268 ANNUAL REPORT SMITHSONIAN INSTITTJTION, 1942 



normal rabbit serum or antisera to bushy stunt, ring spot, or latent 

 mosaic viruses. However, electron micrographs of a mixture of to- 

 bacco mosaic virus and tobacco mosaic virus antiserum from rabbits 

 show particles about 60 m/t wide, about 300 niju, long and having fuzzy 

 profiles (pi. 4, fig. 2). The increase in particle width and the fuzzy 

 appearance are regarded as indicating that the ends of asymmetrically 

 shaped molecules from the serum react specifically with the virus 

 molecules. No reaction between anti-tobacco mosaic virus serum and 

 bushy stunt virus was demonstrable either by means of the electron 

 microscope or the ultracentrif uge (pi. 5, fig. 1) . Bushy stunt virus is, 

 however, specifically precipitated by its own antiserum (pi. 5, fig. 2). 

 In general, a serological relationship may be demonstrated between 

 different strains of the same virus, but not between different viruses. 

 However, Bawden and Pirie have found that a serological relationship 

 exists between tobacco mosaic and cucumber mosaic 3 and 4 viruses. 

 Tliis fact and the fact that other properties of these viruses are very 

 similar may indicate a common origin for these viruses. Bawden and 

 Pirie also noted that the precipitates of rod-shaped viruses with their 

 antisera resembled those obtained with bacterial flagellar antigens, 

 whereas those of the symmetrically shaped bushy stunt virus resembled 

 those with somatic antigens. Tobacco mosaic virus has been found not 

 anaphylactogenic by the Schultz-Dale technique and only weakly ana- 

 phylactogenic when tested in vivo. 



Viruses are inactivated when subjected to excessive amounts of 

 acid, alkali, oxidizing agents, formaldehyde, urea, ultraviolet light 

 or heat. In general, the rate and the amount of the inactivation 

 vary with the virus and with the severity of the treatment. Tobacco 

 mosiac virus is stable between about pH 2 and pH 8. At more 

 acid or more alkaline reactions the nucleoprotein is denatured and 

 broken up into material of low molecular weight, and the virus 

 activity appears to be irreversibly lost. There is some evidence that 

 the virus first breaks into fairly large pieces and these then continue 

 to break up into progressively smaller pieces, but more data will be 

 required before an exact picture of the process may be obtained. 

 The disintegration of virus in urea provides another interesting 

 process for study. In 6 M urea and 0.1 M phosphate buffer at pH 7, 

 tobacco mosaic virus is rapidly disrupted, with appearance of free 

 sulfhydryl groups, into low molecular weight protein components 

 which contain no nucleic acid, exhibit no double refraction of flow, 

 are insoluble in dilute buffers, and possess no virus activit3\ The rate 

 of the disintegration varies widely with the concentration of urea, the 

 concentration of electrolyte, the type of electrolyte, the hydrogen-ion 

 concentration and the temperature. The bonds which hold the com- 

 ponent parts of the virus together appear to be released and satis- 

 fied by those of the urea, for the virus structure literally flies apart. 



