The trimeric intracellular cation channels, of which there are two subtypes (TRIC-A and TRIC-B), are located on endo/sarcoplasmic reticulum (ER/SR) and nuclear membranes where they are thought to provide counter-current for ER/SR Ca2+-release [1, 2]. TRIC-B knockout mice die immediately after birth demonstrating the essential role of this isoform [3]. We aimed to define the key aspects of TRIC-B channel gating and investigate how gating is regulated by voltage. Both TRIC-A and TRIC-B are present in most tissues and therefore to be certain that we record only TRIC-B single-channel events, we incorporated skeletal muscle light SR from TRIC-A knockout mice into artificial membranes under voltage-clamp conditions [2]. Recordings were obtained in symmetrical 210 mM K-PIPES, pH 7.2. Single-channel recordings were idealized using the segmental k-means (SKM) algorithm implemented in QuB software [4]. We developed Markov models of TRIC-B gating, with up to 4 distinct sub-conductance states (S1-S4), using both QuB and our own software. Mean-variance histograms [5] were computed using our own code. There was a high degree of variability in TRIC-B gating behaviour but almost 90% of channels (7 of 8 channels) exhibited voltage-dependent gating where channel activity was higher at positive than at negative holding potentials. At positive potentials, gating was characterised by long closures interspersed with bursts of openings in which the majority of time was spent in the full open state. At negative potentials, channel openings were rare and usually consisted of brief openings to the lower-conductance open states (S3 and S4). Even at positive holding potentials, Po plateaued to relatively low levels (< 0.2) over the voltage range +20 to +60 mV. Transitions to the full open state were typically preceded by brief transitions through a range of sub-conductance states. During an opening burst, TRIC-B often gated rapidly between the full open state and the higher-conductance open states, S1 and S2. Closure of the channel at the end of a burst also usually consisted of a tail of transitions through multiple sub-conductance states. The variable gating of TRIC-B and the frequent, rapid transitions between sub-conductance states prevents the development of reliable gating models using conventional single-channel analysis alone. We have therefore used mean-variance plots to provide additional quantitative and visual information that allows us to describe the key features of TRIC-B gating. Our model of TRIC-B gating provides the basic framework for developing an understanding of how TRIC-B may be regulated in situ and will help us to determine the individual and combined roles of TRIC-A and TRIC-B in supporting intracellular Ca2+-release.