Haemoglobins perform the vital physiological function of transporting oxygen from the external environment to the tissues. Poikilothermic rainbow trout (Oncorhynchus mykiss = Salmo gairdneri = S. irideus) produce multiple forms of haemoglobin that respond differently to varying environmental and physiological conditions. However, the timescale and physiology of these changes are not known. Changes in isomorph abundance may potentially originate from the production of new isomorphs in already circulating erythrocytes. Alternatively, new isomorphs may be produced through the reassembly of extant haemoglobin subunits. The final hypothesis is that changes in isomorph abundance occur through the production of new erythrocytes with red blood cells ‘pre-programmed’ to produce a particular set of haemoglobins. Changes originating from production of new erythrocytes would require longer periods of time before being detectable; weeks to months depending on the temperature regime. To test this, paired groups of rainbow trout were subjected to either 10ºC or 20ºC for 5, 7, 14, 21 and 28 days. A total of 14 isomorphs were observed after the haemolysate was separated using cellulose acetate electrophoresis. However, no detectable differences in isomorph abundance were found between treatment groups. In a follow-up experiment, anaemia was induced in rainbow trout to stimulate the production of new erythrocytes. The trout were then held at either 10ºC or 20ºC for 21 days. This resulted in relative increases in the abundance of anodal haemoglobin isomorphs in the 20ºC acclimated group and a corresponding decrease in cathodal haemoglobin isomorphs. To further confirm that changes in abundance were occurring through the production of new erythrocytes, separation of erythrocytes into age classes was undertaken to compare the isomorphs present in mature erythrocytes with those from erythrocytes produced under amended temperature regimes. Using Percoll density gradients, red blood cells from anaemia-induced trout acclimated to either 10ºC or 20ºC were enriched into mature and young erythrocyte fractions. Further significant differences in abundance were found between both anodal and cathodal isomorphs when compared between treatment groups. From these results it was concluded that changes in haemoglobin isomorph abundance originated from the production of new erythrocytes. Cellulose acetate gel electrophoresis was carried out on seven haemoglobin fractions that had undergone prior separation from whole haemolysate by Fast Protein Liquid Chromatography. Each haemoglobin fraction was found to be composed of two isomorphs for a total of 14 haemoglobin isomorphs. The oxygen binding properties of each fraction was examined under varying conditions of temperature, pH, ATP and chloride concentrations. Cathodal functional groups HbI to HbIII were found to be insensitive to temperature, pH, chloride and the organic phosphate ATP. In contrast, the anodal fractions (HbIV to HbVII) all responded to pH and temperature changes, while HbVI and HbVII responded to ATP. However, no fraction responded to increased chloride concentrations. These results suggest that different varieties of rainbow trout may produce different forms of haemoglobin as part of an adaptive response to local environmental conditions, leading to variation in the functional properties of some of the less abundant functional groups such as HbIII. Despite the theory that cathodal haemoglobins function as emergency back-up supplies of oxygen being proposed more than thirty years ago, no published information can be found for it being tested in the laboratory. Two groups of anaemia-induced rainbow trout were placed in a divided annular flume for 24 days. The high activity group was subjected to a forced swimming speed of 2.5 body lengths (B.L.) s-1 for 6 h d-1. When not undergoing forced exercise the treatment group were maintained at the same speed as the low activity group of 0.5 B.L. s-1. Significant differences in haemoglobin isomorph abundance were present between the initial samples taken at the time of anaemia induction and high and low activity groups. However, only the C4 isomorph demonstrated a significant difference between high and low activity groups. When total anodal and cathodal isomorphs were compared between initial state, high activity and low activity groups, no differences were present. These data suggest that the induction of anaemia had an effect on the composition of the isomorphs but no physiological effect on oxygen delivery. In addition, the swimming velocity of 2.5 B.L. s-1 employed for the high activity group may have been an insufficient stimulus to induce changes in isomorph abundance. It is concluded that changes in haemoglobin isomorph abundance occur in response to chronic changes in the environment. Increases in the abundance of anodal isoforms in response to increasing temperature allows for increased delivery of oxygen to tissues undergoing increases in metabolic activity associated with higher temperatures. The multiple haemoglobin isomorphs of rainbow trout provide an increased efficiency in delivery of oxygen to tissues under varying metabolic conditions of pH, temperature and oxygen saturation. The cathodal and anodal haemoglobin functional groups of rainbow trout exhibit different oxygen affinities in response to temperature, pH and ATP concentration but not to physiologically realistic chloride concentrations. The oxygen binding properties of the isomorphs within the cathodal and anodal functional groups are broadly similar. However, differences in responses by the anodal functional groups to NTPs may exist. An examination of the hypothesis that cathodal haemoglobins act as reserve oxygen delivery sources under prolonged activity produced no significant results. However, this hypothesis still remains viable and needs to be tested under different experimental conditions. This work provides a basis for further research into the adaptive abilities of rainbow trout. The selection of rainbow trout which better adapt to wide ranges of environmental conditions would allow for targeted introduction by fisheries managers to aquatic systems previously not considered optimal for trout growth thereby expanding the fishery.