A unique specimen of an unusual elongate segmented compression fossil, Polylurida aenigmatica n. gen., n. sp., is described and illustrated. The bilaterally-symmetrical, dorsoventral structure (c. 6·5 cm long × 0·8 cm maximum width) narrows towards both ends and is subdivided into approximately 19 similar tapering segments, showing no evidence of tagmosis, appendages or an internal gut. The segments are covered with a substantial papillose organic cuticle, each segment bearing towards one end a pair of transverse calcareous bars. These bars are perforated by simple canals filled with strands of dark organic material which traverse their whole thickness and flare out superficially to form pores on both their lower and upper surfaces. Internal horizontal lamination of the bar material suggests an episodic accretionary pattern of skeletal growth. The bars appear to represent internal skeletal structures and are restricted to one side (dorsal or ventral) of the body, where the adjacent cuticle is locally thickened to form a corresponding series of discrete paired densely papillose transverse bands. Although Polyurida may be an unusual segmented worm, its singular suite of characters and lack of data prevent assignment to any higher taxon.

Tealliocaris is probably one of the best preserved of all Carboniferous crustaceans. It occurs scattered through the 12 cm thick laminated Gullane ‘shrimp-bed’. The cuticle is exceptionally well preserved and the specimens, which are normally complete, are infilled with fluorapatite. The carapace has an elongate rostrum, a defined frontal area, a median ridge and paired gastric and inner and outer lateral ridges. The eyes are borne on a stout peduncle. The first antenna bears a pair of flagella, the second an antennal scale and long flagellum. The mandible is large and heavily sclerotised with molar and incisor process. The maxillae and anterior thoracopods are poorly preserved. The posterior six thoracopods each have a long segmented endopod and annulated setose exopod. More than half the specimens preserve lamellae which may represent branchial epipods, or floor a brood chamber beneath the carapace on either side of the thorax. The pleopods are biramous. The telson and uropods form a tail fan. In the absence of oostegites Tealliocaris cannot be assigned to the order Mysidacea but its affinities lie close to these Eumalacostraca. Tealliocaris walked on the thoracic endopods, but could also swim using the pleopods and thoracic exopods. The mode of feeding is uncertain. Tealliocaris probably inhabited a brackish lake or lagoon, but its distribution elsewhere suggests that it may have been tolerant of a range of salinities.

Coprolites of four distinct shapes, cylindrical, discoidal, beaded and irregular masses, are described from 11 horizons within Bed 16 of the Foulden sequence. These coprolites are composed of either phosphatic or claystone material, although they have not been analysed mineralogically or chemically. Their producers appear to have been fish, probably both nektonic predators and nekto-benthonic scavengers or detritus feeders.

Ostracods are poorly preserved and of two broad types, thick shelled paraparchitids including Shemonaella scotoburdigalensis and thin shelled podocopids, possibly species of Carbonita or Cavellina. This ostracod assemblage represents a conservative ‘carbonaceous fades’ fauna which occurs widely in plant-rich sediments in lower Carboniferous (Courceyan) sequences of northern England and southern Scotland. Other microfauna includes spirorbid worm tubes, fish scales, arthropod cuticle fragments, spat of myalinid bivalves, plant fragments and spores. No significant changes in the vertical distribution of the microfauna could be detected.

The microfauna indicates a general fresh to brackish water lacustrine or lagoonal environment possibly with access to quasi-marine conditions.

Acanthodians are represented by numerous young and poorly cranially ossified Acanthodes ovensi White, the earliest known acanthodid. Study of the small A. ovensi together with the type series shows that squamation begins posteriorly, and spreads anteriorly along an initially triangular front, as in A. bridgei Zidek. Gyracanthus sp. occurs at several horizons in the sequence. The coelacanth remains include the first articulated specimen known from Foulden, but preservation only allows an assignment to cf. Rhabdoderma. Although these taxa are distinctive, resemblances to other taxa occur only in generalised features. Hence, unlike the Foulden actinopterygian fish, the acanthodians and coelacanths cannot be used to suggest closer affinities with any other fish of this age, on which information is sparse.

Macrurous decapod crustaceans are generally uncommon and are often incompletely preserved. However, at certain horizons in the Speeton Clay (Ryazanian–Albian) and Atherfield Clay (Lower Aptian) of Yorkshire and the Isle of Wight respectively, macrurans form the dominant element of the fauna. Two species in particular, Meyeria ornata (Phillips) and Meyerella magna (M'Coy), are not only very common in these formations, but are also very well preserved. Recent collections have yielded a large number of specimens of both species. This material is examined with the aid of SEM techniques, gross morphology is discussed, and the details of the exoskeletons are compared. In the case of M. magna, examples are figured which show traces of a colour pattern and the effects of regeneration on the first pereiopods. A discussion is included of the possible function of the appendages in M. magna and M. ornata, and of their other, unique features. Finally, the mode of life of both species is inferred from a consideration of their morphology, and of the evidence from their preservation, distribution and relationships with associated faunas.

The four stem-group neopterygians from Foulden are redescribed: Carboveles ovensi White, Aetheretmon valentiacum White, Strepheoschema fouldenensis White and Fouldenia ottadinica White. Carboveles is shown to be a junior synonym of Phanerosteon Traquair, Fouldenia ottadinica White is synonymised with Styracopterus fulcratus (Traquair), Strepheoschema rhodesi Moy-Thomas with S. fouldenensis White and Aetheretmon whitei Moy-Thomas with A. valentiacum White. The family Strepheoschemidae is erected to comprise Strepheoschema and Aetheretmon, characterised by paired ventral ridge scales in front of the pelvic fins. Cosmoptychius striatus (Agassiz), recorded here from Foulden for the first time, is widespread in the geologically younger Oil Shale Group. The Foulden actinopterygian fauna is identical to that of the Cementstone Group at Coomsdon Burn, Northumberland, although with two additional taxa. Styracopterus fulcratus occurs also at a further two Cementstone Group localities in Scotland.

This paper reviews the methods which have been used to deduce the life habits of trilobites. The most reliable conclusions are those that satisfy three independent criteria of evidence: (1) functional morphology of the exoskeleton; (2) analogy with living arthropods; and (3) geological evidence, as from facies relationships, or palaeogeography. Pelagic trilobites are one of the better examples for life habit reconstruction. The Ordovician trilobites that were most probably pelagic are those with hypertrophied eyes—Telephinidae, Opipeuteridae, Bohemillidae and Cyclopygidae—which satisfy the threefold requirements for evidence. Opipeuter, Carolinites (and other telephinids) were probably epipelagic, whereas cyclopygids and Bohemilla were likely to have been mesopelagic, living in exterior sites around the perimeter of Gondwana in the earlier Ordovician. These pelagics divide into two morphological groups: poorly streamlined, sluggish species somewhat like the living hyperiid Parathemisto, and well-streamlined species which are larger, and are considered to have been fast and active swimmers, with predatory habits. The streamlining of this group has been demonstrated by experiment. Other trilobites may also have been pelagic, but the evidence is less complete: Irvingella and its homeomorphs, and Remopleurides. Most Olenidae, however, which have been supposed to have been pelagic, were benthic forms inhabiting a specialised environment low in oxygen and possibly below the thermocline; only the leptoplastines may have been pelagic.

It is also shown that there are cases where ventral terrace ridges did not function to grip sediment during filter feeding. This applies not only to pelagic trilobites, but also to benthic ones in which the width and disposition of the doublure renders engagement with the sediment unlikely.

Material of crossopterygian fishes from Foulden, Berwickshire, is important in including the first known entire specimen giving a clear picture of the body form of the Family Rhizodontidae (sensu Andrews 1973). This group is specialised in having numerous subsidiary lateral lines on the body, a ventrally expanded shoulder girdle, fins with peculiar stiffened structure, the pectorals forming greatly enlarged paddles and other fins being reduced in size. The Foulden remains fall into large and small size-ranges occurring largely at different levels in the sequence, but although these may represent growth in one taxon the available parts of the body are mostly complementary. Separate preliminary descriptions of the two size-ranges are therefore given, the small complete form being named ?Strepsodus anculonamensis sp. nov., and the large form remaining unnamed pending further study. An abridged historical account of previously named rhizodont genera (Rhizodus, Strepsodus, Sauripterus and Pycnoctenion) is given in order to explain why the new Foulden form(s) cannot at present be generically assigned more closely. A revised diagnosis of the Family Rhizodontidae leads to a discussion of the functional morphology and mode of life of fish of this type. Analogies with modern forms indicate that, like some sharks and crocodiles, they may have fed by tearing flesh off large prey, rotating or shaking against its inertia. Ways in which variously shaped rhizodont teeth (recurved, sigmoid etc.) may have functioned are suggested. Addenda deal with (1) the possible occurrence at Foulden of a large lungfish and (2) (after considering primitive and advanced rhizodont features) the renaming of Rhizodus ornatus Traquair 1878 from higher in the Scottish lower Carboniferous, as the type of a new genus, Screbinodus.

A model suggesting that the trilobite carapace served primarily as a tight hood, under which sandy sediment could be processed for food, is consistent with morphological and ichnological data. It helps us to understand not only the origin of trilobite design, but also its modification for other habitats and modes of life.

The Foulden Fish Bed was deposited in a semi-permanent brackish water lake on a floodplain between a northerly mountain range and the sea. The lake supported a rich biota mainly of swimming animals (fish and crustaceans) with a benthos of trace fossils only in the earlier and later stages of the lake's development. Xiphosurans are very rare. This biota changed according to the physical conditions of the lake. Salinity, fluctuating according to water level, was a dominant control. Early in the lake's history there was a dominant benthos of euryhaline bivalves; at the same time, large rhizodonts lived in the lake as the top carnivores. Subsequently, crustaceans were the dominant nektobenthos, in a probable brackish-water regime, the rare Bairdops preying on the common Belotelson; there was a limited infauna. Following this, palaeoniscids abound in the middle part of the sequence, perhaps due to a higher water level but crustaceans are absent at this horizon. During the last stages before final silting-up following a mass-mortality episode, biotic diversity reached a maximum with many fishes and crustaceans and some trace fossils, due to optimal salinity conditions.

Comparisons are made with other Carboniferous biotas and with the Jurassic Solnhofen limestone.

Phacopid trilobites, unlike most trilobites, lack functionally articulated libregenae. This distinction presumably dictated ecdysial strategies which were fundamentally different from those forms with sutured cephala. Phacops rana (Green) and Greenops boothi (Green) from the Middle Devonian Hamilton Group (New York) are preserved in circumstances which illustrate certain aspects of their moulting behaviour. Undisturbed exuviae display five different moult patterns; these suggest two basic modes of exuviation in phacopid trilobites (Body-upright and Body Inversion Moult Procedures).

Certain Hamilton beds yield clusters of Phacops comprising complete and/or moulted remains. Complete trilobites in body clusters display varied orientations which, when considered in conjunction with exuvial patterns in moult clusters, suggest ecdysial procedures that involved body inversion and erratic tergal contractions (Body Inversion Moult Procedures). Ensembles where in thoracopygidia are upright and outstretched indicate hyperextension of the thorax and disarticulation of the cephalo-thoracic suture (Body-upright Moult Procedures). Phacops moult ensembles are characterised by inverted and tightly recurved thoracopygidia, whereas Greenops ensembles comprise outstretched, usually upright thoracopygidia with cephala that are frequently upright. Classic ‘Salterian’ patterns are uncommon in Phacops rana and have not been recognised in Greenops boothi. Variations in moult procedure are attributed to differences in functional morphology and substrate consistency.

There is abundant evidence of injury, teratology and pathological conditions in trilobites although it is commonly difficult to distinguish between their morphological results. Many injuries were probably sustained during moulting, with spines, bilamellar fringes and narrow gaps between dorsal exoskeleton and doublure being most vulnerable. Some injuries were the result of predaceous attack and provide important clues to the trophic structure of Palaeozoic communities. Many injuries show evidence of repair. Some supposedly teratological conditions, most notably the number of paradoxidid thoracic segments, are so common that they must be considered part of normal variation. Others represent genetic mutation or ontogenetic malfunction, the latter possibly externally induced. Undoubted pathological conditions are restricted to swellings and borings presumably caused by parasitic infestation.

Abnormalities affecting lateral areas are more common than those involving the rachis and glabella and this probably reflects both greater mortality when the axial region is involved, and greater vulnerability of peripheral regions to predaceous and moulting damage. The first explanation also accounts for the rarity of cephalic abnormalities compared to those of the pygidium.

The commonest arthropod trace fossils from Triassic aquatic red beds are the bilobate traces Isopodichnus and trackways known as ‘Merostomichnites’ triassicus. These trace fossils were probably produced by notostracan branchiopods, similar to Triops. Four arthropod ichnocoenoses from Lower Triassic fluvial sandstones have been analysed in terms of size frequency distribution, behavioural variation and relationship to sedimentary structures and depositional environment. One Isopodichnus ichnofauna associated with flute moulds (Dumfries-shire, Scotland) shows a normal age-structured population of arthropods responding with strong rheotaxis within shallow fluvial channels. The second Isopodichnus assemblage associated with ripple marks (Worcestershire, England) also shows strong rheotaxis but is bimodal in size and morphotype, possibly suggesting change in arthropod behaviour with age. Two ichnocoenoses of trackways with less pronounced rheotaxis associated with ripples (Cheshire, England) and flute moulds (Württemberg, Germany) were produced by larger arthropods than the resting traces. These arthropods probably possessed 6 to 9 pairs of walking limbs.

The conclusions derived from these notostracan trace fossils are compared with data on palaeoecology, population size-frequency, morphology and behaviour of Triops cancriformis derived from the analysis of three Triassic body fossil faunas and literature on living populations. Taxonomic consideration favours retention of the name Isopodichnus but the trackways should be included in Acripes Matthew. Brief review of late Palaeozoic Isopodichnus assemblages which appear to predate known notostracan fossils is inconclusive as regards both identifying producers or infallible means of separation from Cruziana assemblages.

Sediments which occur frequently within Lower Devonian andesites of the Midland Valley of Scotland have so far yielded trace fossils in only two areas: Dunure, Ayrshire and Montrose, Angus.

Smith (1909) interpreted the trace fossils from Dunure as representing an “Upland Fauna” and he named 23 new ichnogenera. Re-examination of his localities and specimens suggests that the environment was a shallow ephemeral lake and when sedimentological and preservational factors, size, behavioural variation and synonymy are taken into account, only 8 ichnogenera are here considered valid. These are redefined as Siskemia, Stiaria, Danstairia, Keircalia, Stiallia, Isopodichnus (=Archatomia Smith), Mermia and Merostomichnites. The Montrose ichnofauna of 5 ichnogenera includes a large arthropod trackway Mitchellichnus ferrydenensis gen. and sp. nov.

The trackways from both localities appear to have been formed by various merostomes, crustaceans and possibly early myriapods. The habitat was essentially aquatic although some arthropods may have been amphibious.

The fossil record of unquestionable decapod remains dates back to Permo Triassic times. Evaluation of all available data from the fossil record, trace fossils and facies reveals various trends within the evolution of Mesozoic decapods, and its relation to their ecology. One of these trends is the protection of the long vulnerable abdomen. The most successful method, gradual reduction and inflection of the abdomen against the ventral cephalothorax, led to the Brachyura. Brachyurans exhibit the broadest spectrum of habitat adaptions.