Identification of excess 40Ar by the 40Ar/39Ar age spectrum technique

⁴⁰Ar/³⁹Ar incremental heating experiments on igneous plagioclase, biotite, and pyroxene that contain known amounts of excess⁴⁰Ar indicate that saddle-shaped age spectra are diagnostic of excess⁴⁰Ar in igneous minerals as well as in igneous rocks. The minima in the age spectra approach but do not reach the crystallization age. Neither the age spectrum diagram nor the⁴⁰Ar/³⁶Ar versus³⁹Ar/³⁶Ar isochron diagram reliably reveal the crystallization age in such samples.     

Investigation of excess argon in ultramafic rocks from the Kola Peninsula by the40Ar/39Ar method

In several xenolithic ultramafic rocks from the Kola Peninsula, including a magnetic separate, abnormally high⁴⁰Ar/³⁹Ar ratios persisted at low and high temperatures. The lowest⁴⁰Ar/³⁹Ar ratio was consistently observed at intermediate temperatures (900–1100°C), indicating an apparent age of 2.8–3.1 b.y.; however, this may not indicate the formation age.The quantity of excess⁴⁰Ar was estimated at each temperature fraction, adopting ages inferred from published Rb-Sr ages or the minimum⁴⁰Ar/³⁹Ar age. Excess⁴⁰Ar is abundantly trapped both in mineral lattices and nonretentive trapping sites, but the trapping sites are different from those of in-situ radiogenic⁴⁰Ar. The high temperature component of excess⁴⁰Ar is considered to represent Ar dissolved during mineral formation in the upper mantle or the lower crust.A correlation between the amount of high temperature excess⁴⁰Ar and³⁶Ar exists for some samples. The⁴⁰Ar_excess/³⁶Ar ratios of the rocks of probable upper mantle or lower crust origin vary from about 10 000 to 35 000, which may suggest large fluctuations of this ratio in the deep interior of the earth. The high value implies that most³⁶Ar was already degassed from the earth's interior at least 2 or 3 b.y. ago.     

40Ar/39Ar dating on volcanic rocks of the Deccan Traps,India

⁴⁰Ar/³⁹Ar dating results on seven volcanic rocks from four areas of the Deccan Traps, India, suggest that volcanic activity more than 70 Ma ago might have occurred at least in limited areas.In the Igat Puri area, the uppermost flow shows an⁴⁰Ar/³⁹Ar age of 63 Ma, whereas a lower flow has an age of around 82–84 Ma.⁴⁰Ar/³⁹Ar ages of samples from the Bombay area also seem to favor the occurrence of volcanic activity more than 70 Ma ago. One rhyolite dyke from the Osam Hill in the Girnar Hill area shows a well-defined plateau age of 68 Ma, whereas two tholeiitic basalts from the Mahabaleshwar area indicate a total⁴⁰Ar/³⁹Ar age of around 63–64 Ma, though they show the effect of secondary disturbance in the age spectra.The volcanic activity(ies) more than 70 Ma ago may correspond to precursory one(s) for the main volcanic activity around 65 Ma ago in the Deccan Traps.     

Excess40Ar in metamorphic rocks from Broken Hill,New South Wales: implications for40Ar/39Ar age spectra and the thermal history of the region

⁴⁰Ar/³⁹Ar age spectrum analyses of samples from Broken Hill, New South Wales, indicate that the region has experienced a complex thermal history following high-grade metamorphism, 1660 Ma ago. The terrain cooled slowly (～3°C Ma^?1) until about 1570 Ma ago, when the temperature fell below about 500°C. Following granitoid emplacement ～1500 Ma ago, the region remained relatively cold until affected by a thermal pulse 520±40Ma ago, causing temperatures to rise to～350°C in some places. During this event, accumulated⁴⁰Ar was released from minerals causing a significant Ar partial pressure to develop. Laboratory Ar solubility data combined with the⁴⁰Ar/³⁹Ar age spectra gives a local estimate of this partial pressure of ～10^?4atm. The region finally cooled below 100°C about 280 Ma ago.⁴⁰Ar/³⁹Ar age spectrum analyses of hornblende, plagioclase and clinopyroxene containing excess⁴⁰Ar are characterized by saddle-shaped age spectra. Detailed analysis of plagioclase samples reveals a complex diffusion behaviour, which is controlled by exsolution structures. This effect, in conjunction with the presumed different lattice occupancy of excess⁴⁰Ar with respect to radiogenic⁴⁰Ar, appears to be responsible for the saddle-shaped age spectra.     

40Ar/39Ar and U-Th-Pb dating of separated clasts from the Abee E4 chondrite

Determinations of⁴⁰Ar/³⁹Ar and U-Th-Pb are reported for three clasts from the Abee (E4) enstatite chondrite, which has been the object of extensive consortium investigations. The clasts give⁴⁰Ar/³⁹Ar plateau ages and/or maximum ages of 4.5 Gy, whereas two of the clasts give average ages of 4.4 Gy. Within the range of 4.4–4.5 Gy these data do not resolve any possible age differences among the three clasts.²⁰⁶Pb measured in these clasts is only ～1.5–2.5% radiogenic, which leads to relatively large uncertainties in the Pb isochron age and in the²⁰⁷Pb/²⁰⁶Pb model ages. The Pb data indicate that the initial²⁰⁷Pb/²⁰⁶Pb was no more than 0.08±0.07% higher than this ratio in Can?on Diablo troilite. The U-Th-Pb data are consistent with the interpretation that initial formation of these clasts occurred 4.58 Gy ago and that the clasts have since remained closed systems, but are contaminated with terrestrial Pb. The⁴⁰Ar/³⁹Ar ages could be gas retention ages after clast formation or impact degassing ages. The thermal history of Abee deduced from Ar data appears consistent with that deduced from magnetic data, and suggests that various Abee components experienced separate histories until brecciation no later than 4.4 Gy ago, and experienced no appreciable subsequent heating.     

40Ar/39Ar incremental-release ages of biotite from a progressively remetamorphosed Archean basement terrane in southwestern Labrador

Gneisses within an Archean basement terrane adjacent to the southwestern portion of the Labrador Trough were variably retrograded during a regional metamorphism of Grenville age (ca. 1000 Ma). Biotites from non-retrograded segments of the gneiss terrane record⁴⁰Ar/³⁹Ar plateau and isochron ages which date times of cooling following an episode of the Kenoran orogeny (2376–2391 Ma). A suite of gneiss samples displaying varying degrees of retrograde alteration was collected across the Grenville metamorphic gradient. Biotites in these samples show no petrographic evidence of retrograde alteration, however they do record internally discordant⁴⁰Ar/³⁹Ar age spectra. Although the extent of internal discordance is variable, the overall character of the release patterns is similar with younger apparent ages recorded in intermediate-temperature gas fractions. The total-gas dates range from 2257±27 Ma (northwest) to 1751±23 Ma (southeast), suggesting that variable quantities of radiogenic argon were lost from the Archean biotites during Grenville metamorphism. The “saddle-shaped” nature of the discordant spectra indicates that argon loss was not accomplished through single-stage, volume diffusion processes.Biotites in portions of the gneiss terrane which were completely recrystallized during Grenville metamorphism are petrographically and texturally distinct. A representative of this phase records a⁴⁰Ar/³⁹Ar plateau age of 2674±28 Ma. This date is markedly inconsistent with regional constraints on the timing of Grenville metamorphism, and indicates the presence of extraneous argon components. Both the extraneous and radiogenic argon components must have been liberated in constant proportions during experimental heating because the argon isotopic data yield a well-defined⁴⁰Ar/³⁶Ar vs.³⁹Ar/³⁶Ar isochron corresponding to an age (2658±23 Ma) similar to that defined by the plateau portion of the spectrum.The⁴⁰Ar/³⁹Ar biotite dates suggest that the effects of Grenville metamorphism extent 15–20 km northward into the Superior Province. The limit of this overprint is approximately coincident with the northernmost development of Grenville age thrust faults in the Archean terrane. Therefore, it is proposed that the northern margin of the Grenville Province in southwestern Labrador should be located along the northernmost Grenville thrust fault because this represents both a structural and a thermal discontinuity.     

40Ar/39Ar age and thermal history of the Kirin chondrite

The Kirin meteorite, a large (2800kg) H5 chondrite, fell in Kirin Province, China in 1976. A sample from each of the two largest fragments (K-1, K-2) yield⁴⁰Ar/³⁹Ar total fusion ages of 3.63 ± 0.02b.y. and 2.78 ± 0.02b.y. respectively.⁴⁰Ar/³⁹Ar age spectra show typical diffusional argon loss profiles. Maximum apparent ages of 4.36 b.y. (K-1) and ～4.0 b.y. (K-2) are interpreted as possible minimum estimates for the age of crystallization of the parent body.The⁴⁰Ar/³⁹Ar ages found for gas released at low temperature are about 2.2 b.y. for K-1 and about 0.5 b.y. for K-2, suggesting that this meteorite may have suffered two discrete collisional events that caused degassing of radiogenic argon. Modelling of possible thermal events in the parent body indicates that samples K-1 and K-2 were at a depth of less than 3 m from the base of an impact melt of a thickness less than 7 m and separated by no more than ～2 m from one another at the time of the heating event about 0.5 b.y. ago. Further, the duration of heating was probably less than a few years.Calculations from³⁸Ar data yield exposure ages for samples K-1 and K-2 of about 5 m.y., similar to that found for many other H chondrites.     

Age studies on slates: Applicability of the40Ar/39Ar stepwise outgassing method

Results of⁴⁰Ar/³⁹Ar stepwise outgassing experiments are reported for six slates, one hornfels and one metasilt-stone from the Lower Paleozoic Meguma Group of Nova Scotia, Canada. The age spectrum of the hornfels and of one slate collected from the vicinity of a granite contact have both been completely reset by the thermal effects of the intrusion. The latter, however, has not produced any recognizable mineralogical modification of the slate. Two samples in the collection contain potassium feldspar, possibly of detrital origin, and both have yielded discordant age spectra which cannot be unambiguously interpreted.Three apparently “clean” slates collected from localities well removed from the effects of the granite batholith have yielded quite well-defined age plateaus at ?415 Ma (revised decay and isotopic constants). The mean total gas⁴⁰Ar/³⁹Ar age or, equivalently, the mean K-Ar age is ?400 Ma. From these data we suggest a minimum value ?400–415 Ma for the time of initiation of the Acadian Orogeny in Nova Scotia which culminated with post-deformational “granitic” intrusion ?370–380 Ma ago.     

The evolution of excess argon in alpine biotites — A40Ar-39Ar analysis

Alpine biotites containing excess⁴⁰Ar have been analysed by step-heating argon analysis of both neutron irradiated and unirradiated samples. In addition to age spectra the data are discussed in terms of the thermal release of⁴⁰Ar,³⁹Ar,³⁷Ar and³⁶Ar and also displayed on a correlation plot of³⁶Ar/⁴⁰Ar vs.³⁹Ar/⁴⁰Ar which is used to interpret the data and present a model of isotopic evolution during metamorphic cooling. This diagram overcomes misleading complications of isochron plots. The samples exhibit the following argon systematics: (1) flat age spectra for 80–90%³⁹Ar release with anomalously old ages but early gas fractions that approximate the accepted cooling ages; (2) each sample shows decreasing³⁶Ar/⁴⁰Ar with increasing temperature of heating step with three samples having a negative correlation of³⁶Ar/⁴⁰Ar vs.³⁹Ar/⁴⁰Ar and one a positive correlation; (3) there appear to be two³⁶Ar components, one released at high temperatures and correlated with radiogenic⁴⁰Ar and one released at low temperatures which is not correlated with radiogenic⁴⁰Ar; and (4) there is no significant effect of neutron irradiation on the release of⁴⁰Ar and³⁶Ar.Interpretation suggests that these biotites contain a record of the evolution and isotopic composition of ambient argon retained within the metamorphic host rocks during cooling. After incorporation of argon of high⁴⁰Ar/³⁶Ar another argon component, of atmospheric composition, was retained at lower temperature and argon partial pressures.     

40Ar-39Ar age determinations on the Owyhee basalt of the Columbia Plateau

⁴⁰Ar/³⁹Ar step-heating analyses have been performed on 11 samples of basalt from sites near Owyhee Reservoir of southeastern Oregon, U.S.A. These rocks were extruded during the great flood basalt episode of the Pacific Northwest. The whole-rock points are highly correlated on a plot of⁴⁰Ar/³⁶Ar versus³⁹Ar/³⁶Ar, corresponding to a common age of the samples of 14.3 ± 0.3 m.y. Inspite of this, individual “plateau” plots of the age versus fraction of³⁹Ar released do not give good plateaux. These age spectra exhibit to varying degrees a common structure in which lower age values are found at higher temperatures. This pattern may result from a closed-system redistribution of the argon isotopes. The usefulness of grinding the basalts in removing a loosely held atmospheric argon component is confirmed.     

40Ar/39Ar ages for the alkaline volcanism and the basement of Gorringe Bank,North Atlantic ocean

Gorringe Bank is situated on the Europe-Africa plate boundary at the eastern end of the Azores-Gibraltar fracture zone. It has two summits, Gettysburg Bank to the Southwest and Ormonde Bank to the northeast.We applied the⁴⁰Ar/³⁹Ar stepwise heating method to date six samples of the alkaline volcanic rocks, two gabbros from the Ormonde Bank and a dolerite from the Gettysburg Bank. The results that the alkaline volcanism lasted probably for less than 6 Ma(66-60 Ma).Although the nature of this volcanism precludes any subduction feature during its setting, the alkaline volcanism of Ormonde is probably linked to Upper Cretaceous/Eocene compressive tectonic events.The basement rocks of Gorringe Bank reveal distrubed⁴⁰Ar/³⁹Ar age spectra. One plagioclase and one biotite from a gabbro give evidence for a thermic event whose age is tentatively estimated at about 75 Ma, and related to a variation in the direction of the relative movement between Europe and Africa. The more probable age given by a plagioclase of another gabbro and by a dolerite (110 Ma) corresponds to tilting northeastward of the Gorringe massif.     

40Ar-39Ar and Rb-Sr age determinations on Quaternary volcanic rocks

⁴⁰Ar-³⁹Ar and Rb-Sr ages have been measured on separated minerals from the potassic volcanics of the Roman Comagmatic Region to test the ability of these methods to accurately data Quaternary geological events. The very high K and Rb contents of the Roman magmas present particularly favorable situations in which the very high concentrations of the radioactive nuclides⁴⁰K and⁸⁷Rb result in well resolved in situ enrichments of the daughter isotopes despite the very young ages. Six leucite separates contained Ar with very high bulk40/36 ratios (above 1000) and in which the⁴⁰Ar and the³⁹Ar were very well correlated, yielding well-defined ages averaging3.38±0.08×10⁵ years. Two leucites contained Ar with lower bulk40/36 ratios (～400), and in at least two release steps from these leucites the⁴⁰Ar/³⁶Ar ratio was significantly lower than atmospheric. Despite the uncertainty in the composition of the trapped component, these two leucites have ages that do not differ significantly from the ages of the other leucites. For the biotites, it was not possible to obtain through stepwise degassing a good separation of in situ radiogenic⁴⁰Ar from trapped⁴⁰Ar and therefore the calculated ages are not as precise as those of the leucites. In three cases the biotite age agrees with the age of the cogenetic leucite, but in the remaining two cases discordant ages are obtained, suggesting caution when using biotites as Quaternary age indicators.Rb-Sr measurements on leucite, biotite, and pyroxene separates hand-picked from each of three tuff samples yielded a dispersion in⁸⁷Sr/⁸⁶Sr as large as 16 parts in 10⁴ and⁸⁷Rb/⁸⁶Sr as high as 218 for leucites, and permitted the determination of internal isochron ages. The ages obtained range from3.8±0.2×10⁵to3.3±0.2×10⁵ years and are in good agreement with the⁴⁰Ar-³⁹Ar ages on the leucites. The data for each tuff sample yield a well-defined uniform initial⁸⁷Sr/⁸⁶Sr. However, different tuffs show small differences in initial⁸⁷Sr/⁸⁶Sr pointing to distinct sources or to assimilation of different materials during the extrusion of the tuffs. These measurements demonstrate the possibility of dating Quaternary materials by both the⁴⁰Ar-³⁹Ar method and the Rb-Sr method. The observation of concordant ages with a precision of a few percent represents a powerful tool in Quaternary stratigraphy.     

39Ar recoil losses and presolar ages in Allende inclusions

Ar analyses are reported for five coarse-grained, Ca-Al-rich inclusions from the Allende meteorite. The samples were neutron-irradiated in individual evacuated ampoules, and the Ar gas in the ampoules as well as the samples was analyzed. A large fraction (up to 60%) of the³⁹Ar from³⁹K (n, p) reactions was lost out of the inclusions into the ampoules. The³⁹Ar losses resulted in substantial increases in the apparent⁴⁰Ar-³⁹Ar ages of the samples.³⁹Ar recoil loss during neutron-irradiation is a major effect and must be accounted for in⁴⁰Ar-³⁹Ar dating. All of the Allende inclusions studied contained substantial trapped³⁶Ar. The origin of the trapped³⁶Ar is unknown, and the possible presence of trapped⁴⁰Ar cannot be excluded. Ar measurements on Allende inclusions which have yielded anomalously old ages must be re-examined in the context of³⁹Ar recoil loss and possible contributions of trapped⁴⁰Ar. Allende inclusions appear on both accounts to be poor candidates to search for relicts of presolar materials with well-defined K/Ar ages.     

40Ar/39Ar age spectrum analysis of detrital microclines from the southern San Joaquin Basin,California: an approach to determining the thermal evolution of sedimentary basins

Detrital microcline grains from sedimentary strata preserve a record of thermal evolution in the temperature range ～ 100° to 200°C which can be revealed by⁴⁰Ar/³⁹Ar age spectrum analysis. Microcline separates from deep drill hole intersections with Eocene to Miocene sediments in the Basin and Tejon Blocks of the southern San Joaquin Valley, California, analysed by the age spectrum approach show radiogenic⁴⁰Ar (⁴⁰Ar*) gradients that record both the slow cooling of the uplifting sediment source ～ 65 Ma ago, and a recent thermal event. This information, in conjunction with the observation of fission track annealing in the coexisting apatites, allows estimation of the temperature-time conditions of this thermal event at about 140°C for ～ 200 ka. Present and paleotemperature data is in accord with heating related to several kilometers of Pleistocene sediment deposition. Heat flow calculations suggest that this recent subsidence has depressed the thermal gradient from about 30°C km^?1 to the present apparent gradient of 24°C km^?1.⁴⁰Ar/³⁹Ar analysis of detrital microcline crystals yields thermochronological information in the temperature-time range of petroleum maturation and provides this technique with potential as both a useful exploration tool and as a means of probing the fundamental geodynamic processes of basin evolution.     

Intercalibration of international and domestic 40Ar/39Ar dating standards

Four international standards, Ga1550, MMhb-1, Lp-6, Bem 4M, and one domestic standard BT-1 have been intercalibrated. The repeated measurements on MMhb-1 with different mass demonstrate that MMhb-1 is inhomogeneous in age and its average age is 519.8 Ma. The results of Bern 4M and Lp-6 reflect that they have an invariable value of ⁴⁰Ar*/³⁹Ar_k (F) and the ages we obtained are consensus with their K-Ar age: Lp-6=127.7Ma; Bern 4M=18.2 Ma. Analyses of BT-1 age spectra, Ca/K and Cl/K spectra as well as inverse isochrons indicate that the sample is homogeneous and invariable and keeps close chemically, with its trapped argon isotope composition close to the atmosphere. The dating results show that age values are reproducible and steady, total fusion age, step-heating age, plateau age and isochron age are in accord with each other within the error range (2σ). Therefore, we recommend 28.7 Ma as the calibrated age of BT-1. We also discuss the variation in neutron flux gradients of Beijing 49-2 reactor. It was found that the neutron flux gradient varies considerably, and more monitors (standard samples) are needed to fix the trend of variation. The coefficient of the 49-2 reactor that transfers the ratio of production rate of ³⁷Ar_Ca/³⁹Ar_K into Ca/K ratio is 1.78. This is different from that reported earlier, 2.0, which may be caused by the reconstruction of the reactor.     

40Ar–39Ar analysis of phlogopite in the Horoman Peridotite Complex, Hokkaido, Japan and implications for its origin

Abstract ⁴⁰Ar–³⁹Ar analysis of phlogopite separated from a plagioclase lherzolite of the Horoman Peridotite Complex, Hokkaido, Japan, has yielded a plateau age of 20.6 ± 0.5 Ma in an environment where the metamorphic fluid was characterized by an almost atmospheric Ar isotopic ratio. The age spectrum is slightly saddle-shaped, implying some incorporation of excess ⁴⁰Ar during the formation of the phlogopite at a depth. As the phlogopite has been inferred to have formed in veins and/or interstitials during exhumation of the peridotite body, metasomatic fluids, to which ground- and sea water might have contributed, were probably involved in the formation of phlogopite in the crustal environment. A total ⁴⁰Ar–³⁹Ar age of 129 Ma of a whole rock sample of the plagioclase lherzolite, from which the phlogopite was separated and is representative of the main lithology of the Horoman Peridotite Complex, indicates the occurrence of excess ⁴⁰Ar. Hence, the age has no geological meaning.     

Studies on40Ar/39Ar thermochronology of strike-slip time of the Tan-Lu fault zone and their tectonic implications

Samples of mylonite, ultramylonite and phyllonite were collected from 5 localities in the Anhui part of the Tan-Lu fault zone for⁴⁰Ar/³⁹Ar chronological studies. Among them 4 samples from 3 localities on the eastern margin of the Dabie orogenic belt yielded⁴⁰Ar/³⁹Ar plateau ages of 128 —132 Ma; and 2 samples from the western margin of the Zhangbalin uplift and eastern margin of the Bengbu uplift gave the same⁴⁰Ar/³⁹Ar plateau ages of 120 Ma. Isochron analyses and other lines of evidence suggest that the data are reliable. The data are interpreted as cooling ages of sinistral strike-slip deformation of the Tan-Lu fault zone. The younger ages from the north might be related to slower strike-slip rising. These results indicate that the large-scale left-lateral displacement in the Tan-Lu fault zone took place in the Early Cretaceous, rather than in Late Triassic (Indosinian) as proposed by some geologists. Therefore, this fault zone is an intracontinental wrench fault rather than a transform fault or suture line developed during formation of the Dabie orogenic belt.     

Radiometric ages from Ryukyu arc region and an40Ar/39Ar age from biotite dacite on Okinawa

A biotite dacite that intrudes metamorphic rocks on Okinawa in the Ryukyu island arc has been dated at 12 m.y. by the⁴⁰Ar/³⁹Ar method. The details of this age measurement and a compilation of radiometric ages for the Ryukyu island arc and adjacent regions are presented. These data suggest that from 65 to 12 m.y. ago the magmatic axis of the Ryukyu arc was confined to a very narrow zone along the arc. In Kyushu and Shikoku, the southern Japanese islands, intrusive and volcanic igneous rocks dated as 21?12 m.y. occur over a much wider zone than in the Ryukyu arc. The apparent difference in width of the magmatic zones may be due to different absolute motions of the overthrust plates of those two regions of subduction. The dissimilarity of available radiometric ages for the Ryukyu arc and for Taiwan suggest different histories for the development of these two features. The occurrence of active volcanoes in association with the Okinawa Trough, northwest of the Ryukyu island arc, may indicate that the trough itself developed in the last 12 m.y.     

39Ar-40Ar systematics of two millimeter-sized rock fragments from Mare Crisium

Two small fragments, L24B, a glass-rich agglutinate (1.9 mg) and L24A, a fine-grained lithic fragment (9.4 mg), from the Luna 24 landing site have been neutron irradiated for the purpose of³⁹Ar-⁴⁰Ar dating. A fairly well-defined³⁹Ar-⁴⁰Ar plateau age of 3.65 ± 0.12 AE was found for the larger fragment. After appropriate corrections the composition of the trapped and spallogenic Ar could be deciphered. The evolution of³⁸Ar_sp/³⁷Ar showed that 660 m.y. and 500 m.y. were the most reliable exposure ages for L24A and L24B, respectively. The Ti contents of ≤0.6% determined by gamma-counting prior to the Ar analysis indicate both fragments being associated with the group of low-Ti or even very low-Ti basalts.     

39Ar-40Ar ages of samples from the Apollo 17 station 7 boulder and implications for its formation

³⁹Ar-⁴⁰Ar ages and³⁷Ar-³⁸Ar exposure ages of samples representing four different lithologies of the Apollo 17 station 7 boulder were measured. The age of the dark veinlet material77015of3.98 ± 0.04AE is interpreted as representing the time of intrusion of this veinlet into the 77215 clast. The data obtained so far indicate that the vesicular basalt 77135 formed 100–200 m.y. later. However, this has to be confirmed by³⁹Ar-⁴⁰Ar investigations on separated mineral and/or grain-size fractions. A small clast enclosed in the 77135 basalt gives a well-defined high temperature age of3.99 ± 0.02AE. A sample of the noritic clast 77215 gave4.04 ± 0.03AE, the highest age found so far in this boulder. The³⁹Ar-⁴⁰Ar ages obtained are in agreement with the age relationships deduced from the stratigraphic evidence.Taking into account the shielding by the boulder itself, an average³⁷Ar-³⁸Ar exposure age of(27.5 ± 2.5)m.y. is obtained for the samples collected from the boulder.