SEROLOGICAL VARIATION 257 



against 39 type A strains from 1933 to 1949, and swine influenza virus 

 against prototype strains of 1933, 1934, 1941, and 1947. The swine virus and 

 WS strain were classified as distinct variants. The authors grouped strains 

 from 1934 to 1945 together because of extensive cross reactions, while 

 differing from later and earlier strains. But they comment on the great 

 individuality and the difficulty of classification within the group except by 

 cross testing with sera of significant heterologous potency. The 1947 and 1949 

 strains differed from the PR8 group but were closely similar to each other. 

 The same relation between 1949 and 1947 strains was noted by Chu et ah, 

 (1950). Hilleman (1951, 1952) introduced an orderly effort to chart antigenic 

 make-up by quantitating HI activity of immune chicken serum to a strain 

 against each of four prototypes — 1933a, 1934b, 1947c, 1950d. The patterns 

 were expressed as shown in Table I. 



TABLE I 



Thus, the basis for cross reaction can be visualized. Again the differences 

 between groups were observed, and the great homogeneity of strains from 

 different parts of the world during a twelve-month period in 1950-51 was 

 noted. But, in addition, attention was called to a series of French, Canadian, 

 and Arctic strains of 1949 resembling PR8; there were other suggestions that 

 strains of earlier configuration might be in circulation in later years. Magill 

 and Jotz (1952) studied with rabbit sera a large number of strains isolated 

 over the same period. The results were similar to those of Hilleman, and on 

 the basis of variations in reaction with selected strains, the strains of 1946-47 

 to 1949-50 were grouped together, while essentially all the 1950-51 strains 

 were congregated in another. Neither series contained significant representa- 

 tion of 1940-41 and 1943 strains. Five subgroups of type A virus were sug- 

 gested. Isaacs and Andrewes (1951), using ferret sera for study of 1950-51 

 strains, reported two subgroups which differed in avidity for immune serum 

 and were thought to be epidemiologically independent. It is doubtful that the 

 reported differences are of antigenic nature, since they are observed only 

 with ferret sera. Serologically, they were not observed by Hilleman or Magill, 

 who considered their 1950-51 strains serologically homogeneous. A number 

 of investigators have described a further antigenic deviation in 1953 strains, 

 suggesting that they were a new group unrelated to 1947 and 1950 strains 

 vol. in — 17 



