This letter presents wide-field optical and and NIR ( UBVRI { H } \alpha K ) images of the galaxy M74 which were taken between 0.6 - 8.3 years before the discovery of the Type Ic supernova 2002ap .
We have located the position of the supernova on these images with an accuracy of 0.3 ^ { \prime \prime } .
We find no sign of a progenitor object on any of the images .
The deepest of these images is the B - band exposure which has a sensitivity limit corresponding to an absolute magnitude of M _ { B } \leq - 6.3 .
From our observed limits , we rule out as the progenitor all evolved states of single stars with initial masses greater than 20M _ { \odot } unless the WR phase has been entered .
Two popular theories for the origin of Type Ic supernovae are the core collapse of massive stars when they are in the WR phase , or the core collapse of a massive star in an interacting binary which has had its envelope stripped through mass transfer .
Our prediscovery images would be sensitive only to the most luminous \sim 30 % of WR stars , hence leaving a substantial fraction of typical WR stars as viable progenitors .
The energetics measured from modelling the initial lightcurve and spectral evolution of SN 2002ap suggest an explosion of a 5M _ { \odot } C+O core .
While WR stars generally have measured final masses greater than this , the uncertainties associated with the explosion model , stellar evolutionary calculations and mass measurements suggest we can not definitively rule out a WR star progenitor .
The alternative scenario is that the progenitor was a star of initial mass \sim 20 - 25 M _ { \odot } which was part of an interacting binary and stripped of its hydrogen and helium envelope via mass transfer .
We discuss future observations of the supernova environment which will provide further constraints on the nature of the progenitor star .