Linear [Q/H]-[O/H] relations, Q = Na, Mg, Si, Ca, Ti, Cr, Fe, Ni, are
inferred from a sample (N = 67) of recently studied FGK-type dwarf stars in
the solar neighbourhood including different populations (Nissen and Schuster
2010, Ramirez et al. 2012), namely LH (N = 24, low-? halo), HH (N = 25,
high-? halo), KD (N = 16, thick disk), and OL (N = 2, globular cluster
outliers). Regression line slope and intercept estimators and related
variance estimators are determined. With regard to the straight line,
[Q/H]=aQ[O/H]+bQ, sample stars are displayed along a "main sequence", [Q,O]
= [aQ, bQ, ?bQ], leaving aside the two OL stars, which, in most cases (e.g.
Na), lie outside. The unit slope, aQ = 1, implies Q is a primary element
synthesised via SNII progenitors in the presence of a universal stellar
initial mass function (defined as simple primary element). In this respect,
Mg, Si, Ti, show ?Q = 1 within -+2^??Q; Cr, Fe, Ni, within -+3^??Q; Na, Ca,
within -+r^??Q, r > 3. The empirical, differential element abundance
distributions are inferred from LH, HH, KD, HA = HH + KD subsamples, where
related regression lines represent their theoretical counterparts within the
framework of simple MCBR (multistage closed box + reservoir) chemical
evolution models. Hence, the fractional yields, ^pQ/^pO, are determined and
(as an example) a comparison is shown with their theoretical counterparts
inferred from SNII progenitor nucleosynthesis under the assumption of a
power-law stellar initial mass function. The generalized fractional yields,
CQ=ZQ/ZaQ O, are determined regardless of the chemical evolution model.
The ratio of outflow to star formation rate is compared for different
populations in the framework of simple MCBR models. The opposite situation
of element abundance variation entirely due to cosmic scatter is also
considered under reasonable assumptions. The related differential element
abundance distribution fits to the data, as well as its counterpart inferred
in the opposite limit of instantaneous mixing in the presence of chemical
evolution, while the latter is preferred for HA subsample.