To-date serial block-face scanning electron microscopy (SBF-SEM) dominates as the premier technique for generating three-dimensional (3-D) data of resin-embedded biological samples at an unprecedented depth volume. Given the infancy of the technique, limited literature is currently available regarding the applicability of SBF-SEM for the ultrastructural investigation of tissues. Herein, we provide a comprehensive and rigorous appraisal of five different SBF-SEM sample preparation protocols for the large-volume exploration of the hepatic microarchitecture at an unparalleled X, Y and Z resolution. In so doing, we qualitatively and quantitatively validate the use of a comprehensive SBF-SEM sample preparation protocol, based on the application of heavy metal fixatives, stains and mordanting agents. Employing the best-tested SBF-SEM approach, enabled us to assess large-volume morphometric data on murine parenchymal cells, sinusoids and bile canaliculi. Finally, we integrated the validated SBF-SEM protocol with a correlative light and electron microscopy (CLEM) approach. The combination of confocal scanning laser microscopy and SBF-SEM provided a novel way to picture subcellular detail. We appreciate that this multidimensional approach will aid the subsequent research of liver tissue under relevant experimental and disease conditions. 4 #19}, heralded a new era in analysing the complex geometric arrangement of biological cells and tissues at the nanometre scale. Throughout the infancy of EM techniques, including both transmission (TEM) and scanning electron microscopy (SEM), the structural information gained was limited to two-dimensional (2-D) space, providing a mere snapshot of innately three-dimensional (3-D) structures.Throughout the mid-1950s, advances in ultramicrotome design and specimen preparation procedures{Porter and Blum 5 #25} facilitated the generation of 3-D ultrastructural information via the superimposition of consecutive 2-D TEM micrographs that were acquired from a serially sectioned specimen -a technique dubbed serial-section TEM (ssTEM){#387 }. Currently, ssTEM is regarded as a partial effort in the generation of 3-D information, primarily due to the fact that it is a highly arduous and labour-intensive technique that is riddled with several sources of error, such as section compression and the loss of serial sections, which ultimately results in an incomplete data set. The tediousness of the technique is further compounded by the necessity to align image stacks in order to correct for misalignments that occur during the manual imaging of successive sections{Williams and Kallman 12 #15}. Moreover, the generation of high-resolution axial resolution is