5 
Seventh, It is not clear how well X1876 dies in the presence of other 
bacteria in a rich environment lacking the growth factors required for 
viability* **, since in the direct measurement of death rate, the conclusion 
that x!876 dies rapidly in the presence of }@89 is based upon a single 
datum-point. This rate of death does not agree with the rate of plasmid 
transfer observed in conjugation experiments performed unier the same 
physiological conditions, assuming that conjugational transfer' is limited 
by viability. Furthermore, the extent to which phenotypic rescue by 
other bacteria of the death of X1S76 in a rich environment lacking the 
growth factors required for viability is due to cross-feeding is not 
measured (see this paper, §§A8c, A9b, A9c, B26). 
Eighth, the ability of xl?76/xl876 to transfer genetic information 
by known resistance (R) factor transfer systems is reduced only by 
one-hundred fold at best at at least one of the two temperatures, 32°C 
and 37°C, for the vast majority of cases, when compared to the unweakened 
parent (see this paper, §§A15, A16, A17, A18, A19, A20, A21, B25, B28, 
B29) . 
Ninth, since markers are frequently lost during routine manipulations 
of a complex auxotroph, especially those markers detrimental to the 
survival of Escherichia coli ( see this paper §B24), it is not certain 
how handling of xl?76/xl876 8y many laboratories will alter the genotypic 
and phenotypic properties of this strain. How dispensation and use of 
any EK-2 candidate system will alter the genotypic and/or phenotypic 
properties of the system must be assessed prior to certification. 
and glc (glycerol or glucose ?). We cite other internal references 
that might bear on this point, such as the survival of (>f y).81 6 
during drying conditions (§A13, Curtiss £t^ al. , Figure 28). 
** See Curtiss et^ al . , Figure 1C. 
[Appendix C — 193] 
