We derive atmospheric parameters and lithium abundances for 671 stars and include our measurements in a literature compilation of 1381 dwarf and subgiant stars .
First , a “ lithium desert ” in the effective temperature ( T _ { \mathrm { eff } } ) versus lithium abundance ( A _ { \mathrm { Li } } ) plane is observed such that no stars with T _ { \mathrm { eff } } \simeq 6075 K and A _ { \mathrm { Li } } \simeq 1.8 are found .
We speculate that most of the stars on the low A _ { \mathrm { Li } } side of the desert have experienced a short-lived period of severe surface lithium destruction as main-sequence or subgiant stars .
Next , we search for differences in the lithium content of thin-disk and thick-disk stars , but we find that internal processes have erased from the stellar photospheres their possibly different histories of lithium enrichment .
Nevertheless , we note that the maximum lithium abundance of thick-disk stars is nearly constant from \mathrm { [ Fe / H ] } = -1.0 to -0.1 , at a value that is similar to that measured in very metal-poor halo stars ( A _ { \mathrm { Li } } \simeq 2.2 ) .
Finally , differences in the lithium abundance distribution of known planet-host stars relative to otherwise ordinary stars appear when restricting the samples to narrow ranges of T _ { \mathrm { eff } } or mass , but they are fully explained by age and metallicity biases .
We confirm the lack of a connection between low lithium abundance and planets .
However , we find that no low A _ { \mathrm { Li } } planet-hosts are found in the desert T _ { \mathrm { eff } } window .
Provided that subtle sample biases are not responsible for this observation , this suggests that the presence of gas giant planets inhibit the mechanism responsible for the lithium desert .