Understanding the mechanism of high transition temperature (Tc) superconductivity in cuprates has been hindered by the apparent complexity of their multilayered crystal structure. Using a cryogenic scanning tunneling microscopy, we report on layer-by-layer probing of the electronic structures of all ingredient planes (BiO, SrO, CuO2) of Bi2Sr2CaCu2O 8+δ superconductor prepared by argonion bombardment and annealing technique. We show that the well-known pseudogap (PG) feature observed by STM is inherently a property of the BiO planes and thus irrelevant directly to Cooper pairing. The SrO planes exhibit an unexpected Van Hove singularity near the Fermi level, while the CuO2 planes are exclusively characterized by a smaller gap inside the PG. The small gap becomes invisible near Tc, which we identify as the superconducting gap. The above results constitute severe constraints on any microscopic model for high Tc superconductivity in cuprates.PACS numbers: 74.72. Gh, 68.37.Ef, 74.50.+r, 74.25.Jb Superconductivity in perovskite-type layered cuprates [1], which thus far hold the record for the highest transition temperature (T c ), ranks among the most challenging and engaging problems in modern condensed matter physics. Despite nearly three decades' tremendous efforts of research all around the world, the key mechanism behind the Cooper pairing that lies at the heart of high-T c cuprate superconductors still remains puzzling. Many intriguing phenomena that intertwine with the occurrence of superconductivity have been discovered, leading to a very sophisticated phase diagram of cuprates [2]. These phenomena include the ubiquitous existence of various sorts of broken-symmetry states (e.g. charge density wave, spin density wave and electron nematicity) and the well-known pseudogap (PG) phenomenology, which has been considered a key finding in the research of cuprate superconductivity. A vast amount of experimental and theoretical studies have been devoted to understanding these phenomena themselves and their possible interplay with superconductivity, but so far most of which fell flat.From the view point of crystal structure, the cuprates consist of superconducting CuO 2 layers and charge reservoir building blocks (e.g. BiO/SrO in Bi 2 Sr 2 CaCu 2 O 8+δ (Bi-2212)) that stack alternatively along the crystallographic c-axis. In Bi-based cuprate superconductors, it is widely thought but empirical that the BiO and SrO block layers are insulating [3]. A considerable amount of surface-sensitive measurements, e.g. via angle-resolved photoemission spectroscopy (ARPES) [4] and scanning tunneling spectroscopy (STS) [5], have been conducted on the vacuum cleaved BiO planes of Bi-based cuprates and contributed largely to the cuprate PG data base. The measurements are generally assumed to reflect the superconducting properties of the CuO 2 planes, despite that they are located 4.5Å beneath the top BiO plane. Yet, such model has not been rigidly tested experimentally thus appears contentious, particularly regarding to the fact tha...