As the site of folding for proteins of the secretory pathway, the endoplasmic reticulum (ER) must be equipped with an efficient quality control machinery. Through ER associated degradation (ERAD) misfolded lumenal and membrane-bound proteins are tagged with poly-ubiquitin chains and degraded in a proteasome-dependent manner. ERAD is a multi-step process which is initiated by the identification of the misfolded protein. While lumenal substrates degraded by the ERAD-L pathway appear to be recognized by a number of factors, it is unknown how misfolded membrane proteins are identified by the ERAD-M pathway. As a rate-limiting factor for ERAD, and a specificity factor as a ubiquitin ligase, we tested whether Hrd1p plays a role in the identification of misfolded proteins. By mutating a number of conserved and hydrophobic residues within the Hrd1p transmembrane domain, we illustrated that Hrd1p can distinguish between misfolded lumenal and membrane proteins, and that specific residues were important for the degradation of specific membrane proteins. Thus, in addition to Hrd1p's role as a ubiquitin ligase, it appears that this protein plays a role in the recognition of misfolded membrane proteins as well. Our studies also demonstrated that substrate binding does not appear to be sufficient to initiate ubiquitination, and that some type of signal is required following binding in order to activate the RING domain of Hrd1p. Following substrate ubiquitination, the protein must be removed from the ER in a process known as retrotranslocation, as the proteasome is located in the cytoplasm. To study retrotranslocation, we established an in vitro retrotranslocation assay, which we demonstrated was capable of retrotranslocating a natural ERAD-M substrate, Hmg2p. Retrotranslocated Hmg2p was soluble and full-length, and the process was dependent on Cdc48. With this assay, we were able to test the current model of retrotranslocation, in order to determine the core set of proteins which are essential for this process. This uncovered a role for the factors Ubx2p and Rad23p/Dsk2p in retrotranslocation, while demonstrating that the putative retrotranslocons Sec61p, Der1p/Dfm1p, and Hrd1p are dispensible for Hmg2p retrotranslocation. These studies have greatly increased our understanding of HRD-dependent degradation.