POPULATION GENETICS 



it impossible to be certain that a single successful mutation 

 might not also have rapidly spread over the continent. 

 Irrespective of whether it was of single or multiple origin, the 

 appearance of a less virulent strain, which can replace the 

 more virulent, raises some interesting problems that are 

 highly relevant to our general theme of clonal selection. 



To understand the situation, we must first consider the 

 process by which myxomatosis is spread in Australia. Acute 

 epizootics nearly always occur in summer and are always 

 associated with a local prevalence of a mosquito which will 

 feed freely on rabbits. The dominant vector is an Anopheles 

 {annulipes) which has developed the habit of resting in rabbit 

 burrows during daytime. Fenner, Day and Woodroofe ( 1 952) 

 have made a detailed study of the process of transmission 

 by mosquitoes. No biological cycle is involved : the virus is 

 not taken in with the blood and there is no multiplication 

 in the tissues of the vector. The process seems to be purely 

 mechanical. The mosquito feeds mainly round the face and 

 ears of the rabbit, and in an animal with myxomatosis lesions 

 the insect will probe through heavily infected tissue to reach 

 blood. The mouth parts will be lightly contaminated with 

 virus which will adhere when the stylet is withdrawn. Once 

 a mosquito has fed through an infected area of skin it is liable 

 to infect any subsequent rabbit on which it feeds. In Fenner's 

 experience, however, the proportion of infective bites is 

 relatively low, and though ' takes ' can be obtained as long as 

 25 days after the infecting feed, they become less and less 

 frequent. The obvious implication is that in at least the later 

 stages almost all infections are initiated by a single virus particle. 



This provides the clue to the possibility of replacing 

 a virulent strain by a less virulent one. We may suppose that 

 in a standard 99*5% strain of myxomatosis virus, mutation 

 occurs to types of significantly lower virulence at a rate around 

 10-^. If transfer to new individual hosts is always by a small 

 number of particles, and usually by one only, then we can 

 reasonably assume that about once in 10^ transfers a rabbit 



25 



