Arguably, the act of reaching constitutes one of the most devoted lines of contemporary developmental research. In addition to the underlying dynamical characteristics of motor coordination, a key element in programming is limb selection; a phenomenon [handedness] that has so far resisted any reasonable unified explanation. Traditionally, it has been assumed that if an individual truly preferred a particular hand, he or she would likely use that limb to carry out an array of unimanual activities, even in awkward positions; in essence, the definition of limb dominance. From a more contemporary view, two factors appear to have the most influence on hand selection for a given task: motor dominance and attentional information related to task demands. This study was designed to determine what factor(s) influence choice of limb for reaching in hemispace in reference to motor dominance, object proximity, and a hemispheric bias favoring use of the hand on the same side as the stimulus. Strong right-handed children (6- to 8 year-olds) were blindfolded and asked to reach and retrieve a small object across right and left hemispace locations beginning with the arms uncrossed and arms-crossed. Order of stimulus position was systematically given using a computer-generated list of random positions, counterbalanced between participants. With the arms-crossed, the participant had the choice of keeping the arms crossed to reach (favoring proximity) or uncrossing the limbs to reach ipsilaterally (favoring hemispheric bias). As expected, results from the uncrossed condition supported previous reported findings for adults and children. That is, participants responded ipsilaterally using the hand on the same side as the stimulus, thus, supporting the case for object proximity and hemispheric bias. In the arms-crossed condition, an interesting observation occurred; the vast majority of participants preferred keeping the limbs crossed in response to right (83% remained crossed) and left (91%) hemispace stimuli; chi-square comparisons, ps< .001. This result leads to the suggestion that object proximity rather than hemispheric bias was the driving factor in this context. Overall, the findings add to the growing acceptance that limb selection is task and context dependent, rather than a biologically-based invariant feature of motor behavior.