The paper describes the algorithm used to select the main galaxy sample for spectroscopy in the Sloan Digital Sky Survey (SDSS) from photometric data. The selection is based on the Petrosian magnitude system, which measures flux in apertures determined by the shape of the surface brightness profile. The main galaxy sample consists of galaxies with \( r \)-band Petrosian magnitude \( r \leq 17.77 \) and \( r \)-band Petrosian half-light surface brightness \( \mu_{50} \leq 24.5 \) magnitudes per square arcsec. This selection criterion ensures a median redshift of 0.104 and targets about 90 galaxies per square degree. The algorithm is designed to be uniform and objective, with a high completeness rate exceeding 99%. Tests demonstrate that the star-galaxy separation criterion effectively eliminates stellar contamination while removing minimal genuine galaxies, the fraction of galaxies eliminated by the surface brightness cut is very small (\(\sim 0.1\%\)), and the reproducibility of target selection based on repeated imaging scans is consistent with random photometric errors. The main cause of incompleteness is blending with saturated stars, which becomes more significant for brighter, larger galaxies. The SDSS spectra have a high signal-to-noise ratio (\( S/N > 4 \) per pixel), allowing for reliable redshift measurements for essentially all targeted galaxies (99.9\%). About 6\% of galaxies that satisfy the selection criteria are not observed due to companions closer than the 55'' minimum separation of spectroscopic fibers, but these can be accounted for in statistical analyses. The uniformity and completeness of the galaxy sample make it ideal for studies of large-scale structure and galaxy population characteristics in the local universe.The paper describes the algorithm used to select the main galaxy sample for spectroscopy in the Sloan Digital Sky Survey (SDSS) from photometric data. The selection is based on the Petrosian magnitude system, which measures flux in apertures determined by the shape of the surface brightness profile. The main galaxy sample consists of galaxies with \( r \)-band Petrosian magnitude \( r \leq 17.77 \) and \( r \)-band Petrosian half-light surface brightness \( \mu_{50} \leq 24.5 \) magnitudes per square arcsec. This selection criterion ensures a median redshift of 0.104 and targets about 90 galaxies per square degree. The algorithm is designed to be uniform and objective, with a high completeness rate exceeding 99%. Tests demonstrate that the star-galaxy separation criterion effectively eliminates stellar contamination while removing minimal genuine galaxies, the fraction of galaxies eliminated by the surface brightness cut is very small (\(\sim 0.1\%\)), and the reproducibility of target selection based on repeated imaging scans is consistent with random photometric errors. The main cause of incompleteness is blending with saturated stars, which becomes more significant for brighter, larger galaxies. The SDSS spectra have a high signal-to-noise ratio (\( S/N > 4 \) per pixel), allowing for reliable redshift measurements for essentially all targeted galaxies (99.9\%). About 6\% of galaxies that satisfy the selection criteria are not observed due to companions closer than the 55'' minimum separation of spectroscopic fibers, but these can be accounted for in statistical analyses. The uniformity and completeness of the galaxy sample make it ideal for studies of large-scale structure and galaxy population characteristics in the local universe.