used as a tool for quality control in fish culture 

 or management. Selection of methods of hemato- 

 logical examination for research purposes depends 

 on the nature of the investigation and must be left 

 to the individual worker. In order to be used gen- 

 erally as quality control tools, the hematological 

 methods selected must be simple, rapid and as 

 free as possible from procedural errors. Hem- 

 atological examination may be used to detect some 

 types of malnutrition (Tunison^^, 1939), chron- 

 ic diseases, or disturbances caused by unfavorable 

 environment or pollution (MjCay, 1929). Red 

 cell counts and estimation of hemoglobin have 

 been the procedures used most frequently in the 

 past. Hematocrits were used rarely, because 

 the early hematocrit methods required large quan- 

 tities of blood which seldom were obtainable from 

 fish of the size raised in hatcheries. A detailed 

 review of reference material pertaining to hema- 

 tology of fishes can be found in papers by Yokoy- 

 ama (1947) and Katz (1949). 



Critique of the significance and of the 

 limitations of red cell counts, hemoglobin and 

 hematocrit determination can be found in recent- 

 ly published manuals of hematology or clinical 

 pathology, as for example the well known books 

 by Wintrobe (1958)and Wells (1956). 



TTie purpose of this investigation was to 

 determine the practicability of large-scale micro - 

 hematocrit measurements in fish culture and to 

 establish values which may serve as normal for 

 trout until more extensive data are collected. 



Determination of normal values for fish 

 is difficult. Changes in concentrations of dis- 

 solved oxygen rapidly affect hematological values 

 in fish (Hall £tal_., 1926; Adrianov, 1936; 

 Phillips, 1947; Dombrowski, 1953, and others). 

 For this reason hematological standards for fishes 

 are likely to have wide amplitudes. Only a very 

 thorough examination of fishes under different 

 concjitions may ultimately result in establishing 

 significant standards within normal amplitudes. 

 Our 'observations are a contribution to this aim . 



While a microhematocrit method was 

 described by Guest and Siler as early as 1934, 

 equipment became commercially available only 

 recently (McGovern et al. 1955). Since then the 

 microhematocrit method has gained general ac- 

 ceptance in clinical laboratories. Procedure 



described by Guest and Siler requires one drop 

 of blood or 20 to 40,ul (microliters or cubic milli- 

 meters). In the ultra -microhematocrit method 

 of Strumia et al. (1954) only 5 to 10, ul of blood 

 are used; thus hematocrits can be performed with 

 fish from which even less than one drop of blood 

 is obtained. 



METHOIDS 



The technique we used is essentially that 

 recommended by McGovern et al. (1955). Trout 

 used for this work were hatchery stock kept in 

 water of 12° to 14°C. (54° to58°F.). They were 

 picked at random, several at a time, and kept in 

 a small tank supplied with fresh and freely flowing 

 water of the same temperature until all were ex- 

 amined. Immediately prior to examination each 

 fish was completely anesthetized by exposure for 

 about one minute in a water solution (approx. 1: 

 2000) of tricaine methanesulfonate (MS 222) . After 

 wrapping in paper toweling to blot the surface and 

 to cover the vent in order to prevent contamina- 

 tion of the sample, the caudal peduncle was rapidly 

 cut off with sharp scissors. Blood which welled 

 from the dorsal vessels was collected in heparin- 

 ized capillaries 75 mm long and one end closed 

 either in flame or with modeling clay. Capillaries 

 were centrifuged in an International Microcapillary 

 Centrifuge Model MB for 5 minutes and hematocrit 

 determined by means of a plastic reader. (The 

 supernatant plasma is used for other tests) . 



The usefulness of the hematocrit deter- 

 mination depends upon its speed, accuracy and 

 close relationship to red cell counts and hemoglobin 

 concentration . In order to perform all three tests 

 simultaneously, blood was collected in a spot test 

 plate depression containing a trace of dried anti- 

 coagulant. The blood was continuously stirred 

 until all needed pipettes and capillaries were filled. 

 Red cell counts were made in the usual manner and 

 hemoglobin was determined with a hemoglobino- 

 meter (Spencer #HB meter No . 1000) . 



Fish selected for these observations were 

 fingerlings and yearlings of eastern brook trout, 

 brown trout (S . trutta) and rainbow trout (S . gaird- 

 neri) fed Cortland diet No. 6 mixed with beef 

 liver and spleen . 



For the purpose of comparison, a separate 

 series of hematocrits were run simultaneously in 



