When the middle molecule (MM) hypothesis was formulated in 1975, no MM had yet been identified as a uremic toxin. Meanwhile, the birth and implementation of the Kt/V concept gained wide acceptance and has remained the world standard for assessing dialysis adequacy. However, over the past 20 years, accumulating evidence has made it clear that MM's are important uremic toxins, and that the dose of dialysis based on removal of small molecular substances does not protect against excessive hemodialysis mortality, morbidity, or the presence of uremic signs and symptoms. These poor results are, in one way or another, linked to the accumulation of MM's and other substances behaving like MM's, such as phosphate. Dialysis schedules yielding the best clinical results, such as longer dialysis and more frequent dialysis, favor increased removal of middle molecular substances. The observation that short daily dialysis is giving results similar to long nocturnal quotidian dialysis supports early observations that the volume from which middle molecular substances are extracted mainly by hemodialysis is small (about as large as the extracellular volume), and that transfer of MM's from cells to extracellular fluid is very slow. This behavior of MM's is markedly different from that of small molecular substances, which are more rapidly transferred from intracellular to extracellular compartments and are more readily extracted from total body water during hemodialysis. In order to achieve even minimum adequate dialysis, it is now scientifically validated that toxic MM's must be removed in larger amounts than currently attained. This can only be accomplished by long dialysis sessions with a 3-times per week schedule or more frequent dialyses. Five hours 3 times per week represents the absolute minimum treatment. Dialy sis 6 to 7 times per week is the ideal schedule for patients who are willing to commit the time and effort in exchange for maximum well-being and long survival.