An extremely simple but instructive, ''toy'' model is presented which
shows that a small excess of pions in the nucleus can produce a signif
icant change in the values expected for the Gottfried sum rule. The ge
neral question of the convergence of the sum rule and of the convergen
ce of the experimental integral is also discussed. It is demonstrated
that conclusions about the sum rule, based on deuterium data, are surp
risingly model dependent. In contrast, it is stressed, that the Bjorke
n sum rule can be tested significantly using deuterium data.