DIAGNOSIS OF INFECTION WITH B. PULLORUM. 15 



of this work, the serum was divided into three portions, to serve the 

 technicians for independent work. The period of making tests for one 

 technician often required more time to which the serum was submitted 

 to laboratory temperature than that of another, and this in a way perhaps 

 explains the keeping quality of one portion over another. However, 

 one can see from the tests made that comparatively recently drawn sera 

 carefully retained on ice yielded the best results. 



Hens Xos. 315, 49, GO, 22, 77, 4, 35, 618, 46 and 34 never gave a positive 

 reaction during this work, and the work of all three technicians checks 

 in this respect. The serum of Nos. 2096, 52 and 464 gave varying reactions 

 after long retention of serum, but no trouble was experienced by any of the 

 workers in concluding that these birds either were harboring or had har- 

 bored Backrixim pullorvm. 



It may be stated here that under the conditions of the tests, if the test 

 fluid is prepared uniformly, the test carefully carried out, the macroscopic 

 agglutination test for detecting the \arus of Bacterhnn pullorum has proven 

 a good laboratory'' method as handled by three laboratory technicians 

 in this laboratory during the past summer. 



The Ixfluence of Test Fluids of Varying Composition (Mono- 

 valent AND Polyvalent Test Fluids). 



For these tests an experiment was planned in which the serum was used, 

 drawn on the 19th of July. The test fluid was composed of equal quanti- 

 ties of the different strains of Bacterium fullorum used throughout this 

 work. In the first experiment or test a test fluid containing Si was used; 

 in the second, a test fluid containing Sai-in a third, a test fluid containing 

 equal quantities Sj and S2; in a fourth, a test fluid containing equal quan- 

 tities Si, S2 and S3; in a fifth, a test fluid containing equal quantities Si, 

 S2 and S7; in a sixth, a test fluid containing Si, S2, S3 and S.5; in a seventh, 

 a test fluid containing Si, So, S3, S.5 and Se; and in an eighth, a test fluid 

 containing Si, S2, S3, S5, Sg and S7. Various dilutions of the serum were 

 used, the dilutions being made as before. In most cases readings were 

 made after twenty-four, forty-eight and seventy-two hours' incubation. 



An anah'sis of Table 9 shows that the serum from all hens which had 

 previousl}^ agglutinated gave consistent positive results in all the sera. 

 Sera from hens Xos. 2096, 452, 792 and 5 appeared to give better results 

 with a test fluid containing several strains of the organism. Although 

 some of the positive reactors showed good reactions with a monovalent 

 test fluid, yet from the data at hand it may be stated as justified that a 

 test fluid containing several different strains is best suited, under most 

 conditions, in laboratory routine for making the test. Here it should be 

 Loted that none of the birds previously tested and found negative re- 

 acted when their serum was mixed with the test fluids of the various 

 compositions. 



The birds Xos. 58, 59, 62 and 63 were cocks. Of these, 59, 62 and 63 

 gave questionable reactions, and 58 gave a very weak or slight reaction 



