I. N. Tolstikhin, I. L. Kamensky, V. A. Nivin, V. R. Vetrin, E. G. Balaganskaya, S. V. Ikorsky, M. A. Gannibal, and Yu. M. Kirnarsky
Geological Institute, Kola Scientific Centre, Russian Academy of Sciences, Apatity 184200, Russia
B. Marty
Centre National de la Recherche Scientifique Centre de Recherches Petrographiques et Geochimiques(C.R.P.G.) B.P. 20, 54501 Vandoeuvre-Nancy Cedex, France
D. Weiss, A. Verhulst, and D. Demaiffe
Petrologie et Geodynamique Chimique Department des Sciences de la Terre et de l`Environnement Faculte des Sciences Universite Libre de Bruxelles Avenue F. D. Roosevelt 50 B-1050 Bruxelles, Belgique
108 down to
6 
104. A comparison of
expected from in-situ closed-system production (calc) and measured (meas)
abundances has shown that
4He
meas 
4He
calc whereas in some
rocks and minerals
3He
meas exceeded
3He
calc up to 1,000 times
indicating occurrence of a mantle fluid in a majority of samples. Gas extraction
by milling of rock/mineral samples liberated fluid-related helium with as low
4He/
3He ratios as
3104. These values, well below
the mean MORB ratio of
(8.9 
1) 104,
show a contribution of
3He-rich plume-like
component.
A reason for highly variable
4He/
3He ratios in similar rocks selected from
different massifs is not well understood yet. The following processes appear to play
major roles: (i) different degree of melt degassing, trapping and subsequent retention
of trapped (initially homogeneous) fluid and (ii) various contributions of radiogenic
in-situ produced He.
Isotope composition of Ne supports the occurrence of plume component:
20Ne/
22Ne ratio varies from 10.4 to 12.0 and
20Ne/
22Ne versus
21Ne/
22Ne correlation
(arising from mixing of the mantle and atmospheric components) is quite similar to
that observed for Loihi hot spot, Hawaii.
20Ne/
22Ne versus
40Ar/
36Ar correlation and
K-Ar systematics give initial
40Ar/
36Ar
3,000 for the mantle end-member, which is
by an order of magnitude lower than that accepted for the upper mantle, source of
MORB. Contamination of parent UAC melts by air argon can not be ruled out.
A comparison of these data with rare gas systematic in the mantle enable us to
suggest a contribution of material from the less degassed deep mantle reservoir,
source of primordial rare gases, to Devonian ultrabasic-alkaline and carbonatite
rocks
on the Kola Peninsula. Available model characterisations of principle terrestrial
reservoirs allow contributions (by mass) of the lower mantle, the upper mantle, and
the atmosphere (an air-saturated groundwater) to be evaluated: 2%, 97.95%, and
0.05%, respectively.