502 ANNUAL KEPORT SMITHSONIAN INSTITUTION, 19 3 8 



you that, according to all the tests that we have been able to devise, 

 the protein is homogeneous in all respects, hence pure, hence is the 

 virus. Among the physical tests that have been used is homogeneity 

 with respect to rate of sedimentation, as measured in the Svedberg 

 ultracentrifuge, and homogeneity with respect to electrochemical 

 behavior, as measured in the electrophoresis cell. Chemical tests for 

 homogeneity include fractionation of the protein by fractional crystal- 

 lization and by centrifugation from solution under a variety of condi- 

 tions, solubility studies, correlation of absorption spectrum with 

 destruction spectrum and studies on partial inactivation. The bio- 

 logical tests include the determination of the virus activity of the 

 various fractions of protein obtained by physical and chemical means. 

 The serological experiments include precipitin, complement-fixation, 

 and both in vitro and in vivo anaphylactic tests. In connection with 

 these tests, you may recall that at the meetings a year ago the protein 

 as then prepared showed a cross reaction with normal plant protein 

 when tested by the very sensitive anaphylactic technique. It was 

 suggested that this cross reaction might be due to the presence of a 

 fraction of a percent of normal protein as an impurity. This has been 

 found to be the case, and it is now possible to remove this impurity and 

 to prepare protein that gives no cross reaction with other material 

 even by the most sensitive serological tests. Thus, by all the tests 

 that we have been able to devise, the tobacco mosaic virus protein 

 appears to be homogeneous. There is always the possibility, of 

 course, that the virus activity is not due to the high molecular weight 

 protein but to an impurity that cannot be detected by means now at 

 our disposal. It should be recognized, however, that, by virtue of the 

 experimental evidence for homogeneity already available, an impurity 

 could hardly be other than a closely related high molecular weight 

 protein, hence it would still follow that the virus is a protein. The 

 possibility that the activity may be due to an impurity must always 

 remain, regardless of the material under discussion. However, since 

 there is no reason to believe that such a situation actually prevails in 

 the case of the virus protein, we are unable, at the present time, to 

 conclude other than that the high molecular weight protein under 

 discussion is the virus. Now, these same tests that indicate that the 

 virus protein is homogeneous may be used to demonstrate that it 

 possesses the ordinary properties of molecules. As a matter of fact, 

 the chemist, after a perusal of the physical and chemical properties of 

 tobacco mosaic virus protein, has no difficulty whatsoever in coming 

 to the conclusion that, despite its huge size, it has all the properties of a 

 molecule and hence is a molecule. 



However, we must remember that in addition to the properties 

 which tobacco mosaic virus protein possesses as a molecule it also 



