The number of different interaction combinations between bacteria and plants and the diversity of types of bacteria with respect to mode of pathogenesis indicate that plants possess various mechanisms for resistance to bacteria. On the basis of present evidence, bacteria may be prevented from causing disease in plants by one of two types of resistance system: constitutive or induced. Constitutive systems may involve: (1) inhibition of bacteria by pre-formed compounds that are toxic
per se
or which are converted rapidly to toxic products when cells are injured; or (2) a combination of adverse physiological factors that currently remain ill-defined. Among examples cited, details are presented of recent research concerning the possible role of a constitutive system in the resistance of corn to soft rot bacteria in the genus
Erwinia
. Corn (maize) plants which were highly susceptible to one pathotype of
E . chrysanthemi
were found to be highly resistant to other pathotypes of
E. chrysanthemi
and to other species of
Erwinia
. A differentially inhibitory fraction (d.i.f.) extracted from corn plants was more toxic to soft rot
Erwinia
spp. than to the corn-stalk rot pathogen. Other plant pathogenic bacteria that do not attack corn and certain (but not all) saprophytic bacteria tested were also inhibited by the d.i.f. In contrast, all of the bacterial corn pathogens that were tested were similar to the corn-stalk rot pathogen in their relative insensitivity to d.i.f. Studies with the arbutin-hydroquinone complex in
Pyrus
spp. also provide evidence for a constitutive type of system for resistance to
Erwinia amylovora
. However encouraging recent investigators of these types of systems may be, unequivocal demonstration of the specific manner in which a preformed system operates
in vivo
is still needed. Induced resistance systems include: (1) the hypersensitive reaction (h.r.), and (2) the protection response. The h.r. is elicited by most phytopathogenic bacteria (in particular, pseudomonads, xanthomonads and
Erwinia amylovora
) when introduced into non-host plants, but not by most saprophytic bacteria. Recent studies of different races and strains of the wilt pathogen,
Pseudomonas solanacearum
, as well as with other bacteria, indicate that there are certain exceptions to the generalization that pathogenic bacteria, or avirulent variants of pathogens, will induce a hypersensitive reaction in non-host plants. Isolates of the banana race of
P. solanacearum
(non-pathogenic to tobacco) consistently induced an h.r. in tobacco leaves; whereas the potato race (also non-pathogenic to tobacco) did not produce this reaction. The second main type of induced resistance, the protective reaction, can be induced by prior infiltration of the inoculation court with living cells of avirulent or incompatible strains and, in some instances, saprophytes. It can be induced also by heat-killed cells of bacterial pathogens. Aspects of the protective reaction have been examined with various isolates of
P. solanacearum
. Infiltration of tobacco leaves with heat-killed cells of the wilt pathogen before infiltration with cells of the banana strain (non-pathogenic on tobacco) prevented the h.r. from developing. It was also possible by this treatment to prevent the virulent tobacco strain from infecting the leaf. The protective response, which was light dependent, was detected in areas adjacent to infiltrated tissues and in other adjacent leaves 2-3 days after treatment. Similar effects were obtained with a relatively heat-stable protein fraction obtained from cells of an avirulent strain of
P. solanacearum
. Recent advances in studies with bacterial pathogens reveal the possibilities of using these systems in order to gain new insight into the nature of resistance to pathogens in general.