75 and 32 mm heparinized and plain capillaries. 

 In another series plain capillaries were filled 

 with untreated blood which was permitted to co- 

 agulate completely. At the same time blood 

 treated with anticoagulants was also introduced 

 into similar untreated capillaries and both were 

 centrifuged in the same manner and at the same 

 time. Hemoglobin values obtained with the hemo- 

 globinometer were checked with a Bausch and 

 Lomb "Spectronic 20" spectrophotometer. 



The presentation of quantitative data in 

 tables including mean values, mean deviations, 

 variances or standard deviations and frequency 

 distributions may seem unnecessary to some 

 readers, but may be of value to others who want 

 to compare statistically their own data with those 

 presented here. 



RESULTS 



Data presented in this report are based 

 on examination of about 300 trout. Red cell counts 

 and estimation of hemoglobin, in addition to micro- 

 hematocrit, were made with 64 of the trout 

 examined. In table 1 data are presented on micro - 

 hematocrits. In most cases to or three hemato- 

 crits were determined for each trout. This was 

 done to determine the reproducibility of the hema- 

 tocrit method when applied to trout blood. The 

 frequency distribution of these replicated hema- 

 tocrits shows that in 80 to 90 percent of the cases 

 the difference between the replicates is + 1 unit 

 or less. This is a very high degree of reproduc- 

 ibility. 



The mean hematocrit values for all three 

 species of trout were between 40 and 50, with an 

 average about 45 . The frequency distribution of 

 hematocrit values seemed to become less regular 

 with increasing age. It seems likely that this was 

 caused by gonad development in yearling trout, 

 and/or high incidence of mycosis -like granuloma 

 (Wood et al., 1955) observed in yearling eastern 

 brook trout. 



The relationship between the hematocrits, 

 red cell counts and hemoglobin as determined in 

 the same trout are presented in table 2 and figure 

 1 . From this it can be seen that as in other 

 vertebrates, there is a correlation between these 

 values. 



Hematocrit readings in apparently normal 

 trout differ only slightly from average hematocrit 

 levels in humans and most other vertebrates. It 

 is believed, therefore, that figures obtained for 

 trout at the Leetown hatchery may be tentatively 

 considered "normal" for trout. How representa- 

 tive these figures are for the same species of 

 trout of the same age but with different environ- 

 mental conditions remains to be demonstrated. 



Comparative data on microhematocrit 

 values from rainbow trout blood pre -treated with 

 anticoagulant with that taken directly into two dif- 

 ferent brands of heparinized capillaries are 

 presented in table 3. Pre-treated samples gave 

 values 18 percent lower than those taken in heparin- 

 ized capillary tubes . 



When fish blood was collected directly into 

 capillaries treated with anticoagulants and such 

 capillaries were kept in vertical position, there 

 was no visible sedimentation of blood cells. How- 

 ever, if blood was treated with the same antico- 

 agulants before introducing into capillaries, blood 

 cell sedimentation proceeded at about the same 

 rate as with similarly treated human blood. Also 

 when untreated and pretreated blood was kept in 

 capillaries for some time and then blown out on a 

 porcelain plate, the pretreated blood was uniformly 

 liquid while blood collected directly into treated 

 capillaries had a gelatinous consistency. Blood 

 collected into plain capillaries was coagulated and 

 the clot was well separated from the clear serum . 



Samples of fresh human blood were run at 

 the same time. Blood collected in treated capil- 

 laries did not clot. Cell sedimentation was rapid 

 and hematocrit values, regardless of the type of 

 handling, were identical within the range of ex- 

 perimental error. 



These observations show that even in treated 

 capillaries some coagulation of trout blood occurs 

 and hematocrit values are somewhat higher than 

 for pretreated blood. 



Data are presented in table 4 on hematocrits 

 run with trout blood in 75 mm and 32 mm capillar- 

 ies (Strumia, 1954) . In all cases determinations 

 were made with blood introduced directly into 

 heparinized capillaries and with blood the coagula- 

 tion of which was prevented by pretreatment with 



