A STUDY OP SEROLOGICAL CROSS REACTIONS BETWEEN THE BRUCELLAE AND CERTAIN SALMONELLAE. WITH SPECIAL REFERENCE TO BRUCELLASALMONELLA PULLORUM CROSS REACTIONS By Irving L. Shklair AN ABSTRACT Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OP PHILOSOPHY Department of Bacteriology and Public Health Year Approved 195>2 Irving .L. snKiair F e l s e n f e l d ejfc al. in 1951 antisera cross (2) reported that Brucella cross reacted with Salmonella pullorum. Such reactions may constitute a source of error in the s e r o l o g i c a l diagnosis of pullorum disease, as well as in the d i a g n o s i s of brucellosis. The present study was under­ t a k e n w i t h a view toward further investigation of the s e r o l o g i c a l cross reaction between S. pullorum and Brucellao R e c i pro cal agglutination tests were carried out with standard strain S. oullorum # 8 9 8 1 7 , intermediate strain S. p u l l o r u m #671, variant strain S. pullorum # B A I , and B r . abortus No S. #2308, Br. melitensis #2500, and Br. suis #1255® cross agglutination reactions were observed between p u l l o r u m #89817 and Brucella. In a number of cases cross r e a c t i o n s were noted between S_. pullorum #671 and Brucella. Cross agglutination reactions were observed between S_, p u l l o r u m #BAI and Brucella. The antigenic structure of S. p u l l o r u m was determined by Edwards and Bruner (1) to be IX, XII-^, (Xllg), XII^o the X I X g antigen, Prom the above results it appears that (found in the intermediate and variant ' strains of S. pullorum) or a part thereof, was responsible for t h e observed cross reactions. Reciprocal agglutination tests were also carried out with B r u c e l l a and other organisms, Salmonella reading and Proteus cross (Gwatkin), which contain the XIl£ antigen. The reactions observed between these organisms and B r u c e l l a were less pronounced than those noted between Irving L. Shklair S. pullorum #BAI and Brucella, There appears to be a quan­ titative difference in the amount of the cross reacting antigen (for Brucella) in S. pullorum #BAI, S. reading, and Proteus (Gwatkin). Whether this is a strain characteristic of the organisms involved is not known. M o n o - 3pecific sera containing agglutinins IX, XII-^, XII g, and XII were prepared. Only the XIIp mono-specific sera 3 produced cross agglutinations with the Brucella antigens. It was noted that only a small fraction of the XII^ antigen was involved in these cross reactions. Agglutinin absorption studies carried out with Si. pullorum #BAI and Brucella indicated that the antigen common to _S, pullorum #B A I , £>. reading, and Proteus (Gwatkin) existed in approximately equal amounts in all three species of Brucella tested. One hundred and seventy-six S. pullorum suspicious and reactor turkeys were tested using B r . abortus #2308 and Br, melitensis #2500 antigens in dilutions of 1:25, 1:50, and 1:100, Of this total, 29 (seventeen percent) showed a cross reaction with the Brucella antigens. A number of salmonellae, pathogenic for man and animals, contain antigen XIIp, These organisms, if they contain the antigenic factor common for Brucella, may produce agglutinins in sufficient quantity to cause false positive agglutination tests for brucellosis. Irving L. Shklair Edwards, P. R. and D. ~W. Bruner. Form variation in Salmonella nullorum and its relation to X. strains. Cornell Vet‘. 32:3l8-32lj., 19l(i>o Felsenfeld, 0., V. M. Young, E. Loeffler, S. J . Ishichara, and V„ F. Schroeder. A study of the nature of brucellosis in chickens. Am. J, Vet. Res« 7 1951 o ACKNOWLEDGEMENT The author wishes to express his sincere appreciation to Dr. H. J. Stafseth for his encouragement, interest, and valuable advice throughout this study. The author wishes further to express his appreciation to the many members of the faculty, staff, and graduate students who gave freely of their time, advice, and equipment during this study. TABLE OP CONTENTS Page INTRODUCTION ............................................ REVIEW OP L I T E R A T U R E .............. 1 3 MATERIALS AND METHODS............... . ................... 13 RESULTS................................... , 2k D I S C U S S I O N ............................................... 63 SUMMARY AND CONCLUSIONS. ........................ 69 LITERATURE CITED ....................................... 71 INTRODUCTION The incidence of pullorum disease in poultry flocks has steadily decreased during the last few years. However, it is still widespread and potentially very destructive. Decrease in the incidence of pullorum disease can be attributed to improved methods of detection and removal of infected breeder stock, plus improved sanitary practices in poultry management. Serological methods are mainly used for the detection of carrier birds. The National Poultry Improvement Plan of 191*1 (51*-) endorses the following serological methods: 1 , the standard tube agglutination test; 2 . the stained antigen, rapid, whole-blood test; and 3 * the rapid serum test, "The choice of method is determined by many factors and objectives peculiar to a state or region. states all three methods may be employed, In many but the whole- blood test is the one most generally used," (5 5 )• The diagnostic tests listed above are based on the well-known principle that serum antibodies react specific­ ally with their corresponding antigens. However, it should be rememoered that antigens may be shared by several bacterial species which may or may not be otherwise closely related. This has caused confusion in the campaign against pullorum 2 Recently, Felsenfeld, Young, Loeffler, Ishichara, and Schroeder (21) reported that Brucella antisera produced cross reactions with Salmonella pullorum. Such cross reactions may constitute a source of error in the serologi­ cal diagnosis of pullorum disease, as well as in the diagnosis of brucellosis. The present study was undertaken with a view toward further investigation of the serological cross reactions between S. pullorum and Brucella. Serological cross reactions were performed with standard, variant, and intermediate strains of S. pullorum, and Brucella abortus, Brucella melitensis, and Brucella suis. Mono-specific S. pullorum antisera were utilized in these studies. Agglutination absorption studies were also carried out xvith the above organisms in order to determine the amount of the common antigen or antigens present in them. Cross agglutination studies employing turkeys naturally infected with S. pullorum were also carried out. REVIEW OP LITERATURE In 1900 Rettger (l|.5) reported, the discovery of the etiological agent of pullorum disease. The disease was first described as a "fatal septicemia of young chicks". In 1909 Rettger (lj-6) called the disease "white diarrhea". Later in the same year Rettger and Stoneburn (JL4.7) suggested that the name "bacillary white diarrhea" be given to the disease and that the causitive agent be known as Bacterium pullorum. In 1925 the name of the organism was changed to Salmonella pullorum according to Bergeys Classification. In 1928 the common name of the disease was changed to "pullorum disease". In 1 9 1 I4. Rettger, Kirkpatrick, and Jones (I4.9 ) described the complete cycle of pullorum Infection in chickens. These investigators stated that to combat the disease success-' fully the infection cycle had to be broken. The most feasible method of breaking the infection cycle was to detect and remove pullorum carriers. An early attempt to detect carriers was made by bacteriological examination of fresh eggs from infected flocks. This method was found to be inadequate and i m ­ practical in eliminating infected birds (55). Jones in 1913 (35) reported the use of the macroscopic tube agglutination test as a means of detecting carriers and recommended that serum dilutions of 1 ;5 0 , 1:100, and k 1 : 2 0 0 be used for routine testing. Rettger, Kirkpatrick, and Jones (50) and Gage, Page, and Hyland (23) confirmed the work of Jones. Gage et_ a l . stated "The macroscopic agglutination test proved to be a good laboratory method for the detection of adult hens harboring Bacterium pullorum." Runnells, Coon, Parley, and Thorp (5l) in 1928 develop­ ed the rapid serum test for the detection of pullorum disease carriers. Two serum-antigen dilutions corresponding to the 1 : 5 0 and 1 : 1 0 0 dilutions of the tube agglutination test are employed. "Positive reactions may occur quickly but delayed reactions may require several minutes. Gradations of reactions occur in this method as in other methods." (5 5 )• Schaffer, McDonald, Hall and Bunyea (52) and Coburn and Stafseth (10) reported the development of the wholeblood method in which a concentrated stained antigen is employed. In view of its apparent simplicity it is now the most widely used test for the eradication of pullorum carriers. Other diagnostic tests have been proposed, the intradermal test (57)j a precipitin test (16), and the comple­ ment fixation test (1 6 ), for the detection of carriers. These methods were found to be either unreliable or im­ practical in control and eradication programs. The question has often been asked why repeated blood testing of flocks has failed to bring about the elimination of pullorum disease. Sometimes one can find management 5 practices which are at fault. In other cases, there are certain factors that tend to interfere with the effectiveapplication of the agglutination tests employed in the diagnosis of pullorum disease in birds. Horton in 1916 (2 9 ) was the first to observe that, occasionally, the sera of birds having pullorum disease failed to agglutinate pullorum antigens. Halpin, (38), This was subsequently observed by Beach, and Lampman (1), Doyle (12), Kaupp and Dearstyne and others. nomenon, Whether this was due to a prozone phe­ or as a result of the organism undergoing form variation, or whether it was due to some unknown physio- „ logical factors in the birds is not known. Hinshaw, Jones, Harr, Hall, and Neimeyer (27) and Bunyea, and Dorset (I4.) pointed out that the whole-blood test was equivalent to a 1 : 5 0 dilution in the tube agglutination test, and that a large number of reactors had only 1:25 and 1:50» titers of It is therefore possible that a large number of birds that were negative to the whole-blood test might be reactors in the 1 : 2 5 dilution of the tube test and thus missed. a comparative Corpron, Bevins, and Stafseth (11) made study of the tube agglutination test and rapid whole-blood method applied in testing turkeys and reported that the tube agglutination method was more sensi­ tive and more consistent in reaction than the whole-blood methodo The reports of various Canadian workers concerning the presence of variant strains of S. pullorum offered another possible explanation concerning the failure of serological tests to eliminate pullorum disease. Younie in 191+1 (62) reported that in 1939 a few outbreaks of pullorum disease occurred in chicks hatched from breeder flocks which were negative to pullorum disease tube tests. In 191+0 an increase in the number of such outbreaks x^as noticed. Only a few reactors were found on retesting the flocks. Neither the post-mortem appearance of affected birds nor the cultural characteristics of the organisms isolated from them differed from the usual findings in pullorum disease. certain serological differences were noted. However, In the tube agglutination test, the variant S. pullorum antisera did not react with most of the standard pullorum antigens, but did react with variant■antigens in dilutions of 1 :8 0 0 . In 191+6 Edwards and Bruner (lip) determined the anti­ genic components of £3, pullorum. The antigenic structure of S. pullorum was found to be IX', XII , (Xllg), XII^. The brackets around the X I I 2 indicate that it is subject to form variation. Edwards and Bruner found that the variant strains of S. pullorum contained much XII 2 and very little, if any, XII-j. The standard strains contained little, if any, X II 2 and varying amounts of XII . A number of cultures have been isolated which are fairly well balanced in XIIg and XII^ antigens. Wright and Edwards (6 1 ) desig­ nated this type of culture as ’'Intermediate11. When an intermediate culture becomes stabilized with either the XII 2 or XII^ antigen predominating, the culture becomes either a variant ( X I ^ ) or standard (XII^) strain* When testing for either type of S. pullorum it is possible for one type to escape detection if an antigen is prepared from the opposite type. This has resulted in the appearance of a variety of antigens on the market. The tester now has at his disposal antigens prepared from the standard or variant strains alone as well as polyvalent antigen composed of a mixture of standard and variant strains. Since it is not practical to employ double testing the polyvalent antigen has found favor where both standard and variant types of pullorum disease are encountered. Recently, Wright (59, 60) employed a stabilized 'intermediate1 type culture in the preparation of a pullorum antigen. He used this antigen successfully for the detection of carrier bi rd s. Another factor that has to be taken into consideration in the effective application of the pullorum tost is the problem of suspicious and ’’pin-point" reactors. In this category are included those birds whose blood gives a late, very fine agglutination in the whole-blood plate test and a fine, easily dispersed, often low titered reaction in the macroscopic tube test. As Williams pointed out (58) "the reactors may be encountered in certain flocks from year to year or may be demonstrated for only a short period of time. In any event they present a definite problem, especially when they are encountered in great numbers." It is now recognized that several factors m ay be involved in these atypical reactions. Studies carried out so far suggest that cross reacting agglutinins for S_. pullorum occur in low dilutions, or, occasionally, titers in the blood of such birds. in high It should also be pointed out that organisms possessing antigens common for S. pullorum can often produce typical agglutination reactions. These typical reactions may be caused by salmonellae other than ^3. pullorum or by organisms morphologically, culturally and biochemically unrelated to S. pullorum. It has been only recently that the agglutination of coliform bacilli by Salmonella sera has been examined more closely from the standpoint of antigenic analysis (lp3) • In most instances the serological relationships observed between coliform and Salmonella strains were due to common heat-stable, somatic 0 antigens. Hobs and Arjona (28) described a culture which contained a portion of antigen XII of the Kauffmann-Mhite classification. stein, and Welker, (1 7 )} and others, Braun, Silber- cited by Peluffo, Edwards, and Bruner described many coliform cultures which contain group Salmonella somatic antigens. Johnson and Anderson in 1936 (33) isolated a lactobacillus that agglutinated with S. pullorum antisera in low dilutions. The organism, at the time of isolation, for lack of complete identification was designated as PI. Later, in 191-1-0, Johnson and Pollard (3lj-) suggested the name Lactobacillus meleagrides for this organism. They also pointed out that Lactobacillus casei could cross agglutinate with £>. pullorum anti sera, Gwatkin (26) reported a strain of Proteus isolated from a hen whose antiserum reacted to variant, but not to regular or standard pullorum antigens. The organism is very rich in XllgJ it has been found to be useful in differen­ tiating between the standard and variant strains of _S, pullorumo Edwards, Bruner, Doll, and Hermann (15) isolated a culture of Staphylococcus that reacted with Xllg antisera 0 However, repeated intravenous injections of dead and living cultures into chickens failed to give rise to any agglutinins of XII 2 cultures of S. pullorum. They doubted if this organism played any part in false positive reactions in the diagnosis of pullorum disease. In 191-1-6, Garrard, McDermott, Burton, and Carpenter (2!)., 25) carried out post-mortem studies on 87 fowl exhibiting non-specific pullorum reactions. many strains of staphylococci, coliforms, They isolated and enterococci. These organisms were classified as followsr A. Gram positive cocci — Micrococci, staphylococci, and enterococci Bo Coliform group Co Miscellaneous group -- Proteus, Alcaligines, and other unidentified genera -- Escherichia coli, Aerobacter aerogenes, and some inter­ mediate types "The majority of the cocci were isolated from the ovaries and liver, while most of the coliform types were found in the intestine. Many representatives of each group 10 gave non-specific reactions with pullorum and polyvalent Salmonella sera.” (25)• An enterococcus isolated from the liver of a hen was inoculated intravenously into a group of 20 White and Barred Rock fowl. Agglutination tests were run using both standard and variant S. pullorum antigens. The majority of reactions were observed with the variant type antigen. In some cases titers as high as 1 r6)4.0 and 1 : 1 2 8 0 were recorded with the variant S. pullorum antigen. Further tests showed that the XII 0 antigen was involved, as absorption of the XII 5 Cantibody produced negative results with Proteins (Gwatkin), Salmonella reading, and standard and variant S. pullorum antigens. Antigens IX and XII^ could not be detected by colony typing (6 ). Burton and Garrard (5) carried out additional experi­ ments on the organisms mentioned above. They inoculated two Golobactrum, and three Paracolobactrum into pullets 0 All the inoculated pullets reacted with either the standard or variant pullorum antigens. Absorption tests showed that antigen Xllg was common to all organisms except Paracolon intermedium which possessed antigen XIT^. P. intermedium produced titers as high as 1 : 2 5 6 0 when tested against standard pullorum antigen. "The fact that representatives of the enterococci and coliform groups of organisms isolated from various organs of non-pullorum reacting fowl have been shown to cause cross reactions with pullorum antigen does not mean that their presence is the complete answer to non-pullorum reactions. There is sufficient evidence to suspect, how­ ever, that organisms common to the intestinal’content of fowl are more commonly being found in other organs, where they cause low-grade infections and induce the production of agglutinins strong enough to cause cross reactions with pullorum antigen." (5>) . In 195>1 Felsenfeld e_fc al. (21) reported that Brucella antiserum produced a cross agglutination with S5. pullorum. They did not specify whether S. pullorum standard, variant or intermediate, strain was used. It is now known that Brucella will cross agglutinate with a number of unrelated organisms. Foshey (22) pointed out "It is apparently not widely recognized that anti­ brucella serum may have very broad agglutinating properties for other genera and families. that we have made. The antibrucella serums . . . agglutinate an extraordinarily large number of different bacteria, most of the colontyphoid dysentery group, and some Salmonella, some Aerogenes, almost all strains of Proteus and Alkalegenes, B. tularense, and even H. influenzae." Foshey, unfortunately, did not specifically mention any of the salmonellae that cross agglutinated, and it is not known whether S.. pullorum was one of the organisms tested. Cioglia (8 , 9) described common H and 0 antigens in Brucella melitensis and a number of salmonellae. However, no mention was made of cross agglutination reactions between the Brucella group and. S. pullorum. Huddleson (30), Stafseth (53)> and others have reported that the first cases of brucellosis in chickens ijere observed by Fiorentini in Italy in 1907. Evidence of Brucella in­ fection was based on the fact that 55 percent of the birds reacted positively to the brucella agglutination test, and that B r . melitensis was isolated from the spleen of those that were ill. Zweck and Zeller in 1913 (63) injected Brucella abortus into chickens in an attempt to produce the disease. The- birds showed no evidence of the disease other than the production of agglutinins. Many other investigators have shown that chickens can produce agglutinins against the Brucella group of organisms. An excellent review of the history of brucellosis in fowl will be found in Biester and Schwarte, Diseases of Poultry by Stafseth (53)« MATERIALS AND METHODS This section is divided, into four groups of experi­ ments. Experiment I deals with the agglutination reactions between £>. pullorum and the Brucella species. Experiment II covers the preparation and agglutination reactions of monospecific sera. Experiment III covers agglutinin-absorption studies using S. pullorum, variant strain, and Brucella. Experiment IV includes the cross agglutination studies on turkeys naturally Infected with S. pullorum. Experiment I Specific and cross agglutination reactions between 3. pullorum and Brucella, a. Cultures Salmonella pullorum cultures pullorum #89817 standard strain Somatic structure IX, XII-^, XII^ S. pullorum # 6 7 1 Intermediate IX, XII S. pullorum #BAI variant IX, XII^, Xllg (XII^), XII^ S. pullorum #8 98 1 7 and #671 were isolated at the Michigan State College Poultry Pathology Laboratory. S. pullorum #3AI xtfas obtained from the Bureau of Animal Industry. Prom this original culture a number of sub-cultures were prepared and lyophilized. The culture used in this experiment is one of the lyophilized sub-cultures. 1^ Brucella cultures Br. abortus #2308 Br. melitensls #2500 Br. suis #1255 B r » abortus #2308 and. Br. melitensls #2500 were obtained from the Brucella laboratory at Michigan State College. Br. suis #1255 was obtained from the stock culture collection of the Department of Bacteriology and Public Health at Michigan State College. All the cultures used were in the smooth phase. b. Groups of animals White Leghorn and White Barred Rock roosters and pullets, 3 to 8 months old, were used in this experiment. All the birds were negative for _S. pullorum and Brucella agglutinins. Group 1: 6 birds Inoculated with S. pullorum #89817 Group 2: 8 birds Inoculated with S. pullorum #671 Group 3: 7 birds Inoculated with s. pullorum #BAI Group Lp: 8 birds Inoculated with Br . abortus #2308 Group 5: 10 birds Group 6: c. 6 birds Inoculated with Br . melitensls #2500 Inoculated with Br . suis #1255 Antigen preparation S. oullorum # 8 9 8 1 7 , #671, and #BAI antigens were pre­ pared according to the directions set forth by the Livestock Sanitary Association in 1932 (Ml). Other S. pullorum anti­ gens used in this experiment included the University of New Hampshire standard antigen, pullorum stained antigen K, regular and polyvalent, (Lederle) and Redigen regular and polyvalent antigen (Columbus Vaccine Co,). B r . abortus #2308, Br. melitensis #2500, and Br. suis # 1 2 £ 5> antigens were prepared using smooth colonies of the above organisms according to the directions of Huddleson (30 ) 0 d. Preparation of antisera The S. pullorum cultures listed above were grown on nutrient agar slants for 2ip hours. The growth was removed with sterile saline and the suspension was prepared, the turbidity corresponding to a reading of Ip to 6 on the McFarland nephelometer scale. The Brucella were grown on tryptose agar slants for lp8 to 72 hours. The organisms were removed with sterile saline, the concentration corresponding to a reading of ip to 6 on the McFarland nephelometer scale. The birds were injected intramuscularly (pectoral muscles) with 1 to 2 ml of the above concentrations. Three to four injections were given to each bird. injections were given a week apart. The Seven to ten days after the last injection 10 to 20 ml of blood was removed from each animal by cardiac puncture. The clotted blood was centrifuged and the serum removed, e. Agglutination tests Tube agglutination tests were set up in serial dilutions of 1:20 to 1:5>160. The tests were incubated for 2ip hours 16 at 37° C and then read. Reciprocal agglutination tests were set up between the organisms listed previously. The S. pullorum antigens were adjusted to a pH of 8.2. The Brucella antigens were adjusted to a pH of 7*6. This figure was used as it closely approximated the pH of chicken blood serum, and it was believed that the most uniform results could be obtained using this pH. Experiment II Mono-specific sera were prepared to aid in finding the factor or factors responsible for the cross agglutination between S. pullorum and Brucella. a. Cultures Somatic antigens Salmonella paratyphi A, var. durazzo II, x n 1 , x n 3 Salmonella reading IV, XI1^ Proteus (Gwatkin) XII2 .... xn2 Br. abortus #2308 B r . melitensls #2500 B r . suis #1255 £3. paratyphi A, var. durazzo, S. reading, and Proteus (Gwatkin) cultures were supplied by Dr. W. W. Fergeson of the Michigan Department of Health. b. Groups of animals Group VII: 2 birds Injected with _S, paratyphi A, var. durazzo Group VTII: 2 birds Injected with S. reading Group IX: Injected with Proteus (Gwatkin) 3 birds 1 17 Birds from Groups 1, I4., 5> and 6 (Experiment I) were also used in this experiment. The birds in Groups VII, VIII, and IX consisted of White Leghorn roosters approxi­ mately 8 months old. All the birds used were negative for Brucella and S. pullorum agglutinins, Co Preparation of somatic antigens lo Somatic antigen preparation of S. paratyphi A, var. durazzo, S. reading, and Proteus (Gwatkin). a. A saline suspension of the organisms was seeded over tryptose agar in 1 6 -oz bottles, b. The bottles were incubated at 37° C for 2l\. to 36 hours. c. A small amount of sterile saline was added to each bottle and the growth was suspended by gentle rocking, d. The suspensions were pooled and 3 volumes of 95 percent ethyl alcohol were added. The suspensions were o incubated in a 37 C waterbath for 1 to 2 hours when p r e ­ cipitation should be complete.e. The supernatant was decanted and discarded. The remaining precipitate was centrifuged at high speed for 30 minutes, f. The supernatant was decanted and discarded. The sediment was resuspended in phenolized saline (0 , 3 percent phenol). g. About 7 ml of phenolized saline was added to 1 ml of packed cells. This was the stock somatic antigen. 18 2. Br. abortus #2308, 3r. melitensls #25>00, and Br, . suls # 1 2 55 antigens were prepared according to the directions of Huddleson (3 0 ). 3. The S. pullorum antigens were prepared as described in Experiment I . d. Preparation of antisera The preparation of S. pullorum and Brucella antisera has been described in Experiment I. Somatic agglutinating sera for S. paratyphi A, var. durazzo, S. reading, and Proteus (Gwatkin), were prepared by injecting boiled saline suspensions of the above organisms into White Leghorn roosters (Groups VII, VIII, and IX). A series of intramuscular injections was given until a suit­ able titer was obtained. Approximately £ days after the last injection the birds were bled by cardiac puncture. The clotted blood was centrifuged and the serum removed. e„ Preparation of mono-specific sera Mono-specific serum Antisera IX S. pullorum #898 17 IX, XII1 , XII^ Absorbing antigen S. paratyphi A, var. cTurazzo II, X I I ^ JIT XII t S. reading IV, XII-^ XII 2 Proteus (Gwatkin) XII 2 ..... XII^ S. reading IV, x n 1, XII 2 S. pullorum #89817 IX, XII-j^, x n 3 XII 3 S. paratyphi A, var. durazzo S. II, IV, x n x, x n 2 x n 1, x n 3 reading 19 A heavy suspension of the antigen was placed in a test tube and centrifuged. carded. The supernatant fluid was dis­ A 1:10 dilution of the antiserum was added to the packed cell suspension and thoroughly mixed. was incubated at 37° C for 2 hours. The mixture It was then centri­ fuged and the partially absorbed serum removed. The absorptions were repeated two more times which was sufficient to produce the mono-specific serum desired. f. Agglutination tests Reciprocal agglutination tests were carried out with S.. paratyphi A, var. durazzo, S_.- reading, a n d Proteus (Gwatkin). Agglutination tests were also carried out using the above sera against the S. pullorum and Brucella anti­ gens. Agglutination tests were then carried out with the mono-specific sera and the above organisms. Agglutination tests using mono-specific sera (diluted 1 :1 0 ) were set up as follows: 0 , 5 ml of antigen (diluted to a reading corresponding to 1 on the McFarland nephelometer scale) Tiras added to each tube. To the initial tube 0.5 ml of the mono-specific serum was added. dilution of 1:20. This produced a From the first tube 0.5 ml of the 1:20 dilution was transferred to the second tube. dilutions were carried out to 1:5120, and serum controls were prepared. Serial Suitable antigen 20 Experiment III Agglutinin absorption studies. Mirror absorption tests were carried out between Si. pullorum #BAI and the Brucella species in order to determine the amount of cross reacting antigen or antigens that were present. a. Cultures The Brucellaj S_. pullorum # B A I , S. reading, and Proteus (Gwatkin) cultures were described in Experiments I and II. b. Antisera preparation The preparation of Brucella and S. pullorum #3AI anti sera was described in Experiment I. c. Preparation of antigens Brucella, S. pullorum # B A I , S_. reading, and Proteus (Gwatkin) antigens were prepared as described in Experi­ ments I and II. do Absorption studies B r . abortus #2308, Br, melitensis #2500, Br. suis #1255, and S. pullorum #BAT., were absorbed with their corresponding homologous antigens as well as with each other. The absorption techniques used in this experiment were the same as those used in the preparation of monospecific sera (Experiment II). Preabsorption titers were first determined. Agglutin­ ation tests were then carried out between the absorbed sera and the antigens listed previously. m m m Agglutination tests were carried out as described in Experiment II. Experiment IV Gross agglutination studies of naturally infected turkeys with S. pullorum. Blood samples from Turkeys were sent to the Poultry Pathology Laboratory at Michigan State College for the serological diagnosis of pullorum disease. Sera that were positive or suspicious to the tube agglutin­ ation test were tested against Br. abortus #2308 and Br, melitensis # 2^00 antigens in dilutions of 1 :2 5 * 1 :5 0 , and 1:100. To rule out the possibility of any natural Brucella infections in turkeys, (which would give erroneous cross agglutination results) a modification of Castaneda's (7) 'filter paper surface fixation test' (plate I) was used. This test was found to be specific for Brucella agglutinins and will not produce a positive reaction unless they are present 0 a. Preparation of Brucella surface fixation antigen Stock antigens of Br, abortus #2308 and Br. melitensis # 2 50 0 were diluted with 0 , 5 percent phenolized saline to an equal volume. A 1 percent aqueous solution of crystal violet was added to the suspension to produce a final con­ centration of 3 ml of dye to 100 ml of the diluted antigen. b. procedure of filter paper surface fixation test 1. Place a sheet of filter paper, Eaton-Dikeman Wo. 609 or Schleicher & Schuell No. 589, or other comparable paper, 22 on a test tube rack. 2. Place a drop of the test serum on the filter paper using a Lp—Lp.^ m m loop, 3. Using a 2 mm loop immediately place one loopful of the stained antigen in the center of the serum drop. Ip. Immediately place two loopfuls of saline on top of the antigen using the 2 mm loop, 5. If the serum contains 'Brucella agglutinins the antigen will remain in the center of the drop, if no Brucella agglutinins are present in the serum the antigen will spread through the drop to color the whole area (Plate Positive Negative / / \ _ Plate I * / Surface 'fixation test RESULTS I n Experiment I reciprocal agglutination tests were carried out between S. pullorum # 8 9 8 1 7 , # 6 7 1 , # B A I , and B r . abortus #2308, Br. melitensis #2500, and Br. suis #1255® S. pullorum and Brucella agglutinated their homologous antigens, Tables 1-6. No reciprocal cross agglutination reactions were observed between £>. pullorum #89817 (standard strain) and Br. abortus #2308, Br. melitensis #2500, and B r . suis #1255, Tables 10-12. A cross agglutination was observed, in a few cases, when S. pullorum #671 (intermediate strain) was tested against the Brucella antigens, Table 8, In a few instances, cross reactions were observed when Brucella antisera were crossed with S_. pullorum #671 and S. pullorum polyvalent antigens, Tables 1 0 - 1 2 0 Cross agglutinations were observed when S. pullorum #BAI (variant strain) was crossed wit h the Brucella antigens. Cross reactions were also observed when Brucella antisera were tested against S. pullorum #BAI antigen, Tables 9-12. In a number of cases, however, no cross agglutinations or only atypical agglutinations were observed between some of the Brucella antisera and the S. pullorum polyvalent anti­ gens. Why the results were not uniform is not known, Tables 10-12. To get a better understanding of the above results the antigenic structure of S. pullorum should be recalled. See pages 6 and 7» It was pointed out previously that no cross agglutin­ ation was observed between the standard strain S. pullorum #89817 and B r u c e l l a . Reciprocal cross agglutinations were observed between the variant strain S. pullorum #BAI and Brucella, while in a few cases the intermediate strain of S. pullorum # 6 7 1 cross reacted with Brucella. Prom the above results it appears that the antigen common to S. pullorum and Brucella is the Xllg antigen or a part thereof« The cross reactions between S. pullorum # 6 7 1 mediate strain) and Brucella are interesting. (inter­ When £5. pullorum #671 was first typed (single colony typing) approxi­ mately 18 percent (12 of the 75> colonies tested showed XII 2 activity) of the colonies showed evidence of the XII^ antigen. Over a period of six months the number of colonies showing XII 2 activity has diminished to the point where now less than 1 percent of the colonies tested show any XII^ activity. This could explain why cross reactions between this organism and Brucella were often erratic. It also indicates that S_. pullorum # 671 was undergoing form variation Form variation in the Salmonella was first described by Kauffmann in 191+0 (36). He described a variation in antigen I (i.e. "I-Formenwechsel") . In I 9I4I Kauffmann (37) described "a variation in antigen XII fundamentally similar to that of I variation. Within the same strain there occur 26 colonies with a well-developed antigen XII (form ++), moderately well developed antigen XII (form +) and a weakly developed antigen XII (form +) which forms dis­ sociate each other.” In determining the antigenic structure of S. pullorum Edwards and Bruner (lip) found that S. pullorum underwent antigen XII form variation. They observed, as did Kauffmann (3 6 ), that it was the X I I 2 antigen that was responsible for the form variation in antigen XII. did not undergo form variation. pointed out ”It seems, then, Antigens XII^ and XII^ As Edwards and Bruner that S. pullorum is subject to form variation which involves antigen XII2• cultures this variation occurs continuously, X I I2 ++ and XII + colonies can be found. I n normal so that the However, it is possible for the organisms to become stabilized in the + or ++ form, strains. thus giving rise to ” standard” and "variant” According to this view, the variant strains are not a special strain of S_. pullorum but can arise from any culture of the type by stabilization in the ++ form.” Experiment II was carried out further to test the hypothesis that the XII2 antigen, or a part thereof, was responsible for the cross agglutination between S. pullorum and Brucella. Cross agglutination studies were also carried out between Brucella and other organisms, S. reading, and Proteus (Gwatkin), which were rich in XII2 antigen. A cross agglutination was observed between S. reading, Proteus (Gwatkin), and Br. abortus # 2308, Br. melitensis # 2^00, and Br. suis #1255. Although the organisms were rich in XII 2 antigen the cross reactions observed were not as pro­ nounced as those between S. pullorum #BAI and Brucella, Tables 13 and l5. It is not known if this is a strain characteristic of the organisms involved. Mono-specific antisera containing IX, XII-^, XII^, and XII^ agglutinins were prepared to rule out the possibility of the other antigenic components being responsible for the cross agglutination. No agglutinations were observed when mono-specific sera containing agglutinins IX, XII^, and XII^ were tested against Br. abortus #2308, Br. melitensis #2500 and B r . suis #1255 antigens. However, a cross agglutination was observed between the Xllg sera and Brucella, Table 11]., In this experiment the Xllg mono-specific serum was prepared from S. reading anti serum. Although the XII agglutinins were present in abundance, only a small fraction of this agglutinin cross agglutinated with the Brucella antigens. Prom the above results one can assume that a portion of the XII 2 antigen is responsible for the cross agglutin­ ation reactions studied. In all probability, the amount of the common antigen (for Brucella) present in S. reading, Proteus (Gwatkin), S. pullorum variant and intermediate strains, will vary with each culture. are examined, When other bacteria it may be found that organisms which contain a great deal of the XIl£ antigen will not cross agglutinate 28 with Brucella, and, on the other hand, organisms may be found with a limited amount of X I I 2 antigen which will cross react to a high degree. In Experiment III reciprocal agglutination tests were carried out between S. pullorum #3AI, Br. abortus #2308, Br. melitensis #2500, and Br. suis #12p5» Antisera of the above organisms were absorbed with their homologous and heterologous antigens. Agglutination tests were carried out using the absorbed sera against the above organisms and also against S. reading, and Proteus antigens. (Gwatkin) This was done in an attempt to determine the amount of cross reacting antigens which were present in the above organisms. The results of the reciprocal agglutination tests (preabsorption titers) between Brucella, S. pullorum 7 #BAI, S. reading, and Proteus (Gwatkin) can be found in Table 15>, Br. abortus #2308, Br. melitensis #2^00, and Br, suis # 125>5> antisera were absorbed with their homologous and heterologous antigens. Agglutination tests were carried out using the absorbed sera against their homologous and heterologous antigens, observed. Bo significant agglutinations were When the above absorbed sera were tested against S. pullorum #8AI, S. reading, and Proteus (Gwatkin) antigens no cross agglutination was observed, Tables 16-18, 2l[_-26o 20-22, and When Br. abortus #2308, Br. melitensis #2500, and Br. suis #1255 antisera were absorbed with 3. pullorum #BAI 29 antigen, noted. a uniform lowering of the antibody titer was The lowering of the antibody titer was less notice­ able when tested against S. reading and Proteus (Gwatkin) antigens, Tables 19, 23, and 27. This indicates that the antigen in Brucella, common to S. pullorum # BAT., SU reading, and- Proteus (Gwatkin), exists in equal amounts in all three species of Brucella tested and it can be removed from each by its homologous and/or heterologous antigen. S. pullorum #BAI anti serum was absorbed with its homolo­ gous antigen. No agglutination was observed 'when the absorbed serum was tested against its homologous and various Brucella antigens, Table 23. However, w h e n S. pullorum #BAI antiserum was absorbed with Br. abortus #2303, Br. m e l i ­ tensis # 2 5 0 0 , and Er. suis #1235 cells a uniform lowering of the antibody titer was observed, Tables 29-31. It would appear from the above results that the antigen common between S_, pullorum #BAI and Br, abortus #2308, B r . melitensis #2500, antigen. and Br. suis #1255 is a part of the Xllg There appears to be an equal amount of the common antigen in the species of Brucella tested. indication, however, There is an that the amount of the cross reacting antigen in S. reading and Proteus (Gwatkin) that which appears in S. pullorum # B A I . differs from Whether this is a strain characteristic of the organism used was not determined. A n interesting sidelight in the work of McCullough, Eisele, and Beal (IpO) in their study of the antigenic re­ lationship between Vibrio comma and Brucella may be mentioned. 30 They found that the H antigen of V. comma produced the most pronounced cross agglutination reactions with Br. abortus, while insignificant reactions were noted with Br. suis* and no cross reactions were observed when tested against Br. melitensis. Felsenfeld _et al. (21) reported that V. cholerae can cross agglutinate with S. pullorum. However, it seems unlikely that the common antigen found in the Xllg antigen is responsible for the cross agglutination reactions between V. cholerae and 3^ pullorum. In Experiment IV an attempt was made to determine the number of birds naturally infected with S_. pullorum that would produce a cross agglutination with Brucella antigens , One hundred and seventy-six £>. pullorum suspicious and/or reactor turkeys were tested against Br. abortus #2308 and Br. melitensis #2^00 antigens in dilutions of 1:25, l:f?0, and 1:100. Of this total 29 (seventeen percent) showed a cross reaction with the Brucella antigens. In all cases the sera cross agglutinated both Brucella antigens. There was no evidence of Brucella agglutinins, as shown by a negative filter paper surface fixation test, in any of the birds tested. The reason why this figure appears small may be explained as follows: there may have been only a few reactor birds which harbored the XII 2 antigen; the titers of the birds m a y have been sufficiently high to produce a positive S, pullorum. agglutination response (1:20), but not 31 high enough to produce a cross agglutination with the Brucella antigens; it should also be borne in mind that these cross reactions may not have been due to the presence of S. pullorum agglutinins. It is known, as Poshey pointed out (22), that a great many organisms, unrelated to Brucella, may cross react with Brucella. It is entirely possible that many of the positive agglutination reactions, on the basis of which diagnoses of brucellosis were made, were not caused by homologous brucella agglutinins but by group agglutinins„ TABLE 1 EXPERIMENT I AGGLUTINATION RESPONSE OP S. PULLORUM #89817 (STANDARD STRAIN) ANTI SERUM WHEN TESTED AGAINST VARIOUS S. PULLORUM ANTIGENS S. pullorum # 8 9 8 1 7 Antisera 852 896 810 S. pullorum # 8 9 8 1 7 320 160 1280 6 S. pullorum #671 320 160 New Hampshire std. antigen 320 K antigen regular K antigen polyvalent Antigens Redigen regular Redigen polyvalent 813 800 888 br0 2560 1280 1280 6 I 4 .O 1280 1280 160 1280 6ip0 2560 2560 + + + + + + + + + + + + + + + + + + + + + + + + Figures indicate highest dilution at which agglutination occurred. 33 TABLE 2 EXPERIMENT I AGGLUTINATION RESPONSE OP S. PULLORUM #671 (INTERMEDIATE STRAIN) ANTISERUM WHEN TESTED AGAINST VARIOUS S* PULLGRUM ANTIGENS S. pull or tun #6?1 Anti sera Antigens 898 812 81+9 833 881). 828 830 831+ S. pullorum #671 61+0 1280 61+0 1280 1280 61+0 6110 320 So pullorum #89817 61+0 1280 61+0 1280 1280 61+0 320 320 New Hampshire antigen 61+0 1280 61+0 1280 1280 6LpO 61+0 320 K antigen regular + K antigen polyvalent + Redigen regular + Redigen polyvalent + + + + + + + + + + + + + + + + •* + 4 Figures indicate highest dilution at i.tfhich agglutination occurred* 3k TABLE 3 EXPERIMENT I AGGLUTINATION RESPONSE OP S. PULLORUM #BAI (VARIANT STRAIN) 'ANTISERUM WHEN TESTED AGAINST VARIOUS S. PULLORUM ANTIGENS S. pullorum #BAI Antisera. 205 289 225 193 236 2 l|.0 21+1 320 1280 6 I4.O 2580 2580 6 L1O 6 J4.O S. pullorum #89817 80 320 80 160 160 1+0 80 S. pullorum #671 80 160 ij.0 160 160 20 I(.0 K antigen regular + K antigen polyvalent + Antigens S. pullorum #BAI + + + + + + + + Redigen regular Redigen polyvalent + + + + Figures indicate highest dilution at which agglutination occurredo 35 TABLE 1+ EXPERIMENT I AGGLUTINATION RESPONSE OF BRUCELLA ABORTUS #2308 ANTISERUM WHEN TESTED AGAINST VARIOUS BRUCELLA ANTIGENS Antigens Antisera Br. abortus Br. melitensis Br. suis #1255 #2308 #2500 827 2560 2560 2560 800 1280 1280 1280 231 2560 2560 2560 216 12 8 0 1280 1280 831 6[(.0 614.0 6 I4-O 1280 1280 1280 869 160 160 160 867 160 80 80 81+14- Figures indicate highest dilution at which agglutination occurred. 36 TABLE 5 EXPERIMENT I AGGLUTINATION RESPONSE OP BR. MELITENSIS #2500 ANTISERUM WHEN TESTED AGAINST VARIOUS BRUCELLA ANTIGENS Anti Antisera Br. abortus Br. melitensi Br, suis #1255 #2308 #2500 835 1280 1280 1280 868 160 320 160 866 80 80 80 81+8 614.0 1280 61+0 239 2560 2560 2560 277 2560 2560 2560 821 61+0 61+0 61+0 875 61+0 61+0 882 160 160 815 80 80 Figures indicate highest dilution at which agglutination occurred. TABLE 6 EXPERIMENT I AGGLUTINATION RESPONSE OP BR. SUIS #1255 ANTISERUM WHEN TESTED AGAINST VARIOUS BRUCELLA ANTIGENS Antigens Anti sera Br. abortus #2308 Br. melitensis #2500 Br. suis #1255 838 1280 1280 61+0 873 2560 2560 2560 876 1280 1280 1280 865 61+0 6!+0 61+0 230 2560 2560 2560 263 1280 1280 1280 Figures indicate highest dilution at which agglutination occurred,, 38 TABLE 7 EXPERIMENT I AGGLUTINATION RESPONSE OP S. PULLORUM #89817 (STANDARD STRAIN) ANTISERUM WHEN TESTED AGAINST VARIOUS BRUCELLA ANTIGENS Antigens S. pullorum #89817 antisera Br. abortus #2308 852 Br. melitensis #2^00 Br, suis #1255 neg, neg. 896 neg, neg, neg. 810 neg, neg. neg. 813 neg. neg. neg. 800 neg. neg. neg. 888 neg, neg, neg. 39 TABLE 8 EXPERIMENT I AGGLUTINATION RESPONSE OF S. PULLORUM #671 (INTERMEDIATE STRAIN) ANTISERUM WHEN TESTED'AGAINST VARIOUS BRUCELLA ANTIGENS Antigens S. oullorum #671 antisera Br. abortus Br. melitensis Br. suis #2308 #2300 #12£5 898 neg. I4O neg, 812 neg. 20 neg. 814.9 neg. 20 neg, 883 neg. I4.O nego 8814 - neg. neg. neg, 828 neg, neg, neg. 830 80 neg. 831)- 20 neg. Figures indicate highest dilution at which agglutination occuredo Neg, - no agglutination observed in dilutions lower than 1 :20 , TABLE 9 EXPERIMENT I AGGLUTINATION RESPONSE OF S. PULLORUM #BAI (VARIANT STRAIN) ANTISERUM WHEN TESTED'AGAINST VARIOUS BRUCELLA ANTIGENS Antigens So pullorum #BAI antisera Br. abortus #2308 Br. melitensis #25)00 Br. suis #1255 205 80 80 80 289 320 160 320 225 16 0 80 80 193 6I4.O 320 6I4.O 236 320 320 320 214.0 80 80 80 2 I4I 80 80 160 Figures indicate the highest dilution at which agglutin­ ation occurredo kl TABLE 10 EXPERIMENT I AGGLUTINATION RESPONSE OP BR. ABORTUS #2308 ANTISERUM WIEN TESTED AGAINST VARIOUS S. PULLORUM ANTIGENS Ahtisera S. pullorum antigens 827 S. pullorum #89317 800 231 216 831 869 867 neg. neg. neg, neg. neg, neg. neg. neg8 neg. neg. neg. neg. neg. neg. neg. neg. 6 I4.O 320 neg. neg. neg. neg. neg. neg0 + neg, S. pullorum #671 S. pullorum #BAI K antigen regular , neg. K ant i gen polyvalent neg. Redigen regular + Redigen polyvalent neg, neg, neg. + + + + neg. + + + + neg. + neg, neg. + 'neg. + neg. nego Figures indicate highest dilution at which agglutination occurred, + - slow or suspicious reactions TABLE 11 EXPERIMENT I AGGLUTINATION RESPONSE OF BR. MELITENSIS #2500 ANTI SERUM WHEN TESTED AGAINST VARIOUS S. PULLORUM ANTIGENS Antisera S. pullorum antigens 835 868 866 314.8 239 277 821 875 882 So pullorum # 8 9 8 1 7 nego neg. neg. neg. neg. neg. neg. neg. neg. S. pullorum #671 neg. neg. 20 20 neg. neg. neg. neg. neg. 320 S. pullorum #BAI . 320 K antigen regular neg. neg. neg. + K antigen polyvalent neg. neg. neg. ■¥ Redigan regular neg. neg. neg. neg. neg. neg. neg. neg. neg. Redigan polyvalent neg. neg. neg. neg. + + neg. -p neg. neg. + neg. + neg. neg. Figures indicate highest dilution at which agglutination occurred,, + slow or suspicious reactions + neg. + neg. k3 TABLE 12 EXPERIMENT I AGGLUTINATION RESPONSE OP BR. S U I S # 1 2 5 5 A N T I S E R U M W H E N TESTED AGAINST VARIOUS S. P U L L O R U M A N T I G E N S Antisera S. pullorum antigens 338 873 876 865 230 263 S. pullorum # 8 9 8 1 7 neg, neg, neg, neg, neg, neg, S. pullorum #671 neg, neg, neg, neg, neg, neg0 6ipO 320 So pullorum #BAI K antigen regular K antigen polyvalent Redigen regular . neg, neg, neg, neg, neg, neg, + + + + + + neg, neg, neg, neg, neg, neg. Redigen polyvalent Figures indicate highest dilution at w h i c h occurred, + - slow or suspicious reactions agglutination kb TABLE 13 EXPERIMENT II AGGLUTINATION RESPONSE OF S. PARATYPHI A, VAR. DURAZZO, S. READ ING. A N D PROTEUS (GWATKIN)" WHE N TESTED AGAINST THEIR HOMOLOGOUS A ND HETEROLOGOUS ANTIGENS Antisera Antigens (Preabsorption titers) S. paratyphi A, v a r # durazzo S. reading Proteus (Gwatkin) 9^0 802 155 839 208 895 893 S. paratyphi A, v a r 0 durazzo 2^60 *0 * 5012 320 320 0 0 0 320 320 320 S. reading *0 * 320 6 I4.O 2560 2560 Proteus (G) »0» 0 0 61p0 6I|0 6 I4.O 1280 1280 So pullorum #89817 1280 1280 bO k-o 0 0 0 1290 1280 160 160 k-0 ij.0 80 So pullorum #BAI 80 320 1280 1280 320 B r 0 abortus #2308 0 0 160 160 B r 0 melitensis #2£00 0 0 80 160 Br. suis #1255 0 S. pullorum #671 160 180 180 J4.0 k-0 k-o i).o k-0 k-0 ko k-0 Figures indicate highest dilution at which agglutination occurred. TABLE ll(. EXPERIMENT II AGGLUTINATION RESPONSE OP MONO-SPECIFIC SERA W H E N TESTED AGAINST VARIOUS ANTIGENS Mono-specific Sera Antigens S. pullorum # 8 9 8 1 7 IX 0 XII1 80 XII2 0 XII^ 320 S. paratyphi A, var. durazzo 0 80 0 1280 S. reading 0 320 6 I4.O 0 Proteus (Gwatkin) 0 0 6 I4.O 0 Br. abortus #2308 0 0 80 0 Br. xnelitensis #25>00 0 0 80 0 B r 0 suis #1255 0 0 80 0 Figures indicate highest dilution at which agglutination occurred. TABLE l5 EXPERIMENT III PREABSORPTION TITERS Antisera Antigen S. nullorum #BAI (193) Br, abortus # 2 3 0 8 (2 1 6 ) S. pullorum #BAI 2560 624.0 S. reading 1280 (Gwatkin) Br, melitensis #2500 (277) Br. siiis #1255 (230) 320 61+0 160 80 160 6 lp0 80 160 80 Br. abortus #2308 6 i{.0 1280 2560 2560 Br, melitensis #2500 320 1280 2560 2560 Br, suis #1255 6 /4O 1280 2560 2560 Proteus Figures indicate highest dilution at which agglutination occurred. kl TABLE 16 EXPERIMENT III AGGLUTINATION RESPONSE OF BR. ABORTUS #2308 ANTI SERUM ABSORBED WITH BR. AB0RTUS~F21T78 ANTIGEN Serum dilutions 86 160 320 Slide agg. 20 ko Br. abortus #2308 - - - - - - - Br. melitensis #2500 - - - - - - - Br. - - - - - - - S. pullorum #BAI - - - - - - - S. reading - - - - - - - Proteus - - - - - - - Antigen suis #1255 (Gwatkin) - no agglutination observed 1+8 TAELE I? EXPERIMENT III AGGLUTINATION RESPONSE OP BR. ABORTUS #2308 ANTI SERUM ABSORBED WITH BR. MELI TENS! S~ #2^00 ANTIGEN Antigen Slide agg, 20 Serum dilutions 80 160 320 1+0 '61+"0 Br, abortus #2308 - - _ - Br, melitensis #2500 r - _ - Br, suis #1255 - - - - S , pullorum #3AI - - - - S, reading - - - - Proteus - - - - (Gwatkin) - no agglutination observed J+9 TABLE 18 EXPERIMENT III AGGLUTINATION RESPONSE OF BR. A BORTUS #2308 ANTISERUM ABSORBED W I T H BR. SUIS #1255 ANTIGEN Slide aggo 20 U-0 Br. abortus #2308 - - - - - Br. melitensis #2500 - - - - - Br. suis #1255 - - - _ - S. pullorum #BAI - - - - - 3. reading - - - - - Proteus - - - - Antigen (Gwatkin) - no agglutination observed Serum dilutions 50 160 320 6 I4.O 5o TABLE 19 EXPERIMENT III AGGLUTINATION RESPONSE OP BR. ABORTUS #2308 ANTISERUM ABSORBED WITH S. PULLORUM~#BAI ~ANTIGEN Slide agg# Antigen Br. Serum dilutions 1 6 0 320 6'lj.O 1 2 8 0 20 ij.0 8-0 3+ 3+ 3+ 3+ 3+ 3+ + + 3+ 3+ 1++ k-+ k+ 3+ + - + M- k+ k+ k-+ 3+ + - - - - - - - - - - - - - - - - - - - - 2580 abortus #2308 Br. melitensis #2500 Br. suis #1255 S. pullorum #BAI 1 S. reading Proteus (Gwatkin) l - + incomplete agglutination - no agglutination observed 51 TABLE 20 AGGLUTINATION RESPONSE OP BR. MELITENSIS #2500 ANTISERUM ABSORBED WITH BR. MELITENSIS #25^0 ANTIGEN . Antigen Slide &gg# 2Q ^ Serum dilutions ^ ^ Br. melitensis 7 # 2500 Br. abortus #2308 Br. suis #1255 S. pullorum #BAI S. reading Proteus (Gwatkin) - no agglutination observed g TABLE 21 EXPERIMENT III AGGLUTINATION RESPONSE OP BR. MELITENSIS #2500 ANTISERUM ABSORBED WITH BR. ABORTUS' ~ # g W ANTIGEN Serum dilutions O 20 CD Slide agg. -po Antigen 160 320 6I|.0 Br. melitensis #2500 - - - - - Er. abortus #2308 - + - - - 3r. suis #1255 - - - - - 3. pullorum #BAI - - - - - S. reading - - - - - Proteus (Gwatkin) - - - - - + incomplete agglutination - no agglutination observed £3 TABLE 22 EXPERIMENT III AGGLUTINATION RESPONSE OP BR. MELITENSIS #2^00 ANTISERUM ABSORBED WITH BR. SUIS #12^5" ANTIGEN Antigen Slide aggo 20 J+o Serum dilutions 80 320 160 6 )4.0 Br« melitensis #25>00 - - - - - - - Br. abortus #2308 - - - - - - - Br. suis #125>5 - - - - - - S. pullorum #BAI - - - - - - - S. reading - - - - - - - Proteus (Gwatkin) - - - - - - - - no agglutination observed i TABLE 23 EXPERIMENT III AGGLUTINATION RESPO N S E O P B R . MELITENSIS #2500 ANTISERUM A B S O R B E D W I T H S. PULLORUM #BAI ANTIGEN Slide aggc 20 IpO Br. melitensis #2 5oo ■f" k+ 1++ Br. abortus #2308 + Br. suis #1235 + S. pullorum #BAI S. reading Antigen 80 Serum dilutions 160 320 6i|_0 1280 k+ 2560 k+ 3+ + k+ k+ k+ 3+ 3+ 3+ + lp+ k+ k+ ¥ ii+ 3+ 3+ + - - - - - - - - - - - - - - Proteus (Gwatkin) - - - - - - - + incomplete agglutination - no agglutination observed TABLE 2k EXPERIMENT III AGGLUTINATION RESPONSE OP BR. SUIS #1255 ANTISERUM ABSORBED WITH BR. SUIS #T255 ANTIGEN . Antigen Slide agg^ ^ p 5o - - - - Br. melitensis #2500 - - S. pullorum #BAI - - S* reading - Proteus (Gwatkin) - Br. suis #1255 Serum dilutions IE0 320 SIpO Br, abortus #1255 - - no agglutination observed 56 TABLE 25 EXPERIMENT III AGGLUTINATION RESPONSE OP BR. SUIS #1255 ANTISERIJM ABSORBED WITH BR. ABORTUS #2j?08 ANTIGEN Slide agg. 5° ip - -■ - - - - - Br. abortus #2308 - + - - - - - Br. melitensis #2500 - - - - - - - S. pullorum #BAI - - - - - - - S. reading - - - - - - - Proteus (Gwatkin) - - - - - - - Antigen Br. suis #1255 + incomplete agglutination - no agglutination observed Serum dilutions 80 160 320 6ip 57 TABLE 26 EXPERIMENT III AGGLUTINATION RESPONSE OF BR. SUIS #1255 ANTISERUM ABSORBED WITH BR. MELITEN S T s f 2500 ANTIGEN Slide agg. Antigen Serum dilutions 20 1^0 80 - - mm «M Br. suis #1255 - - 3r« abortus #2308 - _ Br« melitensis #2500 M So pullorum #BAI S. reading Proteus (Gwatkin) - no agglutination observed 160 _ 320 61+0 TABLE 27 EXPERIMENT III AGGLUTINATION RESPONSE OF BR. SUIS #12^5 ANTISERUM ABSORBED WITH S. PULLORUM #BAI ANTIGEN Serum dilutions 160 320 64.0 Slide agg. 20 ¥ Bo Br. suis #1255 + ¥ ¥ ¥ ¥ 3+ 3+ 2+ Br. abortus #2308 + ¥ k+ ii+ 3+ 3+ 3+ 3+ Br. melitensis #25>00 + ipf- ¥ [).+ ¥ ¥ 3+ 3+ S. pullorum #BA!E - - ~ - - - - 3. reading - - - - ~ - - Proteus (Gwatkin) - - - - - - - Antigen + incomplete agglutination - no agglutination observed 1280 25&( +_ 59 TABLE 28 EXPERIMENT III AGGLUTINATION RESPONSE OP 3. PULLOxRUM #BAI ANTI SERUM ABSORBED WITH S. PULLORUM #BAI ANTIGEN Slide agg. 20 Serum dilutions 80 i”6o 320 i+o S. pullorum #BAI - - - - - - S. reading - - - - _ - Proteus (Gwatkin) - - - - - - Br. abortus #2308 - - - - - - Br. melitensis #2500 - - - - Er. suis #1255 - - - - Antigen - no agglutination observed 61+0 - - 60 TABLE 29 EXPERIMENT III AGGLUTINATION RESPONSE OF S. PULLORUM #BAI ANTISERUM ABSORBED WITH BR. ABORTUS #$308 ANTIGEN Antigen S l i d e ___________ Serum dilutions agg< 20 5-0 80 160 320 6 i|0 1280 256( k+ 5-+ 3+ 3+ 3+ 2+. 5-+ 3+ 3+ 2+ + + k-+ 5-+ 5-+ 2+ 2+ + ± — - - - - - B r 0 melitensis # 2^00 - - - - - - Br. - - - - - - S. pullorum #BAI + S. reading + Proteus (Gwatkin) + 5-+ + Br. abortus # 2308 suis #12^5 + incomplete agglutination - no agglutination observed I TABLE 30 EXPERIMENT III AGGLUTINATION RESPONSE OP S. PULLORUM #BAI ANTISERUM ABSORBED WITH BR. MELITENSIS'1^2500 ANTIGEN Antagen Slide agg# ___________Serum dilutions 20 l+o BO 160 320 0 1280 2 ^6 ( S. pullorum #BAI + k+ I4.+ I4.+ 14.+ k-+ 2+ 2+ + S. reading + k+ k+ 3+ 3+ 3+ + + - Proteus (Gwatkin) + k+ k+ ¥ 3+ 2+ + + - - - - - - - Br. melitensis # 2£00 - - - - - - Br. suis #1255 - - - - - - Br. abortus #2308 + incomplete agglutination - no agglutination observed 62 TABLE 31 EXPERIMENT III AGGLUTINATION RESPONSE OF S. PULLORUM #BAI ANTISERUM ABSORBED WITH BR. SUIS~^Tg5BTANTIGEN Slide agg# 20 40 Serum diluti ons 80 160 320 6 J4.O S . pullorum #BAI + [).+ k+ k+ [(.+ k+ 3+ 3+ - S. reading + !(.+ 3+ 3+ 3+ 2+ + - lj-+ J++ 3+ 3+ 3+ + + - - - - - - - Br. melitensis # 2^00 - - - - - - Br. suis #1255 - - - - - - Antigen Proteus (Gwatkin) + Br. abortus #2308 + incomplete ap'glutination - no agglutination observed 1280 2^60 DISCUSSION It has been pointed, out by Foshey (22), Cioglia (8 , 9), and others that a number of salmonellae can cross agglu­ tinate with Brucella, However, no one as far as I have been able to determine, has made an attempt to determine what the common antigens in these groups of organisms are, Cioglia (8 , 9), however, determined whether the H or 0 antigens were responsible for the cross agglutination between Br. melitensis and the Salmonella which he studied,, In this experiment it was found that a part of the XIIo antigen can cross agglutinate with Brucella. A number of organisms in the genus Salmonella possess antigen Xllg* Some of these are: S. enteridites, S. paratyphi B, S. typhi-murium, S. derby, S^. abortus-ovis, S. abony, S. brandenburg, S. typhosa, and S. essen 173• These organisms are known to cause infection in man and animals. One can speculate on the possibility that some of these organisms when found in cattle can produce agglutinins in sufficient quantity to cause a cross agglutination when being tested for Bangs disease. This speculation can al30 be applied when carrying out agglutination tests for undulant fever. It is known that S. typhosa contains the XIl£ antigen(37)« It would be possible for one who has had typhoid fever, is immunized against typhoid fever to develop reacting agglutinins for Brucella. It should again be re-emphasized that it or if one cross is not known, at present, whether all cultures which contain the XII^ antigen also contain that portion of the antigen which can cross agglutinate with Brucella. I n discussing the problem of cross agglutination reactions between S. pullorum and Brucella the question can be raised as to the prevalence and importance of Brucella infection in fowl. The frequency and importance of brucellosis in b.irds is still under discussion. Van Roekel, Bullis, Flint, and Clarke (5>6) examined 25*202 chickens; the area covered represented approximately every county in Massachusetts. No reactors to Brucella antigen were found in any of the birds when a dilution of 1 :2 5 and 1 : 5 0 were used. McNutt and Purwin (ip.) examined 69 flocks containing over 1 0 ,0 0 0 birds, and found less than 2 percent reactors. No one flock contained more than 12 percent reactors. Ernmel (18) found 16.5 percent reactors in one flock of 90 chickens, Huddleson and Emmel (31) examined four flocks which they believed had brucellosis. birds were tested. In flock number one, lli). Seventeen of these birds' blood sera agglutinated Br. abortus in dilutions of 1:25 to 1:100. In flock number 3* birds were examined. Thirteen of these birds' blood sera were found to agglutinate Br. abortus in dilutions varying from 1:25 to 1:100. "Eleven of these birds were purchased* killed, autopsied, cultured for the genus Brucella. . . . the organs remained sterile. and organs Cultures from all Fifteen eggs were taken from these birds and cultured for Brucella and none were found to be infected." In the fourth flock "three birds were received for diagnosis at the laboratory. sisted of 800 birds. Thirty were sick. The flock con­ Ten had died, all showing emaciation, paleness of the comb, wattles, about the head, diarrhea, and extreme weakness. shown paralysis just before death. dropped l5 percent." A few had Egg production had The three birds examined had titers to Brucella ranging from 1:$0 to 1:100. "A species of Brucella was isolated from the lungs, kidneys, of bird Bo. 1. and and spleen Cultures made from the organs of the remaining birds remained sterile." Anguelov, cited by Stafseth (53) reported that bru­ cellosis is very prevalent in modern poultry plants in Bulgaria. Felsenfeld (20) reported that brucellosis in fowl is a serious problem in Eastern Europe. Brand.ly (3) also reported that Russian workers have stated that brucellosis is a major poultry problem in Russia« In most of the diagnoses of brucellosis in fowl the majority of investigators (1 3 , 3 1 , 3 9 , 1-1-1 , lj-2 ) have been unable to isolate the organism from the affected birds. Most of the diagnoses of brucellosis have been based on the finding of agglutinins which agglutinate Brucella anti­ gens and evidence of the disease in other animals on the farm. "The vast majority of authors have failed to observe symptoms of brucellosis in birds, and it is not certain that the symptoms described by others have actually been due to the disease." (5 3 ). Felsenfeld et_ al. (21) stated "Due to frequent lack of clinical symptoms in birds infected with Brucella and the possibility that the death of some birds may be m i s ­ interpreted as pullorum disease, because of the serologic cross reactions with pullorum antigens, the true cause of some diseases in poultry may be obscured," The most important aspect of brucellosis in fowl is its relationship to public health. Brucellosis in fowl constitutes a definite public health hazard. This has been pointed out by Felsenfeld £t al. (21), Brandly (2), Ingalls (32), Brandly (3), and Felsenfeld (20). Felsenfeld et_ al. (21) remarked "one should keep in mind the possibility of human infection by eating Brucellainfected poultry meat. The meager pathologic signs make it difficult or even impossible, to detect all infected chicken during food inspection." In another paper Felsenfeld (20) further pointed out "Since blood cultures frequently become positive during such infections, chickens may serve not only as vectors of brucellosis on the farm but if slaughtered during the bacteremia, may provide meat that is infected with B ruc ell a," This same thought was expressed by Brandly (3 )> "the carcasses of diseased birds often contain myriads of patho­ genic organisms which are introduced into the kitchen with the carcasses; and knives, sinks, pans, hands, are contaminated by these disease germs. chicken salads, cold chicken sandwiches, towels, etc, In preparing etc, these organ­ isms can be reintroduced into the edible product and cases of food poisoning or infection are the result," "Since birds can become infected with Brucella and may thus serve as agents of transmission of this disease not only to other birds but to mammals as well, one should take steps to prevent fowl from being in contact with infected mammals. Good poultry hygiene demands that poultry should be confined within premises set aside for this type of livestock and not be allowed in barns, hog yards, etc. The practice of throwing dead chickens on a manure pile or elsowhere, where hogs or other birds may eat them, is to be condemned," (53)• At present birds are not routinely tested for brucello­ sis. If, however, Brucella testing of birds becomes necessary, S. pullorum agglutinins may interfere with an accurate serological diagnosis. To overcome this, the use of the filter paper fixation test may prove to be a valuable diagnostic test. To test the hypothesis that the filter paper surface fixation test is specific for Brucella agglutinins, the antisera of a number of organisms that are known to cross agglutinate with Brucella were tested using this method. These included Proteus O X K , Proteus 0 X 2 , Proteus 0X19, Proteus (Gwatkin), S. reading, and S. pullorum # B A I . None of the above antisera pro'duced a positive reaction, while Brucella antisera continued to give a positive test. Although the filter paper surface fixation test may prove to be of value in the diagnosis of brucellosis, the isolation and identification of one of the members of the genus Brucella constitutes the only positive way of diagnosing this disease. SUMMARY AND CONCLUSIONS No cross agglutination reactions were observed between standard strain S_. pullorum #89917 and B rucella. In a number of cases cross reactions were noted between inter­ mediate strain S. pullorum #671 and Brucella. Cross agglutination reactions were observed between variant strain S. pullorum #BAI and Brucella. The antigenic structure of S. pullorum was determined b y Edwards and Bruner (llj.) to be IX, X I I ^ (Xllg), XIl y results it appears that the XII 2 antigen, Prom the above (found in the intermediate and variant strains of S. pullorum) or a part thereof, was responsible for the observed cross reactions. Reciprocal agglutination tests were also carried out with Brucella and other organisms, Salmonella reading and Proteus (Gwatkin) which contain the XII^ antigen. The cross reactions observed between these organisms and Brucella were less pronounced than those noted between S. pullorum #BAI and Brucella. There appears to be a quantitative difference In the amount of the cross reacting antigen (for Brucella) in S. pullorum # B A I , S. reading, and Proteus (Gwatkin). Whether this is a strain characteristic of the organisms involved is not known. Mono-specific sera containing agglutinins IX, XII^, XIIp, and XII^ were prepared. Only the Xllg mono-specific 70 sera produced cross agglutinations with the Brucella antigens. It was noted that only a small fraction of the XIl£ antigen was involved in these cross reactions. Agglutinin absorption studies carried out with So pullorum #BAI and Brucella indicated that the antigen common to S. pullorum #BAI, S. reading, and Proteus (Gwatkin) existed in approximately equal amounts in all three species of Brucella tested. One hundred and seventy-six S, pullorum suspicious and reactor turkeys were tested using Br. abortus #2308 and Br. melitensis #25>00 antigens in dilutions of l:2jp, 1:E>0, and l: 1 0 0 o Of this total, 29 (seventeen percent) showed a cross reaction with the Brucella antigens. A number of salmonellae, pathogenic for man and animals, contain antigen X I ^ . These organisms, if they contain the antigenic factor common for Brucella, may produce agglutinins in sufficient quantity to cause false positive agglutination tests for brucellosis. LITERATURE CITED 1. Beach, B. A., J. G. Halpin, and C. F. Lampman. Results of white diarrhea investigations. J.A.V.M.A., 70:605>6 1 1 , 1929. 2. Brandly, C. A. Poultry diseases as a public health problem. Public Health Reports, 66:668-672, 1951* 3. Brandly, P. J. Poultry inspection as part of the public health problems. J.A.V.M.A., 112:10-17, 19l|-8. ij.. Bunyea, H., V/. A. Hall, M. Dorset. A simplified agglutination test for pullorum disease. J.A.V.M.A., 75:ij.08-l|10, 1929. 5. Burton, W. H. and E. ation reactions. IV. from fowl inoculated Comp. Med. Vet. Sci., H. Garrard. Non-pullorum agglutin­ Reactions with pullorum antigen with coliform types. Can. J. 12:20-25?, 19Il8. 6 . Carpenter, J. A., W. H. Burton, and E. H. Garrard, Non-pullorum reactions. III. Reactions with pullorum antigen from fowl inoculated with an enterococcus. Can. J. Comp. Med. Vet. Sci., 11:163-168, 19^4-7. 7o Castaneda, R. M. Surface fixation. A new method of detecting certain immunological reactions. Proc. S o c 0 Exp. Biol. Med., 73:Lj.6 —14.9 » 195>0. 8. Cioglia, L. Antigen! communi a salmonelle. Immun., 14-2:81-93, 195>0. 9. ___________. Sulla constituzione antigene delle brucelle e gang. G. Batt., Ij.2:3 I4-6 - 3 6 2 , 1950. 10, 11. G. Batt. Coburn, D, R. and H. J. Stafseth. A field test for pullorum disease. Preliminary report. J.A.V.M.A., 79:2lp.-2ij.3, 1931. Corpron, R. A., J. A. Bivins, and H. J. Stafseth, Pullorum disease studies in turkeys. I. Antibody pro­ duction following intravenous and oral infection, relative value of tube and whole-blood plate agglutin­ ation tests, pen contagion and duration of carrier state. Poultry Sci., 36:3l|-0-35l, 191+7• 72 12. Doyle, T. M. Baclllary white diarrhea of chicks. J. Comp. Path. Ther., 38:266-282, 1925* 13. Dubois, M. ill-. Edwards, P. R. and D. W . Bruner, Form variation in Salmonella pullorum and its relation to X strains. Cornell. Vet 32; 318- 3211, 195-8° l?. , , E. R. Doll, and G. J. Hermann. Further notes on variation in Salmonella pullorum. Cornell Vet., 38:257-262, 195-8. 16 . , F. E. Hall. Bacillary white diarrhea and related diseases of chickens. Kent. Agr. Exp., Sta. Bull., 296:237-280, 1929. 17. , W. B. Cherry, and D, W. Bruner. Further studies on coliform bacteria serologically related to the genus Salmonella. J. Infect. Dis., 73:229-238, 195-3* 18 . Enamel, M. W. J.A.V.M.A., 1 ?. fWl. 20 Malta fever in fowl. Rev. Vet., 67:^1-90, 1910. An outbreak of Brucella disease in fowl. 76:565--565> 1930. , and I. F. Huddleson. A b o r t i o n disease in J.A.’V.M.A., 75:573-580, 1929. 0. Diseases . Felsenfeld, man. Iowa State College of poialtry transmissible to Vet., 13:89-92, 1951. 21 . ___________ , V. E, Young, E. Loeffler, S. J. Ishichara, ancr"F7 ¥. Schroeder. A. study of the nature of brucellosis in chickens. Am. J. Vet. Res., 53:5-855-, 1951. 22. Foshey, L. Laboratory diagnosis of undulant fever. Am.' J. Clin, path., 10:176-187, 195-0. 23. Gage, G. E ., B . H. Page, and Ii. V/. Hyland. On the diagnosis of infection with Bacterium pullorum in domestic fowl. Mass. Agr. Exp. Sta. Bull. 15-8, 1915-. 2L\.» Garrard, E. H., L. A. McDermott, W. H. Burton, and J. A. Carpenter, Non-specific pullorum agglutination reactions. I. Preliminary observations on fowl exhib­ iting non-specific reactions over an extended period. Can. J. Comp. Med. Vet. Sci,, 10:35-2-35-7, 195-6. 25. ______________ , ¥. H. Burton, J. A. Carpenter, and L. A. McDermott. Non-specific pullorum agglutination reactions, II. Post-mortem studies on fowl exhibiting non-specific reactions over an extended period. Can. J. Comp. Med. Vet. Sci., 11:102-107, 195-7* 73 26. Gwatkin, R. Studies in pullorum disease. XII .■Antigenic differences in strains of Salmonella pullorum. Am. J. Vet. Res., 8:2014.-208, 191+7• 27. Hinshaw, W. R., E. E. Jones, J. R. Harr, and W . E. Neimeyer. Comparison of the tube and whole-blood tests for uullorum disease in turkeys. Cornell Vet., 30:30-38, i9l)-0. 28. Hobs von Horst and E. Arjona. Uber einem von Bacterium coli mit antigenbegechinger zur Sallmonellagrupoe. Zent. f. Bact., 133:20l|.-209, 1935. 29. Horton, G. D. Apparent recovery of a hen infected with bacillary white diarrhea. As determined by the macroscooic agglutination test. J. Bact., 1:625626, 191&o 30. Huddleson, I. F. Brucellosis in Man and Ani m a l s . Commonwealth Fund, Hew York, 191+3• 31. _________________ and M. W. Emmel. The pathogenicity of the genus Brucella for the fowl. Tech. Bull. 103, 1929o 32. Ingalls, W. L. The public health aspects of poultry diseases. J.A.V.M.A., 117:120, 1950. 33• Johnson, F. P. and G. VI. Anderson. Pullorum disease in turkeys. J. Infect. Dis., 58:337-31+1, 1936. 31+. _____________ and M. Pollard. Further observations on an organism in turkeys whose blood serum agglutin­ ates Salmonella pullorum. J. Infect. Dis., 66:19l+197, 191+0. 35. Jones, F. S. The value of the microscopic agglutin­ ation test in detecting fowls that are harboring Bacterium pullorum. J. Med. Res., 27:1+81-1+95, 1 9 1 3 0 3 6 . Kauffmann, F. The serological Salmonella diagnosis. Proc, 3rd Inter. Cong. Microbiol., 6 2 8 -6 3 6 , 191+0. 37. • A typhoid variant and a new serological variatioh’T n the Salmonella group. J. Bact., 1+1: 127-11+0, 191+1. 38. Kaupp, B. F. and Dearstyne, Bacillary white diarrhea. North Carolina Agr. Exp. Sta. Tech. Bull. 29, 1927. 39. Koegel, A. Bietrage sonr abortusforschung. Tierarztliclie Woehenscarift, 7l+: 6 1 7 , 1923. Mmehener Ik I4.O. McCullough, N. B., C. W. Eisele, and G. A, Beal. Antigenic relationship of Brucella and Vibrio cornua, J. Infect. Dis., 83:55-59, 19^8. )|.l . McNutt, S. H. and P. Purwin. The effect of the Brucella group of microorganisms on chickens. J.A.V.M.A., 77:350-353, 1930. lj.2. Mohler, J. R« Report of the chief of the Bureau of Animal Industry, 1912. i|_3. Peluffo, C. A., P. R. Edwards, and D. W. Bruner. A group of coliform bacilli serologically related to the genus Salmonella. J. Infect. Dis., 70:185-192, 1914-2. )|)| T Report of the Conference of Official Research Workers in Animal Diseases of North America on Standard Methods of Pullorum Disease in Barnyard Fowl. Proceedings Thirty-sixth Annual Meeting of the United States L i v e ­ stock Sanitary Association. J.A.V.M.A., 82:14-87-14-91, 1933* Jj.5• Rettger, L. F.' Septicemia among young chickens. N. Y. Med. J., 71:803, 1900. I4.60 ___________ . ■ Further studies on fatal septicemia or "white diarrhea". J. Med. Res., 21:115-123, 1909. in young chickens, I4.7 . and F. H. Stoneburn. Bacillary white diarrhea of young chicks. Storrs.A.gr. Exo. Sta. Bull., 60:29-57, 1909. 1-1-8o and W. N. Platridge. o'f domestic fowl. Monograph Storrs Pullorum disease Agr. Exu.. Sta. Bull., 173:103-192, 1932. il-9• ___ ___ _______ , W. F, Kirkpatrick, and R. E. Jones. Bacillary white diarrhea of young chicks. Storrs Agr. Exp. Sta. Bull., 77:259-309, 1911+. 50 . ________ Bacillary white diarrhea of young chicks. Exp. Sta. Bull., 8 5 :114-9 -1 6 7 , 1915. _. Storrs 51. Runnells, R. A., C. J. Coon, H. Farley, and F. Thorp. An application of the rapid method agglutination test to the diagnosis of bacillary white diarrhea infection. J.A.V.M.A., 70:660-662, 1927. 52. Schaffer, J. M., A. D, MacDonold, W. J. Hall, and H. Bunyea. A stained antigen for the rapid wholeblood test for pullorum disease. J.A.V.M.A., 79: 236 - 2 li0 , 1931. 7£ 53. Stafseth, H. J. Brucellosis, H. E. Biester and L. H. Schwarte, Editors. Diseases of P o u l t r y , Iowa State College Press-, Ames, Iowa. Chapter 11+7 191+8. £1+. United States Department of Agriculture. The National Poultry Improvement Plan. Misc. P u b . 300, 191+1. 55 . Van Roekel, H. Pullorum Disease, H . E. Biester and L. H. Schwarte, Editors. Diseases o f Poultry. Iowa State College Press, Ames, Iowa. C n a p t e r 8, 191+8. 56. _ _ _______ , K. L. Bullis, 0. S. Flint, and M. E. ClarkeT" Susceptibility o f .chickens to brucellosis. J.A.V.M.A., 80:6141-61+2, 1932. 57. Ward, A. R. and Gallagher. An i n t r a d e r m a l test for Bacterium nullorum Infection in f o w l . U.S.D.A. Bull. 517, 1 9 1 7 / 58. Williams, J. E. U s e of the spot t e s t in the diagnosis of pullorum disease. Poultry S e i a , 30:125-131, 195>1. 59. Wright, M. L. Selection of suitable antigenic strains of S. pullorum by single colony t y p i n g . Can. J, Comp. Med. Vet. Sci., 11:68-71+, 191+7. 60. _________ . The suitability of Sa l m o n e l l a pullorum "strains used in the agglutination t e s t for pullo,rum disease control. J.A.V.M.A., 112: 21+1-21+1+, 191+8. 61. and P. R 0 Ed w a r d s . Th.e serologic differ­ entiation of' Salmonella pullorum f o rms. Am. J. Vet. Res., 33:386-388, 191+8. 62. , Younie, A. R. Fowl infection like pullorum disease. Can. J. Comp. Med. Vet. Sci., 5:l61_+-l67, 191+1. 63. Zweck, W. and A. Zeller. Uber d e n infectosen abortus des Rundus. Arb. a. d. Kauserl. Gesundeitstant, 1+3: 1, 1913.