudosemiglaber, 500, 56; t. texanus, 500,
56; sp. B. 500, 56.

Gomphotheriwn angustidens, 276, 28.

Gorgonisphaeridium wenlockium, 814. 105.

Graptolites: branching in Clonograptus, 707; Isograptus
in Australasia, 45.

Gypidukr, coeymanensis, 39; prognostica, 40.

H

Hahn, G. and Brauckmann, C. Lower Visean trilobites
from Feltrim, Ireland, 391.

Hahn, G. and Hahn, R. Visean trilobites from Holwell,
Somerset, 551.

INDEX

847

Hahn, R. See Hahn. G. and Hahn, R.

HaUetheca, 98.

Hamilton, W. R. A Lower Miocene mammalian fauna
from Siwa, Egypt, 275.

Hancock, C. D. See Palmer, T. J.

Harland, R. Dinoflagellate cysts and acritarchs from the
Bearpaw Formation (Upper Campanian) of southern
Alberta, Canada, 665.

Harpid, 472, 55.

Harrison, C. J. O. and Walker, C. A. Wyleykr. a new
bird humerus from the Lower Cretaceous of England,
721.

llelosphaeridium latispinosum, 105.

Hemideunffia irifurcata, 810, 104.

Hindeodellcr, equidentata, 313, 33, 34; segaformis, 500,
58; sp. A, 314, 34; sp. B, 315, 33;?sp., 316. 34.

Hodgkinson, R. L. See Adams, C. G.

Hughes, N. F. and Croxton, C. A. Palynologic correla-
tion of the Dorset ‘Wealden’, 567.

Hvsirichosphaeridiwn', cf. arborispinum, 680, 85;

dowlingii, 681, 85; salpwgophorum, 682, 85; tuhiferum
brevispinum, 682, 85.

I

Illaeniis udimcus, 108.

Illaenus {Parilloemis): incertiis, 468, 54; sp. A, 472, 54.

India: Jurassic holostean fishes, 149.

Inversidiuiiim, 730; exilimurum, 730, 90.

Ireland: Carboniferous conodont faunas, 335; Carboni-
ferous trilobites, 391; Dalradian Protospongia, 223.

Ischyosporites arkellii, 411, 43, 44, 46.

Isograptiis; cadiiceus, 71 ; c. australis, 74; c. imitatiis, 71 ;
diimosus, 78; forcipiformis, 77; hastatiis, 82; nuwu-
briatus, 84; ovatus, 89; primulus, 57; victoriae. 59;
V. divergens, 69 ; v. lunatus, 59 ; v. maximodivergens, 65 ;
V. maximus, 63 ; v. victoriae, 62.

Isotopic ratios: Wealden environments, 607.

J

Jackson, D. E. On the mode of branching in a new
species of Clouograptus, 707.

Jain. S. L. New specimens of Lower Jurassic holostean
fishes from India, 149.

Jurassic: aptychi of Bajocian Sotminia. 195; ectoprocts
symbiotic with gastropods and hermit crabs, 563 ;
holostean fishes from India, 149; Lias Bryozoa from
Northamptonshire, 219; palaeoenvironment of the
Abbotsbury ironstone, 261.

K

Kayseria. 130; alvea, 134, 7; dividua, 131, 4, 5, 7; lens,
131, 7 ; nohnensis, 1 36, 7.

Keith. M. L. See Allen, P.

Kempf, E. K. Transmission electron microscopy of
fossil spores, 787.

Kier, P. M . A new Silurian echinoid from Scotland. 65 1 .

Knight, R. H. See Adams, C. G.

Krassilov, V. A. Upper Cretaceous staminate heads with
pollen grains, 41 .

L

Lahyrinthidoma, 637; dumptonensis, 638, 78.

Lapworthella dentata, 142, 8, 9.

Lapworthellids: Lower Cambrian of Shropshire, 139.

Leiofusa; algerensis, fusiformis, 104.

ILeiosphaeridia sp., 25.

Lejeunkr, ampla, 673, 84; parva, 672, 84; tricuspis,
673, 84.

Lepidodendron suhdichotomum, 79.

Libya: new Eocene crab, 283.

Ligonodina; alf. salopia, 316, 32-34; sp., 317, 33.

Linguaphillipsia matthewsi, 552, 64.

Lonchodincr, greillingi, 319, 33; walliseri, 318, 33.

Lophosphaeridium ■, microgramdoswn, 818, 106; rugosum,
818, 106.

M

' Macromeriwn scoticum, 187, 16.

Makurath, J. H. See Anderson, E. J.

Mammals: Miocene fauna of Egypt, 275.

Maryland: Tertiary problematical dinoflagellate, 729.

Matthews, S. C. Lapworthellids from the Lower
Cambrian Stremtella Limestone at Comley, Shrop-
shire, 139.

- Notes on open nomenclature and on synonymy lists,
713.

and Naylor, D. Lower Carboniferous conodont
faunas from south-west Ireland, 335.

McLean, D. M. A problematical dinoflagellate from the
Tertiary of Virginia and Maryland, 729.

Meaudrograptus tail, 89.

Megalops bicremdatus, 295, 31.

Megaspores, 1.

1 Membranosphaera cf. maastrichta, 679, 85.

Meseiiteripora wrighti, 20.

Mestognathus beckmanni, 501, 58.

Micrhvstridium: stellatum, 105; sp. A, 696, 88; sp. B,
696,' 88; sp. C, 696, 88; sp. D, 696, 88.

Microdinium cf. irregidare, 675, 85.

Microfossils: acritarchs, 239; Silurian acritarchs of
Ontario, 799 ; agglutinated wall structure of Foramini-
ferida, 777; Cambrian Lapworthellids, 139; Carboni-
ferous conodonts of Ireland, 335; Carboniferous
conodonts from the eastern Mendips, 477 ; Cretaceous
agglutinating foraminifer, 637; Cretaceous dino-
flagellates and acritarchs from Alberta, 665 ; Cretaceous
otoliths, 293; Devonian conodonts of New South
Wales, 307; Tertiary dinoflagellates from eastern
U.S.A., 729; Palaeozoic spores, 765.

Miocene: mammalian fauna from Egypt, 275; soleno-
poroid alga showing reproductive structures, 223.

Miospores, 1.

Mollusca. See Ammonoidea, Bivalvia, Coleoidea,
Gastropoda.

Morris, S. F. See Collins, J. S. H.

Morton, N. The aptychi of Somiinia ( Ammonitina) from
the Bajocian of Scotland, 195.

Muir. M. D. and Diver. W. L. The application of electron
microscopy to palaeontology, 733,

Midtiplicisplu/eridiimr, arbiisculiferiim, 105; eoplank-
tonicum, 105; fisherii, 105; sp., 27.

848

INDEX

Murray, J. W. Wall structure of some agglutinated
Foraminiferida, 111 .

Mutvei, H. and Reyment, R. A. Buoyancy control and
siphuncle function in ammonoids, 623.

N

Naylor, D. See Matthews, S. C.

Neoprionodus; excavatus, 319, 33; sp., 320, 33.

Neosolenopora armoricana, 224, 21, 22.

New South Wales: Devonian conodonts, 307; Ordo-
vician trilobites, 445.

Nomenclature, 713.

North America: Appalachian gypidulid brachiopod
palaeoecology, 381.

Northamptonshire: Lias Bryozoa, 219.

O

Odonotochitina operculata, 695, 88.

Olenidae: eye morphology and evolution, 735.

Olenus wahlenbergi\ eyes, 739, 91, 92.

Oligosphaeridiunr, anthophomm, 686, 86; pidcherrimum,
686, 86.

Ontario: Silurian acritarchs, 799.

Ooidium sp., 24.

Ordovician: trilobites from New South Wales. 445.

Orthosphaeridium sp., 26.

'Osmerus' lobatiis, 301, 31.

Ozarkodina ', media, 322, 33; typica denckmanni, 322, 33.

P

Paladin eichwaldi shimnerensis, 108.

Palaeocarpilius aquilinus, 284, 29, .30.

Palaeoecology : Abbotsbury ironstone, 26 1 ; Appalachian
gypidulid brachiopods, 381; Bearpaw Formation,
Canada, 699; brachiopods Bifidia and Kayseria, 118;
Cretaceous soil beds, 402; isotopic ratios and Wealden
environments, 607; taxodiaceous conifers. Eocene,
208.

Palaeozoic: electron microscopy of in situ spores, 765.

Palmer, T. J. and Hancock, C. D. Symbiotic relation-
ships between ectoprocts and gastropods, and ecto-
procts and hermit crabs in the French Jurassic, 563.

Paltodus sp., 323, 32.

Palynology: early Cretaceous, 399; Wealden, 1 ; 567.

Panchen, A. L. On Crassigyrinus seoticus Watson, a
primitive amphibian from the Lower Carboniferous
of Scotland, 1 79.

Panderodus', gracilis, 324, 32; simplex, 323, 32; uni-
costalus, 324, 32.

Parabolina spinulosa, eyes, 742, 92.

Paradapedium egertoni, 157, 10-14.

Patrognathus variabilis, 356, 35; 501, 59.

Pecten maximus. 60.

Pelekysgnallms; bullyncki, 501, 58; sp. A, 502, 58.

Pellura', minor, 748, 93; scarahaeoides scarabaeoides,
eyes, 746.

Phacops granulatus, 109.

Phillips, W. E. A. See Rushton, A. W. A.

Phillipsia {Phillipsia) holwellensis, 557, 64.

Pilasporites allenii, 416, 42.

Plants: Cretaceous soil beds, 391; electron microscopy
of Palaeozoic spores, 765; Eocene taxodiaceous
conifer, 205 ; Miocene solenoporoid alga, 223 ; palyno-
logic correlation of the Dorset ‘Wealden’, 567; Pre-
cambrian Chuaria, 535; scanning electron microscopy
of latex casts of plant impressions, 645; Upper
Cretaceous staminate heads with pollen grains, 41 ;
Wealden palynologic assemblage-types, I ; see also
Algae and Wood.

Plectospathodus', aff. alternatus, 325, 34; extensus, 324,
33; sp., 326, 34.

Poikilofusa sp., 24.

Pollen: in Upper Cretaceous staminate heads, 41.

Polygnathus', communis carina, 358, 37; 502, 59; c.
communis, 357; 503, 59; flabellus, 358, 37; inornatus,
360,36; 503, 59 ; longiposticus, 363, 36;nodomarginatus,
504, 57 ; symmetricus, 364, 36, 37 ; 506, 57, 59.

Polysphaeridium subtile, 687, 86.

Powell, H. P. See Walter, B.

Precambrian: Chuaria, 535.

Problematica: Precambrian Chuaria, 535.

Protospongia hicksi, 232, 23.

Pseudobasilicus', Ifortis, 461, 53, 54; sp. A, 466, 55.

Pseudopolygnathus', dentilineatus, 365, 38; 506, 57;
multistriatus, 365, 38; 507, 57; postinodosus, 507, 57;
primus, 365, .38; 507, 57; triangulus pinnatus, 508, 58;
t. triangulus, 510, 58.

Pteralhula cantiana, 298, 31.

Pterospermopsis', cf. martinii, 106; onondagaensis, 106;
sp., 26.

Pterothrissus galtinus, 300, 31.

Q

Quadraditum fantisticum, 105.

R

Regresporites, 408; lophus, 410, 41, 42, 44.

Remop leurides', acer, 457, 53; exallos, 456, 51; saenuros,
451, 52; sp. A. 457, 52; sp. B, 460, 51; sp. C, 460, 51.

Retitriletes sp., 41 1, 41, 43.

Reyment, R. A. See Mutvei, H.

Rotundacodina dubia, 328, 32.

Rushton, A. W. A. and Phillips, W. E. A. A Protospongia
from the Dalradian of Clare Island, Co. Mayo,
Ireland, 231.

S

Saccammina atlautica, 99.

Salvinia cerebrata, 101.

Savage, N. M. Lower Devonian conodonts from New
South Wales, 307.

Scaliognathus anchoralis, 5 1 0, 58.

Schizodelphis aff. sulcatus, 276, 28.

Schopfipollinites, 112, 98.

Schopfitheca, 98.

Scotland; aptychi of Bajocian Sonninia, 195; Carboni-
ferous amphibian, 179; Silurian echinoid, 651.

Setosisporites', brevispinosus, 102; hirsutus, 102.

Shropshire: Lapworthellids, 139.

Shumardia sp., 450, 51.

INDEX

849

Silurian: acritarchs from southern Ontario, 799;
echinoid from Scotland, 65 1 ; gypidulid palaeoecology,
381.

Siphonodella\ cooperi, 365, 36, 37; 510, 59; crenulata,
511, 59; isosticha, 366, 37; 511, 59; obsoleta, 366;
511. 59.

Siphuncle: buoyancy control in ammonoids, 623.

Somerset: Carboniferous trilobites, 551.

Sonninia (Papilliceras)\ arenata, 17, 18; mesacantha, 17.

Spathognodus; aculeatiis, 368, 35; 512; cf. costatus, 371 ;
crassidentatus, 369; 512, 59; inclinatus wunni, 328,
32, 34; remscheidensis, 329, 34; scitulus, 514, 59;
stabdis, 370, 35; 514, 59; sp„ 370, 35.

Sphaeronchus rotimdus, 302, 31.

Sphaerophthalmus \ alatus, 754, 95; hunnlis, 754, 94, 95.

Spinidium clavum, 674, 84.

Spiniferites', cf. membranaceus, 692, 87; cf. porosus, 690,
87; ramosus gracilis, 692, 87; r. granosus, 690, 87;
r. multibrevis, 690, 87 ; r. ramosus, 689, 87.

Spores: electron microscopy in situ, 765; transmission
electron microscopy, 787.

Stinton, F. C. Fish otoliths from the English Cretaceous,
293.

Structure; acritarch wall, 241; evolution of the bivalve
shell, 519; wall of agglutinated Foraminiferida, 777.

Synonymy, 713.

Synsphaeridium sp., 25.

T

Tan, F. C. See Allen, P.

Tanyosphaeridium variecalamum, 687, 86.

Tasmanites sp., 26.

Taylor, J. D. The structural evolution of the bivalve shell,
519.

Taylor, T. N. Combined transmission and scanning
electron microscopy of in situ Palaeozoic spores, 765.

Techniques: application of electron microscopy to
palaeontology, 733; scanning electron microscopy of
latex casts of fossil plant impressions, 645.

Tertiary: problematical dinoflagellate, 729.

Tetragonolepis oldhami, 169, 15.

Te.xtularia sagittida, 99.

Thomsonia, 1.

Thracia phaseolina, 60.

Thusu, B. Acritarchs of the Middle Silurian Rochester
Formation of southern Ontario, 799.

Triangulinal sp., 27.

Trichonodelkr, excavata, 330, 33; inconstans, 33.

Tricolpopollianthus burejensis, 41,3.

Trilobites: Carboniferous from Ireland, 391; Carboni-
ferous from Somerset, 551; cuticle microstructure,
827; eyes of Asaphus raniceps, 425; morphology and
evolution of the eye in Upper Cambrian Olenidae,
735; Ordovician from New South Wales, 445.

Trilobosporites ivanovae, 414, 43, 45, 46.

Trimerocephalus caecus, 109.

Tunisphaeridium tentaculiferum, 105.

U

Uvatodiniuml cf. nasutum, 680, 85.

V

Vertebrates: Carboniferous amphibian from Scotland,
179; Cretaceous bird humerus, 721; Cretaceous
otoliths from England, 293; Jurassic holostean fishes
from India, 149; Miocene mammalian fauna from
Egypt, 275.

Veryhachiunr, lairdi, 815, 106; limaciforme, 816, 106;
trispinosum, 106; sp., 24.

Virginia: Tertiary problematical dinoflagellate, 729,

Vishysphaera', dilatispinosa, 106; sp., 27.

W

Walker, C. A. See Harrison, C, J. O.

Walter, B. and Powell, H. P. Exceptional preservation in
cyclostome Bryozoa from the Middle Lias of North-
amptonshire, 219.

Webby, B. D. Remopleurides and other Upper Ordo-
vician trilobites from New South Wales, 445.

Wood, 1.

Wyleyia valdensis, 721, 89.

c

V

V-'

%

•j . .

*.4 •

?

c

i

5

if

'■ ,

THE PALAEONTOLOGICAL ASSOCIATION

The Association was founded in 1957 to further the study of palaeontology. It holds meetings
and demonstrations as well as publishing Palaeontology and Special Papers in Palaeontology.
Membership is open to individuals, institutions, libraries, etc. on payment of the appropriate
annual subscription :

Institutional membership .... £10 00 (U.S. $26.00)

Ordinary membership .... £5-00 (U.S. $13.00)

Student membership .... £3-00 (U.S. $8.00)

There is no admission fee. Institute membership is only available by direct application, not
through agents. Student members are persons receiving full-time instruction at educational
institutions recognized by the Council; on first applying for membership, they should obtain
an application form from the Membership Treasurer. All subscriptions are due each January,
and should be sent to the Membership Treasurer, Dr. E. P. F. Rose, Department of Geology,
Bedford College, Regent’s Park, London, NWl 4NS, England.

PALAEONTOLOGY

All members who join for 1973 will receive Volume 16, Parts 1-4.

All back numbers are still in print and may be ordered from B. H. Blackwell, Broad Street,
Oxford, 0X1 3BQ, England, at £5 per part (post free). A complete set. Volumes 1-15, consists
of 59 parts and costs £295.

SPECIAL PAPERS IN PALAEONTOLOGY

The subscription rate is £8 (U.S. $22.00) for Institutional Members and £4 (U.S. $11.00) for
Ordinary and Student Members. Subscriptions should be placed through the Membership
Treasurer, Dr. E. P. F. Rose, Department of Geology, Bedford College, Regent's Park, London,
NW) 4NS, England. Members only may obtain individual Special Papers from Dr. Rose at
reduced rates. Non-members may obtain them at the stated prices from B. H. Blackwell, Broad
Street, Oxford, 0X1 3BQ, England.

COUNCIL 1973-1974

President: Professor M, R. House, The University, Kingston upon Hull, Yorkshire, HU6 7RX
Vice-Presidents: Mr. N. F. Hughes, Department of Geology, Sedgwick Museum, Cambridge, CB2 3EQ
Dr. Isles Strachan, Department of Geology, The University, Birmingham, B15 2TT
Treasurer : Dr. J. M. Hancock, Department of Geology, King’s College, London, WC2R 2LS
Membership Treasurer : Dr. E. P. F. Rose, Department of Geology, Bedford College, Regent’s Park,

London, NWl 4NS

Secretary : Dr. W. D. I. Rolfe, Hunterian Museum, The University, Glasgow, G12 8QQ
Assistant Secretary : Dr. C. T. Scrutton, Department of Geology, The University, Newcastle upon Tyne,

NEl 7RU

Editors

Dr. R. Goldring, Department of Geology, The University, Reading, RG6 2AB
Dr. J. D. Hudson, Department of Geology, The University, Leicester, LEI 7RH
Dr. D. J. Gobbett, Department of Geology, Sedgwick Museum, Cambridge, CB2 3EQ
Dr. L. R. M. Cocks, Department of Palaeontology, British Museum (Natural History), London, SW7 5BD

Other members of Council

Dr. M. G. Bassett, Cardiff
Dr. D. D. Bayliss, Llandudno
Dr. E. N. K. Clarkson, Edinburgh
Prof. D. L. Dineley, Bristol
Dr. Julia A. E. B. Hubbard, London
Dr. C. P. Hughes, Cambridge*

Dr. J. K. Ingham, Glasgow

Mr. M. Mitchell, Leeds
Dr. Marjorie D. Muir, London
Dr. J. W. Murray, Bristol*

Dr. B. Owens, Leeds

Dr. P. F. Rawson, London

Prof. D. Skevington, Galway

* Co-opted as Editors

Overseas Represenlatives

Australia: Professor Dorothy Hill, Department of Geology, University of Queensland, Brisbane
Canada: Dr. B. S. Norford, Institute of Sedimentary and Petroleum Geology, 3303-33rd Street NW., Calgary,
Alberta

India: Professor M. R. Sahni, 98 The Mall, Lucknow (U P.), India

New Zealand: Dr. C. A. Fleming, New Zealand Geological Survey, P.O. Box 30368, Lower Hutt
West Indies and Central America: Mr. John B. Saunders, Geological Laboratory, Texaco Trinidad, Inc., Pointe-a-
Pierre. Trinidad, West Indies

Western U.S. A.: Professor J. Wyatt Durham, Department of Paleontology, University of California, Berkeley 4,
California

Eastern U.S. A. : Professor J. W. Wells, Department of Geology, Cornell University, Ithaca, New York

Palaeontology

VOLUME 16 - PART 4

CONTENTS

A new Silurian echinoid genus from Scotland

P. M. KIER 651

Dinoflagellate cysts and acritarchs from the
Bearpaw Formation (Upper Campanian) of
Southern Alberta, Canada

R. HARLAND 665

On the mode of branching in a new species of
Clonograptus

D. E. JACKSON 707

Notes on open nomenclature and on synonymy
lists

S. C. MATTHEWS 713

Wyleyia : a new bird humerus from the Lower
Cretaceous of England

C. J. O. HARRISON anjc. A. WALKER 721

A problematical dinoflagellate from the Tertiary
of Virginia and Maryland

D. M. MCLEAN 729

The application of Electron Microscopy to
Palaeontology (EMCON proceedings)

M. D. MViK and w. L. diver 733

Morphology and evolution of the eye in
Upper Cambrian Olenidae (Trilobita)

E. N. K. CLARKSON 735

Combined transmission and scanning electron

microscopy of in situ Palaeozoic spores

T. N. TAYLOR 765

Wall structure of some agglutinated
Foraminiferida

J. W. MURRAY 111

Transmission Electron Microscopy of fossil
spores

E. K. KEMPF 787

Acritarchs of the Middle Silurian Rochester
Formation of Southern Ontario

B. THUSU 799

Trilobite cuticle microstructure and composition

J. E. DALINGWATER 827

Palaeontological Association Report and

Accounts for 1972 841

Index to Volume 16 845

Printed in Great Britain at the University Press, Oxford
by Vivian Ridler, Printer to the University

pp, ^ XtvAsisi?/ m nWSs^

= . cn ■ ± (/> — ±z </> -

VHgn libraries SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHilWS S3ldVyan LIBR/I
2 > w z . w z «/) z •-

- - xT^g>ox ±

o ^y' W^' ^ o

^ 5 'Y?-''' >' 5 Xivosvi!^ >■

</) •*“ Z CO Z

'UTION NOIinillSNI_NVINOSHillNS S3iyvyan libraries SMITHSONIAN INSTITUTION NOiin.

C/> -r V) ~ CO — CO

CO

'■

“ -Z ^

vyan libraries Smithsonian institution NoixnxixsNi nvinoshihms S3iHvyan libr/

z »" z

P ^-03

zo
>

:o ^

CO — v> — CO ^. _ to

TUTION NOlinillSNI NVINOSHIIINS S3iyvyail LIBRARIES SMITHSONIAN INSTITUTION NOIIR
. , to z ^ ^ to z , , to z V- w

O

o

CO
X

^ tr " ^ t-

> 5 5 ^ ^

vyan libraries Smithsonian institution NoiiniiiSNi nvinoshiiws‘^S3 i yvy a n libr/

CO X to ^ X ^

tu ^ xw^Ip^cx tu ✓<u?v:;r\ „ .^s>\ uj

z

>

cc

<
q:

O _ Xi^iisJX o XS' XCfCmsjX o

TUTION^ NOlinillSNI “*NVIN0SH1IWS^S3 iyvyan~’uBRARI ES^SMITHSONIAN~'lNSTlTUTION^NOIin

z ^ V, z r- z z

™ P XC5?IP(oX Z X-Dol^oX O >5^,. m xTp£5oX O XtSl

’ > (pl

35

m x^osv^ ^ m Xiyostiix^

to ”■ !X CO _ to

vyan libraries Smithsonian institution NoiiniiiSNi nvinoshiiws S3iyvyan libr;

5 X w ^ Z r to z ^ ^

5

to z to ' Z CO z

ruTiON NOlinillSNI NviNOSHims S3iyvyan libraries Smithsonian institution Noiin.

to ^ \ to ^ ^ — tn :z to

UJ xtSvrrriTX /« uJ UJ xfoovoX ^

kV*

O X>»osV^

Z _j z -»

vyan libraries Smithsonian institution NoiiniiiSNi nvinoshiuns saiavyan libr/

z r- 2 -

M ^ ^ ^ Jij ^

m ^ rr, I ;;; ^ m

CO — CO E — to X “ to

TUTION NOlinillSNI NVINOSHilWS S3iyvyan libraries SMITHSONIAN institution NOlin

- ^ -i ...A ^ '

^ f;

m xi/vosviix \>va5^ m

_ —in — in

illTHSONIAN mSTITUTION'^NOliniliSNI NVINOSHilWS S3 I y VH 9 11 ^LI B R AR I ES SMITHSONIAN INS

{/> z: in -z

^ s m s * s 5

t

>• '' 2 X^ds»^ >

* • 3! CO ^ CO ‘■•^ ^ ^

^iNOSHiiws S3iyvysn libraries Smithsonian institution Noiiru!iSNi_NviNOSHiiiAis s3

in ^ ^ « z ^ ^ z: '■^ ^

” ^^5) ^ ^ ~ /^W\ ^

5 ® ^ 5

MITHSONIAn'^INSTITUTION NOIinilXSNl"^NVINOSHilWS S3iyVH9ll libraries SMITHSONIAN IN:

^ j— ^ T- z

c» xSkwTx ™ v; P m 00 /

o

z:

33

>

33

m

U3

m ^ rn ^ m

C/3 _ <S) ‘- _ 0^

viNOSHims S3iyvyan libraries smithsonian institution NoiiniiiSNi nvinoshiiiais S3
w z: w z: .sv- 5 ^

. . . "# iW ifOliW

2T in * Z! — </) z in

mithsonian institution NoiiniiiSNi NviNOSHilws S3iyvyan libraries smithsonian_in

to zz. ... in <n 2

iJj f/3

Q pc ^ Q " Q _ "1^'- O

VINOSHimS^Sa I y vy a I1~^LI B RAR I ES^SMITHSONIAN^INSTITUTION^NOlinmSNI^NVINOSHillNS^S;

1 I i I I '<^1^ i /

S ^ ^ ^ CO —

MITHSONIAN INSTITUTION NOlIfUllSNI NVINOSHilWS S3iyvy9ll U B RAR I ES SMITHSONIAN IN

in Zi to z ^ 5

' ' 2 >■ * ' 2 2 ^
jviN0SHims“'s3 1 y vy 9 n'^L! b rar i es^smithsonian institution NoiiniiiSNi_NvmosHiiiNS s:

in — in z \

o

z

O Z O

“Z. «j 2

5MITHS0NIAN INSTITUTION NOlinillSNI NVINOSHilWS SBiavyail LIBRARIES SMITHSONIAN IN

r- Z r- 2: S

O v'^vaSO^TTv ^

m

— CO ^ ± c^

j I yvy an_Li brar I es ^smithsonian institution ^NoiiniiisNi^NViNOSHims^s,

3 . .'I i I vfc i -- -

^ Z