100 F. B. BANG 



obtained by three separate groups of investigators (Bang, 1955b). Two types 

 of particles were found and may be interpreted as representing two phases of 

 development. One type was seen in paranuclear clusters of spherical (oval as 

 compressed in sectioning) particles, having an outside diameter of about 65 

 to 70 m/z. The particles were also scattered throughout the cytoplasm and 

 sometimes encircled small vesicles like a series of beads. They did not appear 

 to have any specific relation to intact mitochondria, but it is possible that 

 some of these small vesicles were remnants of degenerating mitochondria. 

 Pictures suggestive of this have been published (Bang et al., 1956a,b). 

 Bernhard et al., (1956b) has shown that this type of particle is actually 

 double-layered, with an internal diameter of about 38-50 nux and an external, 

 sometimes less dense layer of about 10 rn.fi,. 



2. Morphological Evidence of Virus Release 



The second group of particles is larger, was found either within vesicles, on 

 the surface of the cell, or attached to the microvilli of the cell. The possibility 

 that they are formed by the ejection of the smaller particle either from 

 microvilli (Bang, 1955a; Bang et al., 1956a,b; Bernhard et al., 1956b) or from 

 the free surfaces of the cell follows the pattern of virus release in other cell 

 infections (Figs. 16 and 17). Their presence as dense particles surrounded by 

 less dense "cytoplasm," yet still within the cell, might result from the projec- 

 tion of finger-like extrusions of the host cell cytoplasm into an intracellular 

 vesicle which was the residuum of a collapsed mitochondrial structure. There 

 is, however, no evidence of selective localization in or near mitochondria. 



Serial sections have shown that they are released all along the free surface 

 of a cell and that they may also project by a series of microvilli into fluid 

 pockets between cells (Bang et al., 1956a,b). The particles may be readily 

 distinguished from the larger, more homogeneous milk particles or droplets 

 the secretion of which entails a continuous breakdown of the cell surface 

 (Bang et al., 1956a,b; Bernhard et al., 1956b). 



Except for one preliminary, unconfirmed report (Kinosita et al., 1953), 

 there is no record of their presence in the nucleus of the host cell. 



Since identical particles have been found by three different groups in a 

 number of mammary milk factor tumors and rarely, if ever, in normal cells, 

 and since their characteristic shape, arrangement, and apparent release from 

 the cell surface follow those of other viruses, they are the most promising 

 candidates for identification as the virus of mammary tumors. However, 

 present biological knowledge of the virus is limited to factors known to be 

 transmissible by milk and possibly by seminal fluid. Again, there is agree- 

 ment (Bang et al., 1956a,b; Bernhard et al., 1956b; Dmochowski et al., 1955) 

 that identical particles are found in mammary tumors which lack the milk 

 factor. Either these particles do not represent the milk factor or they may 



