Functions of mammalian neocortex rely on complex but precise neural circuits. Sensory areas of neocortex receive a lot of inputs from the thalamus, and thalamocortical (TC) connectivity is reorganized by thalamic inputs during early postnatal development. However, the dynamic reorganization processes of TC connectivity are not well understood. In vivo time-lapse imaging is a key strategy to obtain such information. In applying this technique, we took advantage of the unique morphological features of mouse "barrels" as a model. Barrels are specialized modular patterns in rodent somatosensory cortex layer 4 (L4), which correspond to the arrangement of whiskers on the face. TC axon terminals are clustered in the barrel center; spiny stellate neurons (barrel cells) are located at the barrel walls; and their dendrites are oriented toward the barrel center (orientation bias). We especially focused on postsynaptic L4 neuronal features. We first examined precise timing of barrel wall formation histologically and found that barrel walls start to emerge between postnatal day 3 (P3) and P4. In addition, we analyzed neuronal morphology at P3 and P6. We found that at P6 majority of L4 excitatory neurons are barrel cells, which have no apical dendrites, and their dendrites have orientation bias as in adult. In contrast, at P3 all L4 excitatory neurons had apical dendrites and dendritic orientation was undetectable because TC axon clusters were not formed yet. These data showed that most of barrel features are formed dramatically between P3 and P6. Recently we reported in vivo time-lapse imaging of barrels over a period of 9 and 18 hours starting at P4 and P5, respectively (Mizuno et al., Neuron 2014). We are improving the imaging system to enable longer-term imaging over a period of 3 days between P3 and P6, which may reveal how barrel walls are formed, whether and how barrel cells loose apical dendrites and how barrel cells acquire dendritic orientation bias. Our studies may advance the understanding of the dynamic reorganization processes of TC connectivity.