We critically examine the evidence available of the early ideas on the bending of light due to a gravitational attraction, which led to the concept of gravitational lenses, and attempt to present an undistorted historical perspective. Contrary to a widespread but baseless claim, Newton was not the precursor to the idea, and the first Query in his {sl Opticks} is totally unrelated to this phenomenon. We briefly review the roles of Voltaire, Marat, Cavendish, Soldner and Einstein in their attempts to quantify the gravitational deflection of light. The first, but unpublished, calculations of the lensing effect produced by this deflection are found in Einsteins 1912 notebooks, where he derived the lensing equation and the formation of images in a gravitational lens. The brief 1924 paper by Chwolson which presents, without calculations, the formation of double images and rings by a gravitational lens passed mostly unnoticed. The unjustly forgotten and true pioneer of the subject is F. Link, who not only published the first detailed lensing calculations in 1936, nine months prior to Einsteins famous paper in {sl Science}, but also extended the theory to include the effects of finite-size sources and lenses, binary sources, and limb darkening that same year. Link correctly predicted that the microlensing effect would be easier to observe in crowded fields or in galaxies, as observations confirmed five decades later. The calculations made by Link are far more detailed than those by Tikhov and Bogorodsky. We discuss briefly some papers of the early 1960s which marked the renaissance of this theoretical subject prior to the first detection of a gravitational lens in 1979, and we conclude with the unpublished chapter of Petrous 1981 PhD thesis addressing the microlensing of stars in the Magellanic clouds by dark objects in the Galactic halo.