Smearing 



The liquid was poured off and the tubes 

 inverted on a paper towel to drain for two min- 

 utes . 1.5 mis . of distilled water were added 

 to the centrifugate . In order to disperse the 

 bacteria in the water, the mixture was stirred 

 with a large glass stirring rod and allowed to 

 sit a few minutes for the debris to settle. 0.05 

 ml. of the liquid was then pipetted onto each 

 slide. The liquid was spread evenly encompass- 

 ing an area of 5 sq. cm. This was most easily 

 accomplished by tracing a 25 -cent coin on the 

 working surface, laying the slide over this out- 

 line, and spreading the liquid within the outline 

 with a sterile inoculating loop. An ordinary 25- 

 cent piece has an area extremely close to 5 sq. 

 cm. 



Staining and reading slides 



The Ziehl-Neelsen technique proved sat- 

 isfactory for staining large numbers of slides . 

 Staining with fluorescent dye and observing under 

 low power (20 X objective, 2D X ocular) with 

 fluorescence microscopy is under investigation. 



One square mm . of stained area was ex- 

 amined using the oil immersion objective. It 

 was necessary to calibrate the microscopic field. 

 A pattern of observation was established to cov- 

 er as much of the 5 sq. cm. smear as possible. 

 The use of wide -field eye pieces is unsatisfact- 

 ory as the curvature of field inherent in all oil 

 immersion objectives gives a hazy image at the 

 periphery of the field and many organisms 

 might be undetected. 



Calculations 



It was determined that for each organism 

 seen in a square mm . there are approximately 

 1 million organisms per gram of liver. This 

 figure was derived by the following procedure: 



A measured quantity of Liver was seeded 

 with a known quantity of acid-fast bacilli. The 

 liver was processed and examined by the com- 

 plete digestion technique. The lowest bacterial 

 density per gram of liver that would consistent- 

 ly give 1-10 organisms per sq. mm. of area 

 examined was the final figure used. 



A 257 gram chinook salmon liver was 

 examined and found to be free of acid-fast bacilli 

 by both digest and smear methods. After heat 

 thawing, the Liver was in a pulpy macerated 

 state and could be stirred to give a homogeneous 

 mixture. 100 grams of the thawed, mixed liver 

 was used for the seeding and the remainder was 

 examined for the presence of acid-fast bacilli. 



A grossly infected liver was used for 

 making a suspension of acid-fast bacilli. This 

 liver was digested as outlined above; however, 

 the material was filtered before centrifuging, 

 the centrifugate was then suspended in water, 

 filtered and centrifuged again. A final suspen- 

 sion showed a high concentration of organisms 

 relatively free of liver debris . 



In determining the bacterial density per 

 ml . of the homogenate, a Breed count was em - 

 ployed to give the desired accuracy (4). As this 

 technique depended upon a random distribution, 

 the viscosity of the water had to be increased 

 since the bacteria gathered toward the periphery 

 of the area smeared. The addition of a few drops 

 of blood alleviated this problem . One square mm . 

 of the smear was systematically examined under 

 oil immersion. 



(4)M.F. = 10,000 



10,000 = 820,000 



area of field .0122 sq.mm. 



(M.F.) (average No. /field = No. organisms / ml. 



(820,000) (21)= 17,220,000 



6 ml. of this suspension was final amount seeded. 



(6) (17,220,000) = 103,320,000/100 grams of 

 liver or 1,033,200/ gm. of liver 



RESULTS 



The last 3 columns of table 2 shows the 

 discrepancies between the 2 methods. The first 

 of these three columns gives the number of posi- 

 tives found by smear method not correlating with 

 those of the digest-concentrate method. For ex- 

 ample, of the 4 positive Elokomin chinook liver 

 found by digestion technique, 3 did not have a 

 reciprocal positive slide from the 5 positive smear 

 slides. Thus, from only one liver acid-fast bacilli 



15 



