With reference to a particle tunneling through two successive barriers, it seems to have been generally accepted that the tunneling time does not depend on the separation distance between the barriers. This phenomenon has been called the {\it generalized Hartman effect}. In this letter, we point out a lack of mathematical rigour in the reasoning by which this effect was deduced about ten years ago. A mathematically rigorous treatment shows us that the tunneling time does indeed depend on the length of the free space between the barriers.
We comment on recent numerical studies concerning coupled 1D electron liquids (F. Mila and D. Poilblanc, Phys. Rev. Lett. 76, 287 (1996)). In particular, we point out that the importance of integrability observed in the results of these authors for the quantity
P(t) is really an indication of the importance of integrability for questions of ergodicity, rather than questions of coherence. We discuss why this is so, and why the quantity
<δN(t)> is a more appropriate function for determining whether or not interliquid hopping is coherent.