Chemical and mineralogical similarity between island-arc mantle (IAM) peridotites from Kamchatka and Philippines and cratonic harzburgite xenoliths indicates a potential genetic relationship. Arc peridotites have low modal clinopyroxene content and refractory olivine, orthopyroxene and spinel compositions (olivine (F091.93), spinel -Cr/(Cr+Al) of 0.65 to 0.81, and A1203 in opx 0.4-1.6 wt. %) (Figure 1). The modal opx content ranges from 10 to

The composition and geochemical signatures of the mantle wedge beneath the Lesser Antilles arc are documented by the ultramafic xenoliths included in alkali basalts (M-series) on Grenada. Xenoliths consist of harzburgites, lherzolites, dunites and subordinate wehrlites and pyroxenites. Primary minerals phases are olivine, low-Al and high-Al orthopyroxene, clinopyroxene and Cr-Spinel. In addition to the primary assemblage, Grenada xenoliths contain metasomatic phases such as Al-rich clinopyroxene, plagioclase, Al-rich spinel, pargasitic amphibole and Si- and Al-rich glasses. The trace-element signatures of pyroxenes and glasses have been determined on selected samples by LA-ICP-MS. Pyroxenes from both lherzolite and harzburgite xenoliths have U-shaped rare earth element (REE) profiles, unusually high Th, U and Sr concentrations and large negative Nb, Ta and Zr, and Hf anomalies. The geochemical signatures of metasomatic clinopyroxene are different from those reported for clinopyroxene from fluid-metasomatised mantle wedge, and are clearly distinct from those of clinopyroxene in equilibrium with host lavas. Si-rich glasses show a narrow compositional range, with trace-element characteristics closely similar to those of reacted pyroxenes. This, along with the general lack of chemical gradients of LILE and LREE over more compatible elements suggests dacitic glasses represent the products of in-situ melting caused by temperature increase before and during the uptake of xenoliths by host lavas. Dacitic melts are believed to represent local re-melts of regions metasomatically enriched by earlier arc magmas that had stalled, fractionated, and solidified in the upper mantle. These local re-melts thus reflect the metasomatic component formed by earlier arc-related metasomatic agents and liable to be re-mobilised. This also appears to be the easiest way to explain the compositional similarities between erupted arc lavas and the metasomatised peridotites.The results of this study suggest that the mantle wedge beneath the Lesser Antilles underwent complex peridotite–melt reaction processes operated by sub-arc melts and, later on, by magmas similar in compositions to the host alkali basalts. The majority of the compositional range of erupted Grenada magmas, but adakites found at surface, can be obtained by the interaction of basalts, possibly formed by hydrous melting of MORB-source mantle, with the overlying mantle wedge.

Pliocene to recent volcanic rocks from the Bulusan volcanic complex in the southern part of the Bicol arc (Philippines) exhibit a wide compositional range (medium- to high-K basaltic-andesites, andesites and a dacite/rhyolite suite), but are characterised by large ion lithophile element enrichments and HFS element depletions typical of subduction-related rocks. Field, petrographic and geochemical data indicate that the more silicic syn- and post-caldera magmas have been influenced by intracrustal processes such as magma mixing and fractional crystallisation. However, the available data indicate that the Bicol rocks as a group exhibit relatively lower and less variable 87Sr/86Sr ratios (0.7036–0.7039) compared with many of the other subduction-related volcanics from the Philippine archipelago. The Pb isotope ratios of the Bicol volcanics appear to be unlike those of other Philippine arc segments. They typically plot within and below the data field for the Philippine Sea Basin on 207Pb/204Pb versus 206Pb/204Pb and 208Pb/204Pb versus 206Pb/204Pb diagrams, implying a pre-subduction mantle wedge similar to that sampled by the Palau Kyushu Ridge, east of the Philippine Trench. 143Nd/144Nd ratios are moderately variable (0.51285–0.51300). Low silica (<55 wt%) samples that have lower 143Nd/144Nd tend to have high Th/Nd, high Th/Nb, and moderately low Ce/Ce* ratios. Unlike some other arc segments in the Philippines (e.g. the Babuyan-Taiwan segment), there is little evidence for the involvement of subducted terrigenous sediment. Instead, the moderately low 143Nd/144Nd ratios in some of the Bicol volcanics may result from subduction of pelagic sediment (low Ce/Ce*, high Th/Nd, and high Th/Nb) and its incorporation into the mantle wedge via a slab-derived partial melt.

Adakites were proposed over a decade ago to be products of the melting of young subducted oceanic crust. In fact, several new localities have been discovered since the original work documented approximately ten localities in modern arcs (e. g., southwestern Japan, Trans Mexican Volcanic Belt, etc.). But work over the past ten years has also shown that adakites can be generated by other processes during subduction (e. g., along the edge of tears in the subducting slab, remnant slabs left in the upper mantle, etc.). In addition, adakites appear to be associated with a suite of rocks including high-Mg andesites resulting from either adakite interaction with the mantle (Adak-type) or melting of the mantle during adakite interaction (Piip-type), niobium enriched arc basalts (NEAB) that are believed to be derived from the partial melting of a mantle metasomatized extensively by adakites, and possibly boninites (several researchers have found an adakite component in boninites). A new rock suite, the adakite metasomatic volcanic series, has been proposed to account for the various associations. In addition, a large number of NEAB have been found to contain ultramafic mantle xenoliths with clear evidence of reaction between ultradepleted mantle and adakites.Several alternative hypotheses have been proposed for the generation of adakites primarily involving the melting of the lower crust rather than the subducting slab. One model proposes that the melting of the lower crust occurs when basaltic melts underplate the lower crust. There are many reasons that appear to rule this method out. The other model proposes that in areas where the continental crust is thick, the lower crust can become eclogitic and separate and sink into the mantle (delaminate). This delamination process will bring the lower sections of the lower crust or the upper part of the delaminated lower crust in contact with relatively hot mantle, which could initiate melting, and the production of adakites. This has led us to conclude that many of the Cretaceous adakites not associated with subduction found in East China are the result of lower crustal melting via delamination. We also emphasize that, if true, the term adakite should not be restricted to processes related only to slab melting but must include those involving the melting of the lower crust.The Archean continental crust consists of primarily trondhjemites, tonalities, and dacites (TTD). It remains problematic as to whether this continental crust was derived from slab melting or lower crustal melting. However, we believe that the higher mantle temperatures during the Archean led to more mid ocean ridges, which generated the subduction of "more" younger crust than today. Based on this, we suggest that the Archean TTD continental crust was generated primarily by slab melting. We also note that the Archean was a period of extensive gold mineralization. Several researchers have also found gold and copper mineralization associated with the adakite metasomatic volcanic series. This series may be an important indicator of ore deposits.

We are grateful for this opportunity to elaborate upon our ideas. R. Conrey has conveniently numbered his thoughts, and therefore, we will address them using his order. But first a few remarks about his general comments.We have never implied that slab melting is a “common process” and that many arc volcanoes are composed “dominantly of adakite.” In fact, we emphasized that Drummond and Defant [1990] were correct in their interpretation that adakite production has vastly decreased since the Archean because less young crust subducts today. Although there are about 10–15 current adakite‐producing regions, this is certainly not a great deal compared with the extent of “regular” arc volcanism on the planet.

To the best of our knowledge, modern adakites have not been documented in a nonarc environment. We report geochemical and isotopic data for Early Cretaceous Anjishan adakitic intrusive rocks that are in a continental setting unrelated to subduction. The Anjishan adakitic intrusive rocks, which are exposed in the Ningzhen area of east China, have high Sr/Y and La/Yb ratios coupled with low Yb and Y as well as relatively high MgO contents and Mg numbers (Mg#; 0.4-4.6), similar to products from slab melting. However, low epsilon(Nd(t)) values (-6.8 to-9.7) and high (Sr-87/Sr-86)(i) (0.7053-0.7066) are inconsistent with an origin by slab melting. The tectonics and geochemistry lead us to conclude that adakitic magmas were most likely derived from partial melting of mafic material at the base of the continental crust. High Sr/Y and La/Yb ratios of the adakitic intrusive rocks suggest that garnet was stable as a residual phase during partial melting, implying that the crustal thickness exceeded 40 km in the Early Cretaceous. The present thickness of the crust in the Ningzhen area is only 30 km, and therefore the crust appears to have been thinned by at least similar to10 km since the Early Cretaceous. The relatively high MgO contents and Mg# of the Anjishan intrusive rocks suggest that adakitic magmas interacted with mantle rocks, possibly coinciding with lower-crustal delamination, which would also account for the observed thinning.

Ultramafic xenoliths from a veined mantle wedge beneath the Kamchatka arc have non-chondritic, fractionated chondrite-normalized platinum-group element (PGE) patterns. Depleted (e.g., low bulk-rock Al2O3 and CaO contents) mantle harzburgites show clear enrichment in the Pd group relative to the Ir group PGEs and, in most samples, Pt relative to Rh and Pd. These PGE signatures most likely reflect multi-stage melting which selectively concentrates Pt in Pt–Fe alloys while strongly depleting the sub-arc mantle wedge in incompatible elements. Elevated gold concentrations and enrichment of strongly incompatible enrichment (e.g., Ba and Th) in some harzburgites suggest a late-stage metasomatism by slab-derived, saline hydrous fluids. Positive Pt, Pd, and Au anomalies coupled with Ir depletions in heavily metasomatized pyroxenite xenoliths probably reflect the relative mobility of the Pd and Ir groups (especially Os) during sub-arc metasomatism which is consistent with Os systematics in arc mantle nodules. Positive correlations between Pt, Pd, and Au and various incompatible elements (Hf, U, Ta, and Sr) also suggest that both slab-derived hydrous fluids and siliceous melts were involved in the sub-arc mantle metasomatism beneath the Kamchatka arc.

Saba is the northernmost volcano along the Lesser Antilles island-are chain, The Lesser Antilles are results from the west-northwest subduction of the Atlantic lithosphere beneath the Caribbean Plate, Sediment thickness along the trench decreases northward away from sediment sources on the continent of South America. We focused our attention on Saba precisely because it is the furthest away from documented geochemical effects in the southern are volcanics of the large sediment thicknesses normally attributed to both source or upper level contamination (i.e. assimilation).Field mapping, petrology, mineralogy, K-Ar dating, and geochemical analyses (major and trace element) indicate a complex history of magma petrogenesis including crystal fractionation, magma mixing, and, surprisingly, crustal assimilation. This is the first time assimilation has been documented in the northern section of the Lesser Antilles are, Magma mixing shows up in the field as banded pumice and petrographically and mineralogically as complex zoning in phenocrysts (such as reverse zoning in plagioclase), disequilibrium mineral assemblages (e.g. quartz and olivine), and disequilibrium between minerals and whole-rock compositions (e,g. forsterite content of olivine). Mass-balance modeling of major and trace elements support our contention that crystal fractionation (including amphibole) played an important role in magma evolution. However, various geochemical trends can only be explained by assimilation-fractional crystallization based on the fact that the trends of various trace elements and trace-element ratios vary with increasing silica, Finally, we could find no evidence of sediment source contamination in the most mafic rocks. It may exist but is overprinted by the later assimilation effects. (C) 2001 Elsevier Science B.V. All rights reserved.

Mantle-derived spinet harzburgites from the Kamchatka arc have fractionated Pd-group element patterns, and Pt is clearly enriched relative to I'd and Rh. This fractionation is also consistent with chondritic Ir-group-platinum-group element distribution and superchondritic Pt/Pd, chondrite-normalized (Pt/Os)(N) and (Pt/Ir)(N) found in harzburgite xenoliths from the Tubaf seamount in the Lihir island group of the Tabar-Lihir-Tanga-Feni island arc in Papua New Guinea. The nonchondritic Pt/Pd ratios and Pt enrichment in the island-are mantle are best explained by extraction of melt in a back-are volcanic-arc setting, which left refractory Pt-Fe alloys in the residual mantle. Pt-Fe alloys selectively concentrate Pt relative to Pd and, overall, increase the Pt/Pd in a subarc mantle wedge. Partial melting of this Pt-enriched mantle wedge above the subduction zone, due to hydrous flux from the subducting lithosphere, is potentially capable of producing Pt-rich primitive melts parental to the Alaskan-type ultramafic-mafic complexes.

Most recent geology textbooks state that subduction‐related volcanism is due to the melting of the down‐going lithosphere. However, for the last 30 years, few in the field have seriously believed that the subducting slab is the source of arc basalts. The accepted hypothesis involves melting of the mantle wedge above the slab via hydrous fluids produced during the transition of the subducting basalt from amphibolite to eclogite. The parental basalts differentiate primarily through crystal fractionation, magma mixing, and differentiation at the Mohorovicic discontinuity into andesites and dacites as they ascend; the basalts are too dense to rise through the lower continental crust. This explains the relative abundance of differentiated rocks in arcs.

The systematics of fluid-mobile trace elements in are lavas from Panama, relative to their Li isotopic compositions, provide unique evidence for the fertilization and subsequent differential extraction of mobile species from the subarc mantle. Calc-alkaline lavas that crystallized between 20 and 5 Ma (Old Group) that possess delta(7)Li as high as +11.2 have low B/Be. Otherwise identical (and similarly old) calc-alkaline lavas with high B/Be (to 23), have mid-ocean ridge basalt (MORB) like delta(7)Li (+4.7 to +5.6). Adakite lavas (<3 Ma; Young Group) possess delta(7)Li from +1.4 to +4.2 and have consistently lower B/Be than Old Group lavas, consistent with derivation from melting of a devolatilized MORB slab. If Li and B had comparable fluid mobility in the subarc mantle, then slab fluids would carry both high B concentrations and elevated delta(7)Li signatures into are sources, and samples with the highest delta(7)Li would also have the highest B/Be. Our data suggest that although bath Li and B are initially derived from the slab, older delta(7)Li signatures may be preserved in the mantle beneath arts. As a result, regions of the lithospheric mantle will develop Li isotope signatures that are heavier than typical MORB mantle.

Neogene and Quaternary lavas from Batan, Babuyan de Clavo, Camiguin and Calayan islands (northern Luzon arc) display temporal increases in incompatible elements including Cs, Rb, Ba, K, La, Ce, Th, U, Ta, Hf and Zr from volcanoes older than 3 Ma to younger ones. These enrichments occur either within a single island (Batan) or within an island group (from Calayan to Camiguin and Babuyan). We show that these enrichments result from incompatible element input into the mantle wedge rather than from partial melting or fractionation effects. The fact that highly incompatible elements display temporal enrichment patterns in Batan lavas whatever their chemical properties indicates that hydrous fluids are not the only metasomatic agents operating in the mantle wedge and that slab-derived melts (adakitic magmas) may also be involved. The coupled temporal variation patterns of large ion lithophile elements and Sr-Nd isotopes su est that the metasomatic budgets beneath the southern group of islands are mainly controlled by hydrous fluid inputs. In contrast, young Batan lavas likely derive from a mantle source mostly metasomatized by adakitic magmas.

The Kamchatka arc (Russia) is located in the northwestern Pacific Ocean and is divided into three segments by major sub-latitudinal fault zones (crustal discontinuities). The southern (SS) and central (CS) segments are associated with the subduction of old Pacific lithosphere, whereas the northern, inactive segment (NS) was formed during westward subduction of young (< 15 Ma) Komandorsky Basin oceanic crust. Further segmentation of the arc is outlined by the development of the Central Kamchatka Depression (CKD) intra-arc rift, which is oriented parallel to the arc and is splitting the CS into the active Eastern Volcanic Front (EVF) and the largely inactive, rear-arc Sredinny Range. The NS volcanics (15-5 Ma) include calc-alkaline lavas, shoshonites, adakites, and Nb-enriched arc basalts. Isotopically all magma types share high 143Nd/144Nd ratios of 0.512976-0.513173 coupled with variable 87Sr/86Sr (0.702610-0.70356). NS lavas plot within or slightly above the Pacific MORB field on the Pb isotopic diagrams. The EVF volcanoes have more radiogenic 143Nd/144Nd (0.51282-0.513139) and 208Pb/204Pb (38.011–38.1310) than the NS lavas. CKD lavas display MORB-like Nd isotope ratios at slightly elevated 87Sr/86Sr values accompanied by a slightly less radiogenic Pb composition. Kamchatka lavas are thought to be derived from a MORB-like depleted source modified by slab-derived siliceous melts (adakites) and fluids (NS), or fluids alone (CS and SS). The NS and EVF lavas may have been contaminated by small fractions of a sedimentary component that isotopically resembles North Pacific sediment. Petrogenesis in the Kamchatka arc is best explained by a three-component model with depleted mantle wedge component modified by two slab components. Slab-derived hydrous melts produced incompatible element characteristics associated with northern segment lavas, while hydrous slab fluids caused melting in the depleted mantle below the southern and central segments of the Kamchatka arc. Trace element characteristics of Kamchatka lavas appear to be controlled by slab fluids or melts, while radiogenic isotope ratios which are uniform throughout the arc reflect depleted composition of sub-arc mantle wedge.