The alteration state of basalt lavas is shown to be largely the result of the combined action of two alteration processes, deuteric oxidation and regional hydrothermal alteration. The former process is well described in the literature while the latter is shown here to be another widespread source of lava alteration. The environment of regional hydrothermal alteration is that of the zeolite metamorphic facies, with temperatures of up to 300°C, and abundant groundwater being the major elements. Independent and rapid spatial variation of the two alteration processes results in the wide range of observed lava properties. Microscopically undetectable titanomagnetite alteration, indicated by steady rise in strong field Curie point, is the first response of a non-deuterically oxidized basalt to regional hydrothermal alteration. With more extreme alteration, titanohematite etc. is seen to pseudomorph titanomagnetite grains. Ilmenite also shows very distinctive mineralogical changes. Initially, high deuteric oxidation specimens appear to be immune magnetically to moderate alteration but these too eventually succumb when conditions are extreme enough for the formation of prehnite in the rock.The geophysical implications of this widespread post eruption lava alteration process is discussed. We conclude that it is probably possible to obtain information on the original TRM of basalts at least in many cases but that the implications for potassium-argon dating need serious investigation.

No abstract

SuriiniaryThe iron-titanium oxides of 45 samples from a single tertiary lava flow from Skye, Scotland, have been subjected to detailed examination and quantitative measurement.A basically simple assemblage of titanomagnetite (93-100 per cent) and discrete ilmenite (0-7 per cent) has, in most samples experienced one or both of two types of alteration, titanomagnetite high-temperature oxidation and ' titanomagnetite granulation ', which we define in this paper.Titanomagnetite high-temperature oxidation, describing the process of ilmenite exsolution and further oxidation, varies rapidly but largely systematically within the flow. Parallel alteration of discrete ilmenite and olivine also occurs.Titanomagnetite granulation, describing the alteration of titanomagnetite to impure rutile granules, only affects samples with limited development of high-temperature oxidation and varies throughout the lava in a rather unsystematic fashion. While high-temperature oxidation is almost certainly wholly of deuteric origin titanomagnetite granulation is shown usually to require burial to greater than 900 m for its formation, and thus may not take place until several million years after the extrusion of a lava flow. The mechanisms by which these two types of alteration take place, and their implications for palaeomagnetic interpretation, are discussed.

No abstract

Working out the history of the Earth's magnetic field will require a better understanding than now exists of the ways that rocks in fact obtain and sometimes change their fossil magnetizations. The contribution in this paper consists of finding correlations between the petrology, Curie points, and polarities of natural magnetization of basalt lavas.The evidence contradicts previous evidence on which was based the hypothesis of exsolution of iilvospinel and magnetite from a titanomagnetite solid solution. It suggests strongly that this exsolution does presenr t rtek ah from ibni irl-a-~?svJ-*ck&s-hxirlrhg ~&emht-d dhs. --r 1iPufi1111a tt' goal is to explain the behaviour of earthquakes. In this preliminary work we should like to point out a few results, which already seem to provide explanations for some phenomena found in nature.(1) In the photoelastic model experiments, with improved boundary conditions as compared with earlier mathematical models, the increase of shear stress towards the ends of a fault is confirmed. There is also no longer an infinite shear stress at the fault ends.(2) The shear stress reaches a much higher absolute value at the left end of the slit in profile I, models A, B (Figure 6) than on its right end. For a profile on the opposite side of the slit, the behaviour is opposite. The strain release density for (the Desert Hot Springs aftershocks (Richter & others 1958, Figure 6) exhibits a we deciuea' m i l t ' sysrCrfik~ic pet?uid&i;ai -u u~r v -d c i~~l r~k a u -h i e ' i~~u s r -~~~~e mental feature. The greatest immediate contribution we could make was to couple petrological classification of types of iron oxide grains with the Curie points and natural remanences of rocks. The following results question the general existence of iilvospinel-magnetite exsolution in lavas, and so contradict earlier evidence for this exsolution. Some new correlations exist which must be incorporated into future theories of magnetization processes in basalt lavas. Curie point analysesCurie points were measured in 1 OOO Oe, using a modified version of Chevallier's (1932) apparatus, with a bifilar suspension. Specimens could be heated to 650 "C in 15 or 20 min, and were small enough that no thermal hysteresis could be measured. We heated 247 specimens, mainly Tertiary plateau basalts, together with some from the sea-floors of the Atlantic and Pacific Oceans.The Curie points of basalts are not always sharply defined. Low Curie points cannot be specified to better than 550°C on this apparatus. Figure 1 contains