Torrentfish (Cheimarrichthys fosteri Haast 1874) are a freshwater fish species endemic to New Zealand and information pertaining to their general distribution, abundance as well as feeding ecology has been well documented. However, there are still fundamental gaps in torrentfish understanding that relate to aspects of their reproductive biology and migration history. In this study the population dynamics and reproductive aspects of torrentfish in two small Waikato streams as well as small scale sampling in the Mangamuka River on the North Island of New Zealand were investigated. The aim was to describe seasonal distribution patterns of male and female torrentfish within study streams as well determine the timing of spawning events by way of gonad assessments. Results indicated that females were significantly more abundant than males in upper reach sites of both streams but were not exclusively dominant as was the case for torrentfish in the Rakaia River. However, in lower reach sites females were equally if not more abundant than males demonstrating considerable overlap between the sexes. Seasonal assessments of male and female gonads using gonosomatic indices (GSI) suggests that spawning occurs over a single protracted period in both streams. Fully ripe females were observed from January through to peak spawning condition in April in the Rangitukia Stream. While in the Mangauika spawning was delayed for a month with ripe females observed from February through to peak condition in May. The distribution of ripe and spent males and females between upper and lower reach sites of both streams during observed spawning periods suggests spawning occurs throughout the upper and lower reaches of both streams. In the second part of this study, known differences in freshwater chemistry (low Sr/Ca ratios) and saltwater chemistry (high Sr/Ca ratios) were used as a premise to track torrentfish movements between fresh and saltwater by analysing the chemistry of their earbones (otoliths). Otoliths from fish in both study streams were analysed by laser ablation inductively coupled mass spectrometry (ICP-MS). Overall trends in otolith Sr/Ca ratios for torrentfish at distance from the otolith core clearly reflected patterns indicative of movements between fresh and saltwater environments demonstrating a marine life history for torrentfish. However, the overall concentration of Sr/Ca ratios in otoliths (3) did not reflect sea water molar ratios (8.6) and were more aligned with estuarine water molar ratios (4). Lower than expected Sr/Ca core ratios may have been due to physiological constraints and a shorter marine life history whereby Sr uptake would occur over a shorter time period. However, as these ratios are the first reported for this species using laser ablation (ICP-MS) further otolith work is necessary to determine if the range in Sr/Ca core ratios is more variable. Additionally, multivariate discriminant function analysis (DFA) of otolith edge values in torrentfish were used to determine whether fish could be classified to their stream of capture or whether male and females could be discriminated between upper and lower reach fishing sites. The results suggested there were significant differences in concentrations of Sr in the Rangitukia and Cr and Mg in the Mangauika Stream. DFA analysis of the same elements including Ba and Mn, correctly classified 100 % of all water samples to their stream of collection. However, elemental concentrations of otolith edge values in torrentfish from both streams demonstrated no significant differences using the same elements identified in water samples or a wider range of elements These indifferences in water chemistry and otolith chemistry may have been attributed to laser spot bias and fish physiology as a result of torrentfish's adaptation to fast flowing streams which may influence brachial uptake of elements in otoliths.However, despite indifferences DFA analysis correctly classified 80% of fish to their stream of capture.