Caldera Formation Research Articles

Satellite reliability engineering : W. A. Hogg, IEE Conf. Reliab. Electron., Conf. Publ. No. 60, 10–12 December (1969), p. 160

New multibeam mapping and whole-rock geochemistry establish the first order definition of the modern submarine Kermadec arc between 30° and 35° S. Twenty-two volcanoes with basal diameters > 5 km are newly discovered or fully-mapped for the first time; Giggenbach, Macauley, Havre, Haungaroa, Kuiwai, Ngatoroirangi, Sonne, Kibblewhite and Yokosuka. For each large volcano, edifice morphology and structure, surficial deposits, lava fields, distribution of sector collapses, and lava compositions are determined. Macauley and Havre are large silicic intra-oceanic caldera complexes. For both, concentric ridges on the outer flanks are interpreted as recording mega-bedforms associated with pyroclastic density flows and edifice foundering. Other stratovolcanoes reveal complex histories, with repeated cycles of tectonically controlled construction and sector collapse, extensive basaltic flow fields, and the development of summit craters and/or small nested calderas.Combined with existing data for the southernmost arc segment, we provide an overview of the spatial distribution and magmatic heterogeneity along ∼780 km of the Kermadec arc at 30°–36°30′ S. Coincident changes in arc elevation and lava composition define three volcano–tectonic segments. A central deeper segment at 32°20′–34°10′ S has basement elevations of > 3200 m water-depth, and relatively simple stratovolcanoes dominated by low-K series, basalt–basaltic andesite. In contrast, the adjoining arc segments have higher basement elevations (typically < 2500 m water-depth), multi-vent volcanic centres including caldera complexes, and erupt sub-equal proportions of dacite and basalt–basaltic andesite. The association of silicic magmas with higher basement elevations (and hence thicker crust), coupled with significant inter- and intra-volcano heterogeneity of the silicic lavas, but not the mafic lavas, is interpreted as evidence for dehydration melting of the sub-arc crust. Conversely, the crust beneath the deeper arc segments is thinner, initially cooler, and has not yet reached the thermal requirements for anatexis. Silicic calderas with diameters > 3 km coincide with the shallower arc segments. The dominant mode of large caldera formation is interpreted as mass-discharge pyroclastic eruption with syn-eruptive collapse. Hence, the shallower arc segments are characterized by both the generation of volatile-enriched magmas from crustal melting and a reduced hydrostatic load, allowing magma vesiculation and fragmentation to initiate and sustain pyroclastic eruptions. Proposed initiation parameters for submarine pyroclastic eruptions are water-depths < 1000 m, magmas with 5–6 wt.% water and > 70 wt.% SiO2, and a high discharge rate.

The geology and geophysics of the ambrym caldera, New Hebrides

Ambrym Island has an unusually large, well-preserved basaltic caldera 13 km across. The caldera occurs in the central region of an early broad composite cone, which formed a north-south line with three smailer volcanoes. Alter the caldera was formed volcanism occurred within it and along fissure lines running nearly east-west. Two volcanic cones are active almost continuously and historic fissure cruptions have been recorded. The caldera formed by quiet subsidence, or by subsidence accompanied by eruption of scoria lappili similar to that erupted prior and subsequent to caldera formation. The collapse was at least 600 metres and radiocarbon dating suggests it took place less than 2000 years ago. The caldera is detined by gravity anomalies 10 to 14 milligals lower than those at its rim suggesting predominantly ash infilling. Aeromagnetic anomalies show a prominent. nearly east-west lineation, with normally magnetised bipole anomalies over the centre of the caldera and over fissure lines east of it. The source of the present volcanic activity is believed to be located along dyke fissures, with a perched magma chamber beneath the caldera. The geophysical evidence on Ambrym, together with that of regional east trending magnetic anomalies and recent bathymetric results, suggests that the volcanic activity is localised by the intersection of an east-west fracture zone with the axis of the New Hebrides island are.

The structural control and genesis of alkaline rocks in central Kenya

In the Nairobi area and the adjacent region of the Gregory Rift Valley of Kenya two series of Cainozoic alkaline volcanic rocks, of mildly and strongly alkaline type respectively, are considered to have been derived from a single parental magma of alkali olivine basalt composition. Vulcanicity is genetically associated with tectonic movements attendant upon doming and rifting and distances from the rift margin decrease with crystal fractionation, the most acid differentiates being located at the maximum elevation of the rift floor. An early period of nepheline-bearing lava extrusion associated with central vulcanism is followed by a later period of welded tuff — trachyte — rhyolite fissure eruption. Caldera formation associated with central vulcanism within the Rift accompanies the later volcanic phase.

22. Some remarks on structure and formation of calderas(Abstract of Papers submitted for the Meetings of the Society)

22. Some remarks on structure and formation of calderas(Abstract of Papers submitted for the Meetings of the Society)

A STUDY ON THE FORMATION AND THE TILTING OF THE MYOKO VOLCANO, CENTRAL JAPAN

The Myoko Volcano is a double volcano formed in the Quaternary Era, which is situated at the northern extremity of the Fuji Volcanic Zone. Author has performed a tephrochronological research of the Myoko Volcano and has considered the process of its formation. Moreover, he has investigated the geological structure of the basement complexes and the geomorphological characteristics of the Myoko Volcano and its marginal area. As the result, he has come to conclude as follows : 1) The basement complexes of the Myoko Volcano are composed of the Neogene system. They have several anticlinal and synclinal axes stretching nearly from the northeast to the southwest with the complicated structure. The Myoko Volcano is located on the eastern wing of the anticlinorium on the top of which the mountains such as Hiuchi-san, Yake-yama Volcano, and others are distributed. 2) The ejectas of the Myoko Volcano are stratigraphicaly divided into some following units, namely Pyroclastic flow, Pumice flow, Middle loam, Upper loam, Mud flow I, Mud flow II, Tsubame lava, Older central cone pyroclastic flow, and Younger central cone pyroclastic flow in ascending order. Moreover, there are Somma lava and Central cone lava. 3) Comparing the volcanic ejectas of the Myoko Volcano with the type locality of Japanese Quaternary system in the Southern Kanto District, it may be assumed that Pyroclastic flow, Pumice flow and Middle loam of the Myoko Volcano are correlated to the deposits of Shimosueyoshi stage and Musashino loam. Myoko Upper loam and Mud flow I•II are correlated to Tachikawa loam. Tsubame lava and Older central cone pyroclastic flow and Younger one are considered to have been deposited in the Alluvial Epoch. 4) The forming processes of the Myoko Volcano are divided into some following stages, namely the construction of strato volcano (from Pyroclastic flow or Pumice flow to Upper loam, Somma lava), the formation of caldera (Mud flow I, 19600±600 Y.B.P. or 19300±200 Y.B.P.), the breaking of northern external wall of somma (Mud flow II, 17900±450 Y.B.P.), the effusion of Tsubame lava, the construction of central cone (Older central cone pyroclastic flow, 8640±100 Y.B.P.; Younger central cone pyroclastic flow, 4790±110 Y.B.P.). 5) It may be assumed that the body of the Myoko Volcano has been suffering the crustal movement tilting eastward throughout its constructing process. Such a movement of the Myoko Volcano is presumed to have been caused by the folding of the anticlinorium of the basement complexes. 6) The evidences for the above mentioned tilting are as follows ; a) Land features of the east side slope of the Myoko Volcano and its foot areas suggest that this area has been deformed by the internal forces pushing from east to west, b) the somma was broken only on the east side, and c) the valley system of east-west diretcion has developed in foot areas. 7) The folding of the anticlinorium of the basement complexes have the scale of more than 20 km in wave length. It may be assumed that the maximum amounts of subsidence of the east side slope and the maximum ones of uplift of the west side slope by the tilting respectively are about 220 m and 240 m, which have been caused during the time from the outflow of Pyroclastic flow and Pumice flow to the present. According to this assumption, the tilting rates have the order of the magnitude of 8.4×10-7 a year.

Isotopic Dating and Geomagnetic Polarity Studies on Volcanic Rocks from Mauritius, Indian Ocean

Results of a geochronological and paleomagnetic investigation of the volcanic island of Mauritius are reported. Potassium-argon dates indicate that the main shield volcano was built subaerially between about 7.8 and 6.8 m.y. ago in the early Pliocene. Some evidence is presented for caldera formation following the construction of the volcanic shield. Profound erosion subsequently destroyed much of the volcano, leaving only peripheral steep-sided massifs. Lavas of the Younger Volcanic Series were erupted from about 3.5 m.y. ago until less than 0.2 m.y. ago, flooding the denuded stumps of the shield volcano. This activity was intermittent and a hiatus is recognized from about 2 to 0.7 m.y. ago, which we have taken as the break between the eruption of the Early and Late Lavas of the Younger Volcanic Series. The mean directions of magnetization of the three groups of lavas are not significantly different from one another and are close to an axial dipole direction. The combined paleomagnetic and dating studies on the lavas also yield further information on the geomagnetic polarity time scale; 16 results on rocks with ages lying between 3.5 and 0.17 m.y. are consistent with the established scale for this period of geological time. Of particular significance are results that confirm the Mammoth reversed event at 3.0 m.y. ago, and the age of the boundary between the Gauss normal and Gilbert reversed polarity epochs at 3.35 m.y. Nineteen results were obtained on rocks whose ages are between 4.6 and 7.9 m.y., and these data generally are consistent with the polarity time scale based upon extrapolations from the sea-floor spreading hypothesis. The evolution of the Mascarene Islands, comprising Mauritius, Reunion and Rodriguez is briefly reviewed and it is concluded that each island had developed independently, and that no correlation of age with respect to distance from the Mid-Indian Ocean Ridge is observed. The volcanism that has built Mauritius and Reunion may be a continuation of the development of the Mascarene Plateau with migration of the activity generally southward with time.

Some problems of the theory of caldera formation

Some problems of the theory of caldera formation

The evolution of the volcanoes of Aden and Little Aden, South Arabia

Along the south coast of Arabia, between the town of Aden and the southern entrance to the Red Sea, there are six central-vent volcanoes of Miocene-Pliocene age. The degree of erosion is ideal, and the abundance of exposures makes detailed structural and evolutionary studies possible; data are presented for the three easternmost cones, those of Aden, Little Aden and Ras Imran. Each volcano evolved through a complex cone-building stage during which the predominant rock types were trachybasalt, trachyte and peralkaline rhyolite. This was followed by periods of caldera formation during which trachyandesites, trachytes, and peralkaline rhyolites were erupted. A final stage of parasitic activity included small amounts of olivine-tholeiite amongst the eruptive products.

A contribution to the petrology of Tutuila, American Samoa

Eight new chemical analyses of rocks of Tutuila Island are given, and 20 new analyses of rocks of the Manu'a Islands will be reported elsewhere. Like the older analyses, all are of alkalic rocks. Tholeiitic rocks have not been found in Samoa, although they may be present in the submerged basal parts of the volcanoes. The differentiation trend of Samoan rocks for the most part closely parallels that of the Hawaiian alkalic suite. In both areas the probable parent magma is oceanitic. Quartz trachytes of Tutuila and rhyolitic obsidians of Easter Island are clearly members of the alkalic suite, although the rhyodacite of Hawaii still appears to be a tholeiitic derivative. Differentiation of alkalic olivine basalt magma to trachyte in some instances and to quartz trachyte or even sodic rhyolite in others may result from differences in the proportion of pyroxene to olivine in the separated material, which may in turn be a function of the depth at which crystallization takes place. In the Samoan Islands caldera formation took place after a considerable period of eruption of alkalic rocks, instead of before the appearance of alkalic rocks as is the case in Hawaii. Thus the details of the Hawaiian pattern cannot be extended to all other oceanic volcanoes.

Crustal structures that produce eruptions of welded tuff and formation of calderas

Accumulation of a huge amount of welded tuff and subsequent formation of calderas are typical examples of volcanisms on a large scale. Crustal structures which are related to the phenomena may be a key for the study of volcanisms on the earth. In Japan there are some 10 calderas of low gravity anomaly type, large and small in diameter, and accordingly in volume of their ejecta, welded tuff. The low gravity anomalies are very characteristic inside the calderas and just concentric with their rims, and are due to mass deficiency caused by eruptions of a huge amount of ejecta. The author (1957), hitherto, has clarified the subterranean structures resulting from the caldera formations by a gravimetric method. In this paper, the author describes the crustal structures beneath Sikotu Caldera in Hokkaido, Japan, deduced from the gravimetric, geomagnetic, seismic and geologic surveys. And referring several examples of the subterranean structures of the other calderas in Japan, the author wishes to discuss the crustal structures that would produce the caldera formations.

Acid volcanism on the south Arabian coast

Along the south coast of Arabia, between Aden and the southern entrance to the Red Sea, there are six central vent volcanoes of probable Pliocene age. All are characterised by the interstratification of basic and acidic extrusives, the formation of large central calderas at a late stage in the volcanic cycle and the subsequent infilling of these calderas with horizontal acidic ignimbrites and basic lavas.

Ignimbrite eruptions in the volcanic history of the glencoe Cauldron subsidence

AbstractThe field relations and petrography of two thick ignimbrite horizons are described from the volcanic pile within the Glencoe Cauldron and are correlated with two pyroclastic horizons in the Lorne Plateau lavas. Their recognition enables two cycles in the volcanic history, culminating in cauldron‐subsidence and caldera formation to be discerned. The role of variation in gas pressure in depth is discussed in relation to the formation of the ring‐faults bounding the Cauldron‐Subsidence, and the volcanic history of the Glencoe Cauldron Subsidence reviewed.

Petrological studies of the Usu volcano

Usu, an active volcano situated in southwestern Hokkaido, consists of a somma, central lava domes, parasitic lava and crypto-domes. In early Holocene, on the southern wall of the Toya caldera, Usu volcano was constructed as a stratocone by repeated eruptions of lava flows and scoriae of basalts and andesites. After a quiescence, its summit was broken by violent explosions which led to the formation of a small caldera. Then the composition of magma changed to dacitic, and consequently the type of activity was converted into formation of lava domes and explosive eruptions of pumice or volcanic ash. Such activities have taken place intermittently also in historic times. Showa-Shinzan is the latest lava dome which protruded on the eastern foot of this volcano in 1943-1945. In the present paper, petrology of the somma lavas is mainly concerned. Compositions of the somma lavas varied gradually from augite-hypersthene-olivine basalt to pigeonite-hypersthene andesite, in accordance with the sequence of eruption. Simultaneously the composition of olivine and Ca-poor pyroxene, contained in the later somma lavas, became richer in Fe, and finally pigeonite crystallized out as phenocryst. Mineralogical and chemical studies on the somma lavas show that the somma lavas were derived from tholeiitic magma by fractional crystal-lization. Dome lavas are all very similar, but have quite different characters from those of the somma lavas both in mineralogy and chemistry. Another interesting problem about Usu volcano, is what kind of process took place during the crystallization of magma at the stage of sudden change of character as above mentioned. The writer describes some petrographical observations on the dome lavas, and he considers that contamination of granitic rocks by mafic magma is most suitable to explain these observations and sudden change in the composition of magma.

Volcanic Eruption and Nature of the Pyroclastic Deposits

This article deals with the mechanism of eruption and transportation of the pyroclastic material and the nature of the resultant deposits from the geological standpoint.In Japan, the method of tephrochronology is best applied to pyroclastic deposits of the Quaternary central volcanoes and those related to the Krakatoan calderas. Most of the rocks are andesitic in composition with subordinate amount of basalt and dacite.Three modes of volcanic eruption may be distinguished: 1) projection of pyroclastic materials which form pyroclastic fall deposits, 2) eruption of pyroclastic flows, and 3) outflow of lava flows or extrusion of dome and spine. Table 1 shows characteristic features of the deposits formed by the three modes of volcanic eruption.Tephra, as originally defined by Thorarinsson, signifies only the air-fall pyroclastic materials and its relation to pyroclastic flow is not clear. In this article, all the pyroclastic materials directly connected with volcanic eruptions, irrespective of their origin (i. e. essential, accessory, or accidental) and of their mode of emplacement, are included in the term tephra. The chronology using the deposits of pyroclastic flows are included in the tephrochronology.The small-scale vesiculation occurring at or close to the top of the magma column results in the so-called Strombolian and Vulcanian eruptions. Larger scale vesiculation with longer time duration leads to the Plinian eruption. The greatest vesiculation takes place within the magma reservoir resulting in the formation of a depression caldera. The larger the size of eruption column, the more effective the sorting of the erupted pyroclastic fragments. The larger and denser particles fall first and closer to the vent while the smaller and more vesicular fragments fall farther away. Consequently the deposits of pyroclastic falls are well sorted and exhibit pronounced lateral regular grading in texture and composition. This is in strong contrast with the poorly sorted character of pyroclastic flow deposits, in which all particles travel en masse in a state of turbulent flow.Welding of the deposit is not uncommon in the pyroclastic flow deposits while it is rare in pyroclastic fall deposits except those deposited near the vents of basaltic eruptions.To reconstruct past eruptions from volcanic deposits, it may be necessary to establish definite correlation between stratigraphic units by which volcanic deposits are grouped and time duration by which specific eruptive activity is grouped. A single eruptive cycle, the deposits of which represent such a time unit, is defined as a series of eruptive events limited by fairly long intervals of quiescence. Historic examples indicate that the duration of a single eruptive cycle ranges from a day to several years in most cases. The intervening periods are generally far longer than the duration of single eruptive cycle.From many examples of single eruptive cycles, a rule has been established: the degree of vesiculation of magma gradually decreases toward the end of the cycle. This is expressed in successive eruption of pyroclastic fall, pyroclastic flow, and lava flow from the same vent in case of felsic magma, and of pyroclastic fall and lava flow in case of mafic magma, which fact may indicate that the original magma column responsible for the eruptive cycle was more enriched in volatiles in the upper part than the lower.The close correlation between the recorded sequence of single eruptive cycles and the reultant beds of volcanic materials is described for a few examples. The beds produced by a single cycle of witnessed eruption conformably superpose each other and do not include a layer representing weathering break. It is stressed that such a group of beds of volcanic ejecta, volcanic deposits of a single eruptive cycle, should be taken into account as a stratigraphic unit when precise tephrochronology is undertaken.

II.—Volcanic Neck Emplacement and Subsidence Structures at Dunbar, South-East Scotland

SynopsisStructures in and around volcanic necks at Dunbar are interpreted as belonging to two different stages in volcanism. Some, which are related to an early neck emplacement stage, include small ring structures—probably incipient (crypto-volcanic) necks—filled with marginally flow-banded breccias, dykes of similarly margined unbedded tuff, and tuff-penetrated wall rocks. Among the last are sandstones veined along planes of minor fracture by very fine sand.Other structures belong to a later stage during which bedded tuffs, that had originally accumulated sub-ærially, subsided into the necks in a process here likened to caldera formation. In one large neck the bedding is relatively undisturbed, but in smaller necks there are high inward dips from the margins and collapse structures, sometimes accompanied by reconstitution of the clastics, at the centres. Diagnostic features of subsidence margins are anastomozing veins of carbonate, locally accompanied by hæmatite, and these appear to be related to dykes of flow-banded sandstone and sandy tuff intruded along fractures initiated inside the necks during subsidence. Other structures described are ring-fracture, with or without gouge, and down-drag of wall-rocks: these suggest a solution to the space problem posed by caldera mechanism.

Apercu sur le Quaternaire de Salvador (Amerique centrale)

Abstract El Salvador is essentially a volcanic region in which an older, presumably late Tertiary, complex of andesite and basalt flows and breccias and younger, more acid Quaternary rocks are represented. Following a long period of inactivity during which the Tertiary volcanic masses were considerably eroded, episodic explosive activity occurred in the Quaternary, accompanied by the formation of extensive calderas and ejection of considerable ash. Paleosols were developed in the intervals between explosions, which permit relative dating of the successive episodes. The last stages of activity were characterized by extrusion of mud flows, torrential gullying, and deposition of thick piedmont detrital beds accompanied by reworking of volcanic ash which was redeposited in dammed lakes and other depressions. The development of calcareous crusts in places constitutes evidence of significant climatic fluctuations. There is also evidence of differential subsidence in coastal areas.

The Ard Bheinn Area of the Central Igneous Complex of Arran

Summary The complex is largely surrounded by Old Red Sandstone formations, which were evidently domed up at an early stage in its development. A succession of sediments ranging in age from Permian to Cretaceous, surmounted by Tertiary lavas, is represented within the complex as “remanié masses” which are interpreted as relics of a block that foundered during the formation of a volcanic caldera, the main subsidence being estimated at 3000 feet. Subsequently there arose on the caldera floor a series of volcanic cones, characterized in general by the sequence andesite-dacite-rhyolite, but also showing a great diversity of pyroclastic formations. The position of the various centres is recognized by the disposition of concentric and radial structures. Quasi-horizontal formations in the most elevated parts of the Ard Bheinn area are considered to correspond to remnants of the basal portions of the volcanic superstructures. The andesites and andesitic breccias are evidently extrusive in character, but the later dacites and rhyolites are represented only by their intrusive equivalents. Finally, granites were emplaced, mainly in conformity with the general structure of the caldera, and in all but the Ard Bheinn area are the prevalent formations at the present level of erosion. I. Introduction The complex occupies an oval area about three miles from north to south and nearly four miles from east to west almost in tne centre of the Isle of Arran and is emplaced within strata ranging in age from Old Red Sandstone to Permian. Topographically the area is a dissected plateau, the highest

Formation of calderas and magmatic evolution

Japanese calderas may be classified into two types. The first type includes calderas formed by collapse following the eruption of a tremendous amount of pumice (calderas of Krakatau type of H. Williams). The second type includes those formed without the pumice eruption (calderas of Glencoe type of H. Williams). In both types, post‐caldera eruptions build up cones or domes, usually inside the calderas but sometimes on their margin. Most of the pre‐caldera lavas consist of rock series originating through fractional crystallization of basaltic magma. In the calderas of Krakatau type, the post‐caldera lavas comprise a rock series formed through contamination of the same magma by granitic material, whereas in those of Glencoe type the post‐caldera lavas represent an advanced stage of crystallization of the pre‐caldera lavas. In the former type, the explosive eruption of the pumice probably shattered the roof of the magma reservoir and facilitated the contamination, whereas in the latter the subsidence took place along a simple system of ring fissures without fracturing of the subsided mass and without giving opportunity for contamination.

IV.—The Geology of Slieve Gullion, Foughill and Carrickcarnan: An Actualistic Interpretation of a Tertiary Gabbro-granophyre Complex

Synopsis:An interpretation of the Tertiary rocks of the area, which are shown to be denudation relics of a gigantic volcano, is found by following the actualistic principle, that is by interpreting the unknown with reference to the known or actual. “Actualistic”, i.e. with reference to actual causes—commonly used by continental geologists—is chosen in preference to “uniformitarian” since the latter stands in contradistinction to “catastrophic”, and is not, therefore, altogether appropriate to some of the more violent aspects of volcanic activity. Gabbros and granophyres, hitherto thought to be plutonic intrusions emplaced within Caledonian granodiorite, are shown to be highly metamorphosed basaltic and rhyolitic lava-flows, agglomerates and tuffs, together with some gabbro sills. These rocks form a layered series which is almost horizontal at Slieve Gullion, and dips at various angles at Foughill and Carrickcarnan. A Tertiary ring-dyke of granophyre and felsite encircles the layered rocks and outlines a caldera within which the basaltic lavas (including pillow-lavas) of Slieve Gullion accumulated. During the formation of a later caldera these lavas became inset within the underlying Caledonian granodiorite. The acid layers of Slieve Gullion represent rhyolites, incandescent tuff-flows, and agglomerates, the materials of which were derived from the Caledonian granodiorite. The time of eruption of these acid rocks is correlatable with periods of caldera formation.The layered series is dissected by arcuate faults, and contacts previously thought to be intrusive are now demonstrated to be faulted. Transgressive granophyres follow the faults and form intrusion-breccias and net-vein systems with the layered series. Evidence is presented for concluding that these granophyres originated as glassy acid tuffisite—an intrusive pyroclastic rock (Cloos, 1941)—derived from the basement Caledonian granodiorite, and emplaced by rising gas-streams. That the gas sometimes had a temperature high enough to fuse basalt is witnessed by the fact that the basaltic rocks commonly become exceedingly fine grained where they adjoin transgressive granophyre. Such fine-grained edges are cross-cutting to the layers and cannot therefore be interpreted as normal chilled edges. Hybrid rocks, e.g. marscoite, are associated with the transgressive granophyres. They are shown to represent mechanical mixtures of acid tuffisite, carried upwards by rising gas-streams, with fragmented (tuffisitised) and partially fused basic material derived from the walls of the channels. The transgressive granophyres and associated hybrids are related in origin to incandescent tuff-flows, and are regarded as marking the routes by which such flows reached the surface.The metamorphism of the basaltic flows, and the transformation of the acid layers and transgressive acid tuffisites to granophyre, were probably accomplished during the pneumatolytic and hydrothermal phase of volcanic activity. Transformation of the acid rocks to granophyre involved introduction of K and Si, and removal of Na, Al, Fe, Ca, Ti, P and Mn. Similar chemical interchanges transformed basaltic rocks at Foughill and Carrickcarnan to quartz-dolerites. The subtracted materials have been located in small-scale basic fronts margining granophyres; within masses of gabbro that sank within gas-tuff streams from higher levels; and in dispersed form within the basic rocks of Slieve Gullion.