Kaszab Zoltán (szerk.): A Magyar Természettudományi Múzeum évkönyve 72. (Budapest 1980)
Embey-Isztin, A.: Major element patterns in Hungarian basaltic rocks: an approach to determine their tectonic settings
nodules (for details see: EMBEY-ISZTIN 1976a, 1976b, 1977, 1978) so they can be regarded as primary by the first criterion. Secondly, the evidence from mantle-derived rocks (lherzolite xenolithsfrom basalts •and kimberlites, diapiric intrusives of high P and high T peridotites, peridotite bodies exposed in modern mid-oceanic ridge and old ophiolite complexes) indicates that the least refractory upper mantle compositions have 100 Mg/Mg+ Fe 2+ (Mg-values) around 88-89. If such a source material becomes melted (up to about 30 per cent) the basaltic melt must have Mg-values «68-75 using K D (ót/u?) = 0.3 published by ROEDER & EMSLIE (1970). Considering this criterion in the case of the young alkali basalts, we can say that the basalt of Badacsony with a Mg-value of 67-70, and that of Sághegy (68-71), Nagyláz (70), Sarvaly (68), Pécskő (71) can be identified as primary partial melts. The basalt of Pálhegy (Pauliberg) with a very high concentration of MgO («12 per cent) and NiO '(«300 ppm) (MAURITZ et al. 1948, SCHARBERT pers. comm.) must be regarded as a very primitive partial melt. On the contrary, basalts of Tátika (Mg = 51-67), Haláp (64-65) and Kabhegy (63-68) as well as some Cserhát basalts are to be considered as fractionated liquids. The depth of the fractionation cannot be defined with certainty, it may have occurred either in the mantel itself or at lower tcrustal levels. The third criterion which is likely useful in distinguishing between primary melts and derivative magmas involves compatible trace elements such as Sc, Cr, Co, and Ni. Since these elements have -solid/liquid partition eoeficients greater than unity for at least one of the major mantle phases, their concentrations in a magma are very sensitive to fractional separation of mafic minerals. Consequently, rocks having the highest concentrations from these elements have likely experienced very little fractionation or none at all. Especially Ni abundances have been used to identify magmas with very >low fractionation rate of olivine (Gast 1968; Kay et al. 1970; Allègre et al. 1977; Frey et al. 1978). Evidence from NiO.concentrations published by MAURITZ et al. (1948) is not always in accordance ^with the one we can obtain from Mg-values. While the Pálhegy basalt has both a high Mg-value •and NiO abundance (300 ppm) and the Tátika basalt with a low Mg-value only 0-100 ppm NiO, the Haláp basalt though lower in Mg-value contains also 300 ppm NiO. The basalt of Kabhegy has an intermediate amount of NiO (200 ppm). Summing up, the result of the discriminant analysis has shown that the young alkali basalts of Hungary are typical WPB-basalts that were erupted through a continental-type crust not too different from the present one, judging from geological and geophysical evidences. The basaltic magma has not been modified by chemical interactions with wall-rocks to any sensible extent. Among the lavas some represent primary melts, while others, liquids modified by olivine and perhaps pyroxene fractionation. iLower Cretaceous alkali basalts As it was mentioned earlier only a few analyses were published from these rocks. As a matter of fact, only four of them have MgO+CaO between 12 and 20, therefore analyses with more and less MgO+CaO have also been plotted (by different symbols) to evaluate the effect -of differentiation (Fig. 3). It is remarkable that all the four "true" basalts fall in the WPBfield and so do some of the trachyandesitic differentiates though a part of them lies in the shoshonitic (SHO) field of the diagram. On the contrary, the cumulative-type rocks (with more MgO +CaO than 20 per cent) have very low values of F 2 with respect to their F r values, consequently, they do not fall in any of the fields, however their vicinity to the WPB-field must be highly significant. Now if we consider the disposition of all the analyses, we can see that they roughly follow two lines, one with increasing F 1 and F 2 and another one with slightly increasing Fj and decreasing F 2 . The first line represent the trachyandesitic trend of differentiation which leads to the highly undersaturated phonolites, the second one may be a complementary differentiation process essentially due to olivine and clinopyroxene accumulation. Considering all these facts, we can quite safely say that in spite of the small number of "true" basalt analyses the parental magma of the Lower Cretaceous volcanism must be definied as a WPB-basalt. Since Pearce gives a set of scaled eigenvectors too, we are able to evaluate the contribution of a given major element to the total discriminating power expressed by F lt .F 2 , and F 3 . For example, the largest scores on the second function (F 2 ) are for MgO (-0.74)
