I review the uncertainties in two observational local constraints of the Galactic disc chemical evolution : the metallicity distribution of long-lived dwarfs and the age-metallicity relation .
Analysing most recent data , it is shown first that the observed metallicity distribution at solar galactocentric radius , designed with standard methods , is more fit to a closed-box model than to the infall metallicity distribution .
We argue that this is due to the specific contribution of the thick disc population , which has been overlooked in both the derivation of the observed metallicity distribution and in standard chemical evolution models .
Although this agreement disqualifies the metallicity distribution as the best supportive ( indirect ) evidence for infall , we argue that the evolution must be more complex than described by either the closed-box or standard infall models .

It is then shown that recent determinations of the age-metallicity distribution from large Strömgren photometric surveys are dominated by noise resulting from systematic biases in metallicities and effective temperatures .
These biases are evaluated and a new age-metallicity distribution is obtained , where particularities of the previous determinations are phased out .
The new age-metallicity relation shows a mean increase limited to about a factor of 2 in Z over the disc age .
It is shown that below 3 Gyrs , the dispersion in metallicity is about 0.1 dex , which , given the observational uncertainties in the derived metallicities , is compatible with the small cosmic dispersion measured on the ISM and meteoritic presolar dust grains .
A population that is progressively older and more metal-rich arises at a metallicity greater than that of the Hyades , to reach [ Fe/H ] \approx +0.5 dex at ages greater than 5 Gyrs .
We suggest that this is best explained by radial migration .
A symmetrical widening of the metallicity interval towards lower values is seen at about the same age , which is attributed to a similar cause .
Finally , the new derived ages are sufficiently consistent that an age-metallicity relation within the thick disc is confirmed .
These new features altogether draw a picture of the chemical evolution in the solar neighbourhood where dynamical effects and complexity in the age-metallicity distribution dominate , rather than a generalised high dispersion at all ages .