A Deep Hot Zone develops when numerous mafic sills are repeatedly injected at Moho depth or are scattered in the lower crust. The melt generation is numerically modelled for mafic sill emplacement geometries by overaccretion, underaccretion or random emplacement, and for intrusion rates of 2, 5 and 10mm/yr. After an incubation period, melts are generated by incomplete crystallisation of the mafic magma and by partial melting of the crust. The first melts generated are residual from the mafic magmas that have low solidi due to concentration of H20 in the residual liquids. Once the solidus of the crust is reached, the ratio of crustal partial melt to residual melt increases to a maximum. If wet mafic magma, typical of arc environments, is injected in an amphibolitic crust, the residual melt is dominant over the partial melt, which implies that the generation of I-type granites is dominated by the crystallisation of mafic magma originated from the mantle and not by the partial melting of earlier underplated material. High ratios of crustal partial melt over residual melt are reached when sills are scattered in a metasedimentary crust, allowing the generation of S-type granites. The partial melting of a refractory granulitic crust intruded by dry, high-T mafic magma is limited and subordinate to the production of larger amount of residual melt in the mafic sills. Thus the generation of A-type granites by partial melting of a refractory crust would require a mechanism of selective extraction of the A-type melt.

In this feasibility study, Sm/Nd and 143Nd/144Nd ratios have been determined on garnets and coexisting mineral and whole-rock phases from a selection of amphibolite facies rocks, granulite facies rocks and an eclogite. In two micaceous schists and an eclogite the distribution coefficient of the Sm/Nd ratio in garnet relative to other phases is in the range 2-3, i.e. comparable to values for phenocrysts in igneous rocks. In two granulite facies rocks the garnets have Sm/Nd distribution coefficients relative to other phases in the range 13-20 and have yielded precise ages.

Garnet from a granulite facies shear zone from the Limpopo belt in southern Africa, of Archaean age, has yielded concordant duplicate ages of 1974±14 Ma and 1988±14 Ma. These results have been interpreted as relating to the last tectonic event in the region, long after the main gneiss-forming event. Garnet from a granulite facies xenolith from a Carboniferous vent in the Midland Valley of Scotland has yielded an age of 356±10 Ma; this is related to the onset of basic magmatism in the region.

Electron and ion microprobe analyses for P, Si, Ti, Al, Cr, V, Sc, Fe, Mn, Mg, Ni, Co, Ca, Sr, Na, K and Li in olivine, pyroxenes and garnet in forty-two cold and twelve hot garnet lherzolites from kimberlites, nine spinel lherzolites from kimberlites and eighteen from alkali basalts, and one cold garnet lherzolite from the Malaita alnöite, are compared with published data for minerals occurring in lherzolite, harzburgite and eclogite xenoliths, for silicate megacrysts in kimberlites, and for silicate inclusions in diamonds. Despite wide ranges in the chemistry of minerals from garnet and spinel lherzolites, there are distinct regions in composition space that would enable determination of the parent lithology of disaggregated minerals in kimberlites and alkali basalts. Titanium correlates with Fe3+ in garnets. Chromium, Al, V and Sc are distributed similarly between silicates in lherzolites. Sodium correlates with trivalent ions in olivine, and increases with temperature. The distribution of Na, but not of K and Li, between olivine and clinopyroxene correlates with temperature. The regular partitioning of Ti, Mn and Ni places constraints on crystal-liquid partition coefficients. Above the stability temperature of mica, ‘metasomatising fluids’ may scavenge Cr and other trivalent ions as they increase Ti in silicates.

Extensive preparation since 1975 has yielded thousands of specimens of various minute arthropods, mainly phosphatocopid ostracods and other crustaceans. The whole exoskeleton is exceedingly well preserved in three dimensions, showing even delicate mo1 phological details. The material permits not only the description of the morphology and systematic status of the animals but also interpretation, with a high degree of confidence, of the function, life habit, and ontogeny of these early arthropods.

The initial adaptive radiation of the Placodermi took place rapidly following the development of the basic placoderm adaptive complex after the ancestral scale covering of the trunk fused into a rigid shield, and not long before the group appears in the fossil record in the Lower Devonian. The radiation was mainly concerned with different ways of living in the benthos of a variety of marine and fresh-water environments; a few nektonic species appear late in the history of the Arthrodira. The fossil record shows the evolution of the orders in their adaptive zones. The zones become increasingly distinct as the orders evolve and become more specific in their adaptations, and the arthrodire, antiarch and rhenanid zones segregate into successively occupied sub-zones. The evolution of the Placodermi has been previously described in terms of improvements in the locomotor mechanism by an analysis of changes in the trunk-armour and pectoral fins. A more detailed description can be given by considering the feeding mechanism as well; this is particularly true of the largest order, the Arthrodira. Study of the feeding mechanism involves the cervical joints as well as the jaws and gnathals. The cervical joints had the same functions in feeding as the anterior part of the vertebral column (“the neck”) in many higher fish. In arthrodires jaw action involved vertical movements of the mandibular lever; the upper jaw apparatus is comparable to the rigid palatoquadrate-maxillary complex of primitive bony fish. The mandible was transformed into a bent lever in Brachyosteus by the development of a small “coronoid” process, but the arthrodire jaw apparatus remained undeveloped in comparison with Actinopterygii and Elasmobranchii. Arthrodire jaw suspension was autostylic. Evidence from the Rhenanida and Ptyctodontida has been interpreted to suggest that this condition was secondary, and that primitive placoderms had an elasmobranch or holocephalan-like palatoquadrate with hyostylic suspension. This view is not entirely supported by the state of the palatoquadrate in primitive arthrodires, but there is no good evidence that placoderms had a complete, open spiracular gill-slit (the aphethoyoid condition). Arthrodire phylogeny cannot yet be described in vertical lines, but four successive levels of organization of increasing efficiency can be recognized; the actinolepid, phlyctaenaspid, coccosteomorph and pachyosteomorph levels. These levels can be defined by simple characters relating to broad adaptations in the locomotor and feeding mechanisms. Evolutionary trends in the Arthrodira include the enlargement of the scapulocoracoid and base of the pectoral fin and the reduction of the spinal plate and flank armour, as the fish gain better control in the water and more myomeres become available for use in swimming; and the enlargement of the nuchal gap and development of the cranio-thoracic joint as powerful muscles are developed to raise the head to give a wide gape, accompanied by the specialization of the gnathals for different modes of feeding. Some of these trends are reversed in compressed, nektonic species. The description of arthrodire phylogeny in terms of changes that can be understood from a functional point of view reveals interesting examples of mosaic and parallel evolution. Parabelosteus n.gen. is erected.

A detailed examination of the boundary between Peach and Home's (1899) Northern and Central Belts, in the Lower Palaeozoic rocks of the Southern Uplands of Scotland and their Irish continuation in Longford–Down, demonstrates it to be a major Caledonian sinistral wrench fault. Maps and descriptions of outcrops of the fault at Slieve Glah in County Cavan, at Orlock Bridge on the NE coast of Down, at Cairngarroch on the Rhinns of Galloway and at Garvald, 11 km SW of Dunbar, are presented. A distinctive fault fabric, clearly the result of repeated sinistral slip at the first three localities, characterises these outcrops along the 400 km trace. The fabric elements include: (1) lenticular shearing of arenites in the fault protolith, (2) a fault-associated phyllonitic fabric, with abundant pressure-solution seams, overprinting the regional S1, (3) numerous foliation-parallel quartz segregations, (4) refolding of the regional S1 cleavage, segregation veins and the phyllonitic fabric in at least two generations of steeply plunging, sinistrally-verging folds, (5) a non-penetrative crenulation cleavage, (6) a locally developed sinistral S–C fabric, and (7) typical protomylonite, mylonite and ultramylonite textures in thin sections of the fault rocks. The zone of fault-associated deformation varies from a few metres across at Garvald to over 1 km at Slieve Glah. The fault thus differs from other tract-defining faults in the Southern Uplands, particularly in its clear evidence of ductile, quasi-plastic, deformation at depth, in its lack of associated imbrication, in its significant refolding and overprinting of the regional S1 cleavage and in its uniquely large stratigraphic effects.

The main fabric-generating movement postdates the regional, accretion-related, S1 cleavage but predates minor Caledonian (c. 400 Ma) intrusions. For its entire length the fault separates the Northern Belt, with proof only of Upper Ordovician turbidites, and the Central Belt, where turbidite deposition began in the Silurian. The age of the base of the turbidites decreases southward by some six graptolite zones at the fault trace. The fault thus either (a) excises the equivalent of four tracts from a single accretionary-prism terrane, or (b) juxtaposes two distinct terranes, one of late Ordovician and the other of Silurian age. In either event the sinistral slip is probably in excess of 400 km.

The structure of the fishes constituting the Devonian genus Cheirolepis is investigated. Articulated remains from the Scottish Old Red Sandstone and the Canadian Upper Devonian were studied and two species are recognised, the Middle Devonian C. trailli and the Upper Devonian C. canadensis. Recently described fragmentary material from Europe assigned to the genus is best regarded as incertae sedis. Cheirolepis is the earliest actinopterygian genus with extensive material. Neurocranial remains are described, with a crossopterygian-like parasphenoid. There seems to have been a mobile rostral region with several small bones between premaxilla and postrostral extending to the anterior corner of the small dermal orbit. New palatoquadrate and pectoral girdle material is described. The apparently primitive nature of the head skeleton is related to the cranial dynamics and the likely mode of life. The small scales and the elongate-fusiform body shape are a corollary of the method of swimming. The ecology of the fishes is touched upon. A taxonomic investigation of the two species was carried out but although interesting variation in the scale-row numbers came to light no changes at the specific level were deemed necessary. Revised generic and specific diagnoses are given.

Water is perhaps the most abundant and most universally distributed form of matter on the earth. It has to perform more varied functions and more important duties than any other kind of matter with which we are acquainted. From its close connection with all forms of life, it has been the subject of deepest interest in all ages. It is constantly changing from one of its states to another. At one time it is solid, now liquid, and then gaseous. These changes take place in regular succession, with every return of day and night, and every successive season; and these changes are constantly repeating themselves with every returning cycle. Of these changes, the one which perhaps has the greatest interest for us, and which has for long ages been the subject of special observation, is the change of water from its vaporous state, to its condensation into clouds, and descent as rain. Ever since man first “observed the winds “and “regarded the clouds,” and discovered that “fair weather cometh out of the north,” this has been the subject of intensest human interest, and at present forms one of the most important parts of the science of meteorology, a science in which perhaps more observations have been made and recorded than in all the other sciences together.

The Mollusca possess two growth axes associated with the body and the mantle/shell respectively. Evolution of the Lamellibranchia involved assumption by the mantle/shell of responsibility for growth and form; also the formation of anterior and posterior adductors. Change from this dimyarian to a heteromyarian and finally monomyarian condition involved changes in the relations of body and mantle/shell. These are considered with regard to their two major axes in the saggital plane. Although body form may be greatly influenced by changes in that of the mantle/shell, its proportions are only altered where the body is attached by byssus. Then the anterior half may be reduced and the anterior adductor finally lost. This occurs in many Anisomyaria, these monomyarians being divisible into five groups according to habit. Apart from the Limidæ, all are bilaterally asymmetrical. In those which have lost the foot, both axes of the body are effectively lost. Primary pallial attachment is lost but secondary attachment has been achieved in most cases. In the Tridacnidæ (Eulamellibranchia) the mantle/shell has rotated in the saggital plane in relation to the fixed body with resultant loss of the anterior adductor. There is hypertrophy and extension antero-dorsally of the originally posterior siphonal tissues in which zooxanthellæ are contained. Evolution may have proceeded by way of now extinct heteromyarians such as Lithocardium. In the freshwater Acostœa (Mulleria) the anterior adductor is lost during growth following cementation by one or other valve. The form of the body is little affected. Variation and natural selection account for the variety of form and habit exhibited by these diverse monomyarians.

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.

Owing to the comparatively rare occurrence of most tectibranchs it is perhaps not surprising to find that, in spite of the ever-increasing knowledge of the histological structure and functioning of the alimentary canal of molluscs, little attention has been paid in this respect to these forms. The structure of the radula and gizzard has previously aroused interest, but so far as the rest of the alimentary canal is concerned most workers have been content with a somewhat superficial description of the gross morphology such as was given by Vayssière (1880), Bouvier (1893), Pelseneer (1893, 1894), and Guiart (1901). A more recent account of the digestive tract of Philine aperta has been given by Brown (1934), but this includes no histological or physiological consideration.

The Miocene Baltoro granite forms a massive plutonic unit within the Karakoram batholith, and is composed of comagmatic monzogranites and leucogranites with a mineralogy consisting of quartz-K-feldspar-plagioclase-biotite ± muscovite ± garnet, with accessory sphene, zircon, monazite and opaques. Geochemically the Baltoro granites are mildly peraluminous, and show a calc-alkaline trend on trace-element normalised diagrams with high LIL/HFS element ratios and negative Nb, P and Ti anomalies. REE are strongly fractionated with little or no Eu anomaly. Leucogranites are depleted in most elements compared to monzogranites with notable exceptions being Rb, K and the HREEs. Initial 87Sr/86Sr ratios are 0·7072-0·7128, considerably lower than High Himalayan leucogranites (0·74-0·79), and are indicative of a lower continental crust source. The probable petrogenesis of the Baltoro granite involves dehydration melting of a biotite-rich pelite to produce a voluminous, hot, water-undersaturated magma which could then separate from its source and intrude through an already thickened and still hot crust. Fractional crystallisation of the monzogranites produced the leucogranites and a pegmatite dyke swarm. A suite of lamprophyre dykes including amphibolerich vogesites and biotite-rich minettes intrude the country rock, dominantly to the north, around the Baltoro granite. These calc-alkaline shoshonitic lamprophyres are volatile-rich mantle-derived melts intruded around the same time as the granite, indicating simultaneous melting of the mantle and lower crust beneath the Karakoram during the Miocene, approximately 30 Ma after the India-Asia collision which initially caused the crustal thickening. Intrusion of mantle melts provided heat to promote crustal melting and may have selectively contaminated the granite magma.

The Baltoro granite intrudes sillimanite gneisses with melt pods along the southern margin indicating temperatures above 700°C at the time of intrusion. Locally, internal fabrics and numerous aligned xenoliths along the southern margin in the Biafo glacier region indicate steep, southward-directed thrusting during emplacement. Along the northern contact, the Baltoro granite intrudes anchimetamorphic to greenschist facies metasedimentary rocks with an andalusite-bearing contact aureole. Northward-directed culmination collapse normal faulting during Miocene emplacement is inferred, in order to explain the P-T differences either side of the pluton. This also provided an extensional stress regime in the upper crust to accommodate the rising magma.

The purpose of this communication is to describe, and to discuss the significance of, (1) the extensive morainic deposits and the associated glacial phenomena of the tract of country situated immediately west of the Ochil Hills (border of Perthshire and Fifeshire)—between the River Teith near Doune, north of Stirling, and the River Tay in the vicinity of Perth; and (2) a well-marked later glaciation studied in the region of Loch Lomond and the Upper Forth valley.

In both instances these glaciations involved readvances of the ice during the retreat of the last general ice-sheet, and for brevity and convenience of description they will be referred to as the Perth Readvance and the Loch Lomond Readvance.

Many hypotheses have been proposed for the rise of dinosaurs, but their early diversification remains poorly understood. This paper examines the occurrences, species diversity and abundance of early dinosaurs at both regional and global scales to determine patterns of their early evolutionary history. Four main patterns are clear: (1) sauropodomorph dinosaurs became abundant during the late Norian–Rhaetian of Gondwana and Europe; (2) Triassic dinosaurs of North America have low species diversity and abundance until the beginning of the Jurassic; (3) sauropodomorphs and ornithischians are absent in the Triassic of North America; and (4) ornithischian dinosaurs maintain low species diversity, relative abundance and small body size until the Early Jurassic. No one hypothesis fully explains these data. There is no evidence for a Carnian–Norian extinction event, but sauropodomorphs did become abundant during the Norian in some assemblages. No clear connection exists between palaeoenvironment and early dinosaur diversity, but environmental stress at the Triassic–Jurassic boundary is consistent with changes in North American dinosaur assemblages. Elevated growth rates in dinosaurs are consistent with the gradual phyletic increase in body size. This study demonstrates that early dinosaur diversification was a complex process that was geographically diachronous and probably had several causes.

The Scottish specimens described in this memoir are all the property of Mrs Robert Gray, and I am exceedingly indebted to her, not only for entrusting them to me for description, but for allowing them to remain in my hands for a lengthy period. None of the collections of the Geological Survey, nor those of Glasgow University, contain any cystids from Girvan. Besides her Cystidea, Mrs Gray has submitted to me her Crinoidea and several interesting specimens of Edrioasteroidea.

Four tiny obvious larvae unassignable to larger “Orsten” arthropods are described and reconstructed. Two of them have a naupliar appearance and may belong to crustaceans, whereas the other two represent as yet unknown types of euarthropodan larvae.

A description of the head, mandible, pectoral girdle, humerus, medial fins and their supports, and the dissociated vertebral column has been prepared for Onychodus jandemarrai n. sp. from the Gogo Formation (Frasnian) of Western Australia. This is the most completely known species of the genus. The feature influencing most of the head morphology is the retractable parasymphysial tusk whorls. Their presence has caused a reorganisation of the braincase, palate (including the loss of the vomers), and lateral displacement of the nasal capsules. The extensive mandibular articulation is in cartilage, and the mandibular symphysis is weak. This makes for a kinetic skull. There is a single submandibular on each side. The vertebral column is poorly ossified, consisting of intercentra which have no ventral contact, and pleurocentra. The neural arches have no longitudinal ligament, have unequal sides, and asymmetrical placing of the dorsal and ventral nerve root foramina. Each arch has an anterior surface that often attaches to the next anterior arch. The caudal fin is almost diphycercal; all the medial fins have strong support structures. An attempt is made to discuss the functional morphology of many features of the skeleton.

A growing body of evidence indicates that some, and perhaps most, plutons are highly composite. However, the geometrical forms of increments and the processes by which they are added are poorly known. Magmatic crack-seal probably is an important incremental assembly process, particularly in the upper crust where wall-rock fracture is important. Evidence for magmatic crack-seal is clearest where it is antitaxial, i.e., new fractures form at the contact between wall rock and a growing intrusion. Local deviation of antitaxial cracks into wall rocks isolates wall-rock bodies that therefore mark increment contacts. Wall rock isolated by this process remains in situ and thus is likely to preserve a ghost stratigraphy. Previously described plutons are identified, and interpreted to have grown by antitaxial magmatic crack-seal. In contrast, it remains unclear what observable geological record may remain in plutons formed by syntaxial crack-seal, in which new cracks form in the middle of the growing pluton. Several plutons are identified that preserve possible indirect evidence for growth by syntaxial crack seal, but conclusive identification of a direct record of the process remains elusive. However, plutons with sharp discordant contacts but few xenoliths may have been emplaced incrementally by syntaxial magmatic crack-seal.

The nature of the fauna of brackish-water environments is reviewed. It is concluded that: (a) a specific brackish-water macrofauna does not exist; (b) in salinities of >c. 5-8‰ the fauna is one that also occurs in soft sediments under fully marine conditions when circumstances (possibly the absence of competing species) permit; (c) in salinities of <c. 5‰ the fauna is essentially a freshwater one; and (d) inability to cope physiologically with brackish water is not a factor of major importance in limiting species diversity in these habitats, except in the vicinity of 5‰ salinity. Caution is therefore advised in assigning brackish status to past environments on the basis of their preserved fauna.

Torridonian and Lewisian rocks lying in inverted order between the Kishorn thrust and the Moine thrust, and Moine rocks, lying above the Moine thrust, exhibit a remarkable parallelism of structure.

New evidence shows that three sets of minor structures have been developed in the formations during the Caledonian movements. At least two of these sets pre-date the Moine thrust movements. The mylonites, which are not restricted to the vicinity of the Moine thrust outcrop, belong to an earlier movement phase than these structures and are not directly related to the clean-cut thrust movements. They appear to represent narrow zones of shearing and sliding, mainly within the Lewisian gneisses that developed early in the Caledonian orogeny.

There is reason to suppose that the inversion of the rocks to the west of the Moine thrust occurred before the formation of the minor structures recognized in the paper.

The minor structures are described and their order of formation established. The plastic, para-crystalline style of the earlier deformation is contrasted with the post-crystalline brittle style of the later deformations.