The color-magnitude diagram of the lower main sequence , as measured from a volume-limited sample of nearby stars , shows an abrupt downward jump between M _ { V } \sim 12 and 13 .
This jump indicates that the observed mass-radius relationship steepens between 0.3 and 0.2 M _ { \odot } , but theoretical models show no such effect .
It is difficult to isolate the source of this disagreement : the observational mass-radius relationship relies upon transformations that may not be sufficiently accurate , while the theoretical relationship relies upon stellar models that may not be sufficiently complete , particularly in their treatment of the complex physics governing the interior equation-of-state .

If the features in the observationally derived mass-radius relationship are real , their existence provides a natural explanation for the well-known gap in the orbital period distribution of cataclysmic variables .
This explanation relies only upon the observed mass-radius relationship of low-mass stars , and does not require ad hoc changes in magnetic braking or in the structure of cataclysmic variable secondaries .
If correct , it will allow broader application of cataclysmic variable observations to problems of basic stellar physics .