cell. The probability of such an occurrence (precise excision and 
ligation) theoretically is expected to be very rare. As mentioned 
above, Agracetus scientists have not seen any gall spontaneously 
appearing in at least a thousand HADH2 tobacco plants examined 
over the past two years. If one assumes that a plant contains 100 
million cells and Agracetus scientists have looked at 1,000 plants, 
then one may estimate that the chance of a precise excision 
occurring is less than one in 100 billion cells. 
However, if one assumes that an exact excision will occur at some 
finite frequency (perhaps in one cell from one thousand plants), it 
is possible that a crown gall might appear from that cell. The 
only symptom that should result would be a single crown gall. 
There is no reason that such a crown gall will spread to other sites 
on the same plant. No Ti plasmid virulence genes are present. 
Unlike'a true crown gall, no Agrobacterium cells are present to 
initiate new galls in the areas of the plant that may become 
injured. Crown galls that Agracetus scientists have induced (by 
inoculating Agrobacterium into injured sites) in non-genetically 
engineered tobacco lines have, in no case, spread to other sites on 
the same plant. It is even more unlikely that the crown gall will 
spread spontaneously to another susceptible plant. In Agracetus' 
crowded greenhouses, where crown gall infected plants have been 
randomly spaced among plants susceptible to crown gall, there has 
been no case of a crown gall appearing in a plant not intentionally 
infected (by wounding) with Agrobacterium . Many experiments by 
other scientists cited in the published scientific literature confirm 
Environmental Assessment 
Page 35 
[505] 
