FITZGERALD AND FRASER 817 



The microscopic-slide agglutination test is widely used for determining the presence or 

 absence of a specific agglutinin for micro-organisms, especially those of the enteric-fever 

 group. In collecting blood for the test, a few drops are allowed to dry on the ground surface 

 of a glass slide or other impervious material; or a large amount is collected, suflicient to yield 

 a few drops of serum. Saline or bouillon in appropriate amounts is added to the dried blood 

 with a Pasteur pipette. With small amounts of serum, dilutions may be made with a hemo- 

 cytometer pipette, either "red" or "white." In order to approximate the desired initial dilu- 

 tion with any degree of accuracy, it is necessary to adopt some color standard with which to 

 match the blood of the specimen. Such a color standard, using defibrinated sheep's blood, has 

 been described by Gilbert and Parsons.' The dilutions i/io, 1/20, and 1/40 are commonly 

 employed. Hanging-drop preparations are made after adding equal amounts of an eighteen- 

 hour (incubated at 25° C.) broth culture of B. typhosus or other desired species. The cover- 

 slip margin is streaked with vaseline to insure that the small chamber is sealed to prevent 

 evaporation of the suspended drop. The final dilution after the addition of the bacterial 

 suspension becomes 1/20, 1/40, and 1/80. A control of micro-organisms and diluent is of 

 course essential. The test is incubated at 37° C, examined at once and at fifteen-minute 

 intervals for one hour. Complete or almost complete agglutination with loss of motility at 

 1/40 dilution of serum is usually considered diagnostically positive for typhoid fever. 

 Normal serum rarely if ever manifests a complete agglutination reaction at 1/40 dilution 

 when the microscopic method is employed. 



The microscopic-slide agglutination test is widely used, as Hull^ found in answer to a 

 questionnaire relative to the technique employed, sent to state and municipal laboratories. 

 To quote from his paper: "It is safe to say that to-day probably no two laboratories perform 



the Widal test in exactly the same manner " The answers received from fifty-three 



laboratories leave one in no doubt as to the desirability of adopting a definite standard meth- 

 od based upon the control of the known factors involved. In regard to the choice of suspen- 

 sion of bacilli, Gilbert and Moore^ and the writers of the Medical Research Council, "Special 

 Report Series," No. 51,4 have demonstrated the impractibility of using killed cultures 

 (formalinized o.i per cent) for microscopic agglutination. Further, the culture used must be 

 one of satisfactorily proved agglutinability and employed at a definite density. A barium 

 sulphate standard of opacity as described by McFarland^ may conveniently be employed. 

 For microscopic agglutination tests a density of 1,000 million bacilli per cubic centimeter is 

 desirable. Controls with a known standardized serum should be set up. For this purpose the 

 lowest dilution of serum is used which, by microscopic-slide agglutination, definitely aggluti- 

 nates the homologous organism and agglutinates slightly or not at all the heterologous 

 organism.^ A uniform system of recording results is essential. Such a system as described in 

 the foregoing publication deserves general adoption. 



The microscopic-slide agglutination method possesses certain distinct advantages: It can 

 be carried out with very minute quantities of serum, even with less than a drop; it can be 

 carried out quickly; it requires a minimum of apparatus. Over against these advantages one 

 must balance the limitations. It is quantitatively inaccurate; its end-point is very ill defined; 

 it is qualitative rather than quantitative. 



■ Gilbert, R., and Parsons, E. A.: /. Lab. &° Clin. Med., 10, 206. 1924-25. 



= Hull, T. G.: Am. J. Pub. Health, 16, 901. 1926. 



3 Gilbert, R., and Moore, A. C: J . Lab. cr Clin. Med., 7, 547. 1922. 



"* "Medical Research Council, Special Report Series," No. 51. 1920. 



s McFarland, J.: J.A.M.A., 49, 1176. 1907. 



* Wadsworth, A. B.: Standard Methods, p. 174. Williams & Wilkins, 1927. 



