Rainbow trout of the Kamploops variety were sampled at intervals from October to the end of March. During this period a decline was noted in red cell count, haematocrit, haemoglobin concentration and plasma osmotic concentration. Increases were seen, however, in mean cellular volume, mean corpuscular haemoglobin, mean corpuscular haemoglobin concentration and the water content of liver and dorsal muscle. Sexual differences were found in all values with the exception of mean corpuscular haemoglobin concentration and the water content of both tissues. Males always had higher values in those parameters in which sexual differences were noted. All trends, with the exception of male haemoglobin and mean cellular volume and female mean cellular volume and osmotic concentration, were significantly linear.None of the findings in this study could be correlated with temperature or photoperiod. Neither could the declining plasma osmotic concentration be correlated with the rising mean cellular volume or tissue water content. However, correlations were noted between haemoglobin and haematocrit, red cell count and haematocrit and between osmotic concentration and haematocrit. A negative correlation was seen between mean corpuscular haemoglobin concentration and mean cellular volume.

Anaemia can be induced in Rainbow trout by removing 12 % of the blood volume by cardiac puncture. Recovery from such an anaemia has been monitored in both normal and splenectomized fish. After the bleeding there is a decrease in red cell numbers, haematocrit and blood haemoglobin concentration. Although recovery in normal trout was not complete within 27 days, which was the period of the observations, it was seen as a replenishing of red cell numbers, haematocrit and blood haemoglobin concentration. Mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC), both depressed by the bleeding, did not appear to recover during the experimental period.Splenectomy reduced erythropoietic intensity. After bleeding all the parameters studied were more depressed than those found in normal trout, and recovery was not noted. Blood haemoglobin and red cell numbers did not have the same post-bleeding pattern as normal individuals; MCH and MCHC values remained markedly depressed.Splenic involvement was further demonstrated by bleeding trout for a second time from both normal and splenectomized groups. In contrast to the response of normal animals, there was a further deterioration of red cell numbers and blood haemoglobin concentration in splenectomized individuals.

Peripheral blood of trout contained two populations of red cells: those with polyribosomes located in the cytoplasm, and those without polynbosomes. Starvation of trout for 30 days was accompanied by a proportional decline of the polyribosomal-containing (PRC) red cells. One week after a 15% bleeding of both fed and starved animals fed individuals showed a proportional decline of PRC red cells whilst starved fish showed a proportional increase of the same cell population. In fed individuals the bleeding response was accompanied by the appearance of many red cells with senescence-related characteristics. PRC cells in both groups of animals were arbitrarily subdivided into three subgroups according to the density of polynbosomes present. No statistically demonstrable differences were evident between the means of the three PRC cell groups of control animals and those subjected to starvation and bleeding. However, there was an apparent rise in the proportion of red cells with the highest density ofpolyribosomes as a result ofboth treatments.

Velocity sedimentation at unit gravity separated peripheral erythrocytes of adult rainbow trout according to age and size. The cells located at the top of the gradient contained numerous polyribosomes, few mitochondria and an essentially electronlucent nucleus. The cells at the bottom of the gradient no longer contained these cytoplasmic organelles and had an essentially electron-dense nucleus. The middle of the gradient contained cells with either very few polyribosomes or cells devoid of this organelle.The length, width and mean corpuscular volume (MCV) ofthe average cell increased and the width to length ratio decreased progressively from the top of the gradient to the bottom.No statistical significance could be determined in the changes of the mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) although they appeared to follow theoretical projections.The use of haematocrit, red cell count, haemoglobin concentration, MCV, MCH and MCHC in assessing erythropoietic activity and the size and haemoglobin content of the maturing cell is discussed.

The nucleotide triphosphate/haemoglobin (NTP/Hb) and nucleotide triphosphatdred cell (NTP/cell) ratios of rainbow trout increased during prolonged starvation. A decline was noted in blood lactic acid concentration. Red cell count, haemoglobin concentration and haematocrit also declined. Changes in mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) were found not to be significant. The NTP/Hb and NTP/cell ratios of both fed and starved trout rose seven days following a 15% reduction in blood volume by cardiac puncture. A rise in whole blood NTP concentration was found only in the bleeding response of fed animals. No significant change was noted in blood lactic acid concentration. The decline in haematocrit was significant only in the starved group. In both groups, however, red cell count and blood haemoglobin concentration fell. MCV rose whereas MCHC declined in all bled animals. Changes in MCH were not significant in either group. Negative correlations were noted between red cell count and both the NTP/Hb and NTP/cell ratios and between haemoglobin concentration and the NTP/Hb ratio. Positive correlations were seen between the two ratios and between red cell count and haemoglobin concentration.