VIRUSES — STANLEY 267 



Among these are alfalfa mosaic virus, which has a molecular weight of 

 about 2 million and a diameter of about 16 mix, and the Shope rabbit 

 papilloma virus, with a molecular weight of about 25 million and a 

 diameter of about 40 m/x. The elementary bodies of vaccinia have a 

 diameter of about 225 ni/x. There is, therefore, a group of rod-shaped 

 viruses and a group of viruses which are essentially spherical in shape, 

 although with the development of more precise techniques some of the 

 latter may be found to be definitely ellipsoidal in shape. It should be 

 emphasized that each virus has a shape and a size which appear to be 

 quite definite and characteristic, regardless of the conditions or the 

 host in which the virus is produced. However, neither this statement 

 nor the statements relative to the homogeneity of virus preparations 

 in the ultracentrif uge or electrophoresis apparatus are meant to imply 

 that all the particles in a given virus preparation are exact replicas. 

 The fact that variants continually arise during the production of virus 

 would always insure the presence in purified preparations of a small 

 amount of closely related although slightly different particles. There 

 is other evidence, such as that obtained by Loring in solubility studies 

 on tobacco mosaic virus, which indicates that the purified virus prepa- 

 rations are not absolutely homogeneous chemically but consist of a 

 family of very closely related structures. The general situation may 

 not be far different from that which is now known to exist in the case 

 of even very simple structures, such as sulfur, nitrogen, and hydro- 

 gen, where families of isotopes are the rule rather than the exception. 

 In this connection, it may be stated that the problems and relation- 

 ships which obtain with the tremendous virus structures are not well 

 clarified at present. However, from a practical standpoint, there has 

 been little difficulty as yet, for there are several instances in which, 

 according to present techniques, there is a very high degree of chemical 

 homogeneity. 



Tobacco mosaic, bushy stunt, and other viruses which have been ob- 

 tained in purified form are good antigens. It is necessary, however, 

 to inject animals with the viruses, for antibodies do not appear to be 

 produced in plants. This may be due to the nature of plants for, 

 despite much effort, no conclusive proof of the existence of antibodies 

 in plants has been obtained, although Wallace secured some suggestive 

 results with curly top virus. Tlie serum of a rabbit injected with 

 tobacco mosaic virus gives a specific precipitate with tobacco mosaic 

 virus and specifically neutralizes tobacco mosaic virus activity. This 

 reaction has been studied Avith Dr. Anderson by means of the electron 

 microscope and the ultracentrifuge. Electron micrographs of a mix- 

 ture of virus and normal rabbit serum show virus particles of normal 

 size, indicating little or no adsorption of particles from normal serum 

 on the virus molecules (pi. 4, fig. 1). The sedimentation constant of 

 tobacco mosaic virus is essentially unchanged in mixtures containing 



