Perovskite nanocrystals have recently emerged as one of the most interesting optical materials for their various renewable energy applications like solar cell, LED and so on.[1] With tremendous progress in last five years, origins of their phase instability and enhancement of photoluminescence intensity are more clearly understood; but their growth mechanism leading to transformation from one size domain to other has not yet been achieved. Furthermore, to increase the stability of perovskite nanocrystals with respect to exposure to polar media, layers growth or shelling with different material is in demand. Unfortunately, no success has been achieved for epitaxial secondary growth, even with growth of a single monolayer of same material to any facet or on entire surface of these nanocrystals could not be established yet.[2] The possible reasons for this might be the ionic nature of these materials, their rapid formation, and difficulty in decoupling the nucleation and growth processes during their formation. To address this, herein, a secondary growth approach leading to creation of a secondary lattice with subsequent expansion on preformed CsPbBr­₃ perovskite nanocrystals is presented.[3] Traditional direct layer growth by adding precursors was not successful owing to the composition sensitivity with added Cs(I) or Pb(II) on preformed CsPbBr₃ nanocrystals resulting in the high possibility of instantaneous composition variations and phase changes. To achieve this secondary growth, the approach of Cs-sublattice extension on preformed seed CsPbBr₃ nanocrystals followed by their lattice expansion with Pb(II) insertions was adopted for the secondary layer growth. Asymmetric CsPbBr₃ nanocrystals having both {110} (or {002}) and {200} facets were seen to be more suitable compared to traditional cube shaped nanocrystals as the seed nanocrystal and Cs-lattice were allowed to be extended following CsBr introduction. In the first step CsBr was observed to be connected with lattice matching to the {200} facets of the parent nanocrystal. Further with Pb(II) incorporation, these CsBr lattices were expanded to CsPbBr₃. Interestingly, during lattice expansion with Pb(II), lattice mismatch was then created at the interface. This led to epitaxially connected twin or triple coupled nanocubes, which had a lattice mismatch between their {200} and {002} facets at the interface even though all components of the nanostructure were of the same phase. This lattice mismatching at the junction restricts the complete shell growth over all facets of preformed CsPbBr₃ nanocrystals, indicating growth of these nanostructures from one size domain to another is practically very difficult. While growth of CsPbBr₃ has remained a major question to date, this result suggests that partial layer growths on perovskite might be possible. However, for complete shelling with layer-by-layer growths which requires formation of a uniform sub-lattice structure of Cs on all facets surrounding the seed nanocrystals might be a difficult process from a practical perspective. Hence, the finding here certainly opens possible routes for secondary growth, directional growth, size/shape tuning, and also for the long desired heterostructure formation with epitaxial connection of perovskite nanocrystals.