Trial measurements allow analytical results to be obtained by matching
the sample volume with the dynamic range of the detector. In flow ana
lysis, implementation of this approach can be accomplished by carrying
out sequential injections or using concentration gradients. Trial mea
surements are performed using an increasing concentration sequence and
the results are compared with a previously set threshold value. Once
this value is surpassed, the remainder of the sample by-passes the det
ector and goes to waste. The potential and limitations in unsegmented
and monosegmented flow systems are discussed. Improved systems for the
automated determination of manganese in rocks by atomic absorption sp
ectrometry are proposed. The typical measurement precision of flow inj
ection was maintained and no baseline drift was observed. Results were
in agreement with those obtained by inductively coupled plasma atomic
emission spectrometry.