Scientists are users of science journals in two senses-to enter their
own work and to learn of others-but are rarely the direct purchaser of
them. The health of research is inextricably bound up with the learne
d journal. Much is contributed by scientists to their journals in invi
sible ways, such as refereeing and editing. The creation of an archive
of information is the primary service the journal supplies. Libraries
pay for its entry into the system; research organizations and univers
ity faculties must recognize that this cost has to be met. The views e
xpressed here are those of a publisher as well as of a user.