CALIFORNIA INSTITUTE OF TECHNOLOGY 
PASADENA. CALIFORNIA 91125 
DIVISION OF BIOLOGY 
May 3, 1976 
Dr. Donald Fredrickson 
National Institutes of Health 
Bethesda, Maryland 20011+ 
Dear Don: 
I expect you have a few other matters to occupy your time in addition to the 
recombinant DNA issue. However, this is and will doubtless continue to be an 
important issue and I therefore feel obliged to comment upon Bernie Davis' 
recent letter to you on this subject (of which he sent me a copy). 
Bernie takes issue with my "speculative suggestion" of a possible genetic 
barrier between prokaryotes and eukaryotes which may be of some importance in 
the prevention of genetic or viral interaction between these great classes of 
living organisms — and, if so, might be breeched only at considerable peril. 
His characterization of this hypothesis as a "speculative suggestion" is quite 
correct; the distinctions, as yet poorly defined, in the control of DNA repli- 
cation and transcription between these classes may be of minor biological 
importance (indeed I feel obliged to hope this will prove to be the case). 
But I do not believe anyone can state that conclusion with confidence at this 
time. We just do not know. And I do not find the arguments in Bernie 's letter 
particularly persuasive. 
The possibility that some genetic exchange between prokaryotes and eukaryotes 
does occur at least at historical rates was previously proposed by David Baltimore. 
Obviously this possibility cannot be excluded but neither can the rate be defined. 
I do not wish to enter into a numbers game of 10 -x , although I could take 
exception to some of the x's used in Dr. Davis' letter. I would only point out 
that the (very small) likelihood that E. coli can take up and incorporate exogenous 
DNA is known to be grossly influenced by a) the culture conditions in which the 
cells are grown, b) the ionic conditions of exposure to the DNA, c) the condition 
of the surface components of the cell, d) the presence or absence of external 
nucleases, e) mutations affecting the cell surface, cellular nucleases, enzymes 
involved in recombinatory processes, etc. In these circumstances it is simply 
impossible to predict the likelihood of such events in the intestines of man or 
other animals. As Dr. Davis says, it is also impractical even to attempt to 
obtain an empirical estimate. 
Dr. Davis also advances the evolutionary argument that we need not be concerned 
about possible hazards from organisms containing recombinant DNA since (in his 
view) opportunities for the formation of such organisms have occurred (repeatedly?) 
[ 537 ] 
