ARTHROPOD RELATIONSHIPS WITH VIRUSES 25 
the time required for the treated insects to regain infective power is 
a latent period while the virus is multiplying up inside them to the 
necessary concentration for infection and explains the inability of the 
insects, after being heated for twelve days, to regain infective power 
at all on the assumption that all the virus has been destroyed. This 
explanation may be the correct one but it seems somewhat facile, 
since we do not really know what the effect of heat on virus and insect 
may be. It is possible, for instance, that the virus is destroyed in one 
part of the insect only and the time lag is due to the necessity of more 
virus moving to take its place. 
On the other hand it is quite clear that many plant viruses do not 
multiply inside their insect vectors and indeed are rapidly lost, prob- 
ably by the action of digestive enzymes. As regards certain viruses 
which are retained for long periods by their vectors, that of sugar-beet 
curly-top for example, there is some evidence that the prolonged 
infectivity is due to a storage, rather than multiplication, of virus. 
This is suggested by the fact that there seems to be some correlation 
between length of time of feeding on a source of virus and the length 
of time the virus is retained by the insect. Again, if there is multi- 
plication of the curly-top virus in the leaf-hopper vector, it is not 
sufficient to keep the virus content of the insect at infection point, 
since in a series of daily transfers to healthy plants the power to infect 
generally falls off. 
Mechanical Transmission of Virus by Insects 
There is probably no sharp line of demarcation between a purely 
mechanical transmission of viruses by insects and transmission by 
specific insects only, for between these two extremes can be found 
intermediate relationships. Presumably the least mechanical form of 
transmission is found in those insects which retain virus for very long 
periods, or in those where multiplication is known to take place, or 
where it can be passed on to the progeny. It has been suggested that 
rickettsiae were originally insect parasites and certainly the louse is 
doomed to die from the typhus rickettsiae more certainly than the 
human being it may infect. But this seems to be the only instance 
known where an insect vector is adversely affected by the virus it 
carries. 
Possibly a truly mechanical vector would be one which is not para- 
sitic on the organism it infects, as, for example, house flies and other 
non-blood-sucking muscid flies which are said to transmit the virus 
