The Hyllingen Series, comprising the southern part of the 160 km2 Caledonian, synorogenic, layered, mafic Fongen-Hyllingen intrusion, southeast of Trondheim, Norway, crystallized from a basaltic parent magma at 5–6 kb. Well developed modal layering strikes directly towards the magma chamber wall at the southern margin. The lithologies of inclusions match those of the adjacent country rock envelope. Eleven sample profiles (274 samples; over 1000 mineral analyses) allow the Hyllingen Series to be subdivided into four stages which generally decrease in thickness from north to south. A sporadically developed Stage I (< 100 m thick) of mostly unlayered ferrodiorites with relatively evolved compositions occurs at the base. Stage II (400–1500m) consists of layered, broadly ferrodioritic rocks. Modal layering is undisturbed by numerous, plate-like, metabasaltic inclusions which occupy about 22% of Stage II. Cryptic variation is slight except at the top where more evolved compositions are developed. Stage III (350–550 m) has fewer inclusions and is characterized by a gradual regression to more primitive compositions. Stage IV (800–2300 m) shows normal fractionation patterns, and a few minor reversals, with extremely evolved rocks in contact with the roof. Occasional metapelitic, platy inclusions occur near the southern margin.

The rocks gradually become more evolved along the strike of modal layering towards the southern margin. Olivine and plagioclase vary systematically over about 7 km from Fo23:An46 to Fo7:An35 along the Stage II/III boundary and from Fo75:An63 to Fo13:An42 along the Stage III/IV boundary. While the apparent angle of discordance between modal and cryptic layering is usually less than 20°, in Stage IV near the wall of the magma chamber they are highly discordant (approaching 90°) over a thickness of about 900 m.

In situ crystallization along an inclined floor took place from magma which became stratified by double-diffusive convection during Stage II. Modal layering developed concordant with the crystallization front while cryptic layering developed essentially parallel to the liquid stratification. Gradual influx of dense hot primitive magma caused elevation of the stratified magma column in Stage III. During this uplift, progressively more primitive liquid came into contact with earlier crystalline products along the inclined crystallization front. Crystallization during uplift of the magma column gave rise to the gradual regression of Stage III. Highly discordant modal/cryptic layering relations in Stage IV require that the magma became zoned with a horizontal, as well as a vertical, component. This may have occurred in response to lateral cooling establishing horizontal thermal gradients within individual magma layers, the resulting tendency towards density increase being compensated by material diffusion to more evolved compositions.

The original roof during the early period of magma chamber evolution is possibly among the inclusions in Stage II. Repeated magma addition during Stage II resulted in the top magma layer having an extremely evolved composition. During the later stages this buoyant, evolved liquid ultimately crystallized to produce the quartz-bearing syenite in contact with the roof.

Graptolite ultrastructural studies using SEM and TEM are now more than a quarter of a century old and have reached the stage of routine procedure where material of suitable preservation is to hand. (Specimens may be chemically isolated from the rock or, in SEM and STEM studies, may be retained in the matrix.) Description of ultrastructural elements has evolved from the purely empirical to a conceptual/biological classification following (a) the recognition of the role of collagen fibrils in skeletal construction, and (b) the realization that at least two models of skeletal secretion are possible, each essentially opposed to the other. In this paper old terminology is explained with reference to current practice and the latter is tabulated and referred to the increasing number of described ultrastructural elements. Some of the outstanding problems of interpretation are outlined; and a set of concise, in some cases revised, ultrastructural definitions is given together with illustration, where appropriate, of the ultrastructural element defined.

Two late Quaternary ash-flow tuffs erupted as single flow units from the Menengai caldera volcano, Kenya, show marked facies variations. Dense welding of the tuffs close to source may have been promoted by their peralkaline trachyte compositions. Coarse, near-vent breccias are partly welded and were emplaced as lithic-rich parts of the body of pyroclastic flows. Non-welded tuff forms the bulk of the outflow sheets which include valley-fill deposits up to 520 m above the present caldera rim, indicating ash-flow generation by collapse of moderately high eruption columns. The flows overran varied topography which controlled development of ash-flow tuff facies. On hilly terrain, the flows became turbulent and separated into a lithic-rich, rapidly deposited part and a fine-grained lithic-and crystal-poor part which was subsequently deposited farther from source. Caldera walls belonging to an earlier caldera probably formed a major barrier to the spread of the second ash-flow from its site of collapse. There are many similarities in the facies variations in these peralkaline tuffs and subalkaline tuffs. Differences in welding characteristics and the shape of ash particles reflect the lower viscosity of peralkaline magma.

Young stages isolated from limestone from Girvan indicate that the first two thecae had unsclerotized walls but apertural rings showing that they opened downwards beside the sicula. The development can thus be matched with that of Glossograptus and the two families Glossograptidae and Cryptograptidae merged. The recognition of downward-opening thecae in C. tricornis allows the earlier species C. schaeferi to be more easily accommodated in the genus. Relationships in the family are reviewed.

The permeability of rocks fractured by random, planar cracks, is expressed as a classical bond percolation problem on a random lattice, by Voronoi partition of space. The percolation threshold is determined as a function of the statistical characteristics of the cracks, or of their traces on an arbitrary face of the rock, by using an empirical quasi-invariant of percolation theory.

The Jarawa Younger Granite complex is composed of high silica alkali granites that were emplaced 161 Ma ago. The granites are characterized by high contents of Rb, Li, F, Sn, Nb, W above normal low-Ca granitic rocks and have typical S-type characteristics that are indicative of a substantial component of crustal melt.

Mineralization in the complex is associated with the biotite granite which was emplaced as a sheet-like body at relatively shallow depth and occurs as disseminations and as greisen lodes and veins.

Chemical studies of the granites have shown that the biotite granite represents a highly fractionated rock that crystallized from a residual magma from which the hornblende-biotite granite had previously crystallized. However the biotite granite is characterized by steep gradients in some minor and trace elements that apparently indicate that liquid-state differentiation and/or volatile complexing processes made significant contributions to their differentiation. Enrichment of Th, Li, Rb, Sn, W and Nb may be more closely linked to roofward migration of F.

Acidic welded tuffs of the Capel Curig Volcanic Formation were deposited from ash-flows that transgressed from a proximal subaerial to a distal submarine environment. Isolated pods of welded tuff enveloped by marine sediments represent a distal facies of the formation deposited beyond the limits of the continuous ash-flow tuff sheets. Various mechanisms for the separation of the tuff pods from a parent flow are discussed and it is concluded that it is most likely that the formation of the pods was due to disruption of a subaqueous ash-flow during transport rather than by downslope collapse at the front of a previously deposited ash-flow tuff sheet.

The stratigraphy and fauna of terrestrial deposits at Portland Bill is described. The fauna is restricted to the head and is indicative of a cool climate rather than the temperate conditions suggested by earlier authors. An age in the Devensian Glacial Stage is tentatively suggested.