We present a kinetic model that can account for several experimental f
indings on short- and long-term potentiation (STP and LTP) and their p
harmacological modulation. The model, which is consistent with Hebb's
postulate, uses the hypothesis that part of the origin of LTP may be a
consequence of an increased release of neurotransmitter due to a retr
ograde signal. The operation of the model is expressed by a set of irr
eversible reactions, each of which should be thought of as equivalent
to a set of more complex reactions. We show that a retrograde signal a
lone is not sufficient to maintain LTP unless long-term change of the
rate constant of some of the reactions is caused by high-frequency sti
mulation. Pharmacological manipulation of LTP is interpreted as modifi
cations of the rate constants of one or more of the reactions that exp
ress a given mechanism. The model, because of its simplicity, can be u
seful to test more specific mechanisms by expanding one or more reacti
ons as suggested by new experimental evidence.