b F. M. BURNET 



numbers of virus types yet to be examined. Until a much larger fraction of 

 the whole universe of viruses living in mammalian cells has been made 

 available for study, any general survey will be heavily biased toward a con- 

 sideration of types actively pathogenic for human beings and their less 

 dangerous congeners. 



Even here, however, some order can be seen. There is at least a suggestion 

 that ecological niche and physical structure have a significant relation to one 

 another, with the implication of common descent and evolutionary radiation 

 in the history of the group. 



The poxviruses are primarily skin pathogens, multiplying in epidermal 

 cells and spread by mechanical transfer to traumatized skin, often by mos- 

 quito or other biting vector. Depending on strain and host species, an addi- 

 tional viremic phase may or may not be manifest. Their complex structure, 

 presence of intrinsic components like copper and biotin, which point to an 

 autonomous intermediate metabolism of some sort, large size, and possession 

 of DNA point almost categorically to evolution by parasitic degeneration 

 from a bacterial form. 



The psittacosis viruses have even closer resemblance to small bacteria and 

 it would be easy to make a case for choosing Brucella suis, Coxiella burnetii, 

 psittacosis virus, and vaccinia virus as modern representatives of the general 

 line of evolution to vaccinia virus. 



The myxoviruses give every indication of being preeminently parasites of 

 respiratory mucous membranes with, however, potentialities for invasion 

 and generalized spread, as in fowlplague and mumps. They differ from all 

 the other groups in the relatively high content of host cellular material that 

 is included in the infective particle. This statement would perhaps be ques- 

 tioned by some, while others might remark that it may only be because 

 technical approaches to the influenza virus studies have not been applied to 

 other viruses that the apparent contrast exists. Everything, however, points 

 strongly to an active participation of the surface layers of the host cell in the 

 fabrication of the surface of the virus particle, whether spherical or filamen- 

 tary. Another striking feature is the lack of uniformity of size and the ease 

 with which functionally and/or morphologically anomalous forms can be 

 produced. The infective particles contain only RNA and are unique among 

 viruses in carrying an active specific enzyme, neuraminidase, presumably as 

 a surface component of the particle. 



The arborviruses have exploited the same niche for survival as an immense 

 group of parasitic protozoa, most of which are also obligate intracellular 

 parasites. It is of interest that the infective sporozoites of malaria injected by 

 the mosquito are very small, although of course much larger than any virus, 

 that they are stored in the insect's salivary glands, and that on entry into 

 the body they first undergo extraerythrocytic development in liver cells. 



