The summary table shows a chronological order only of studies that report ganglion cell number as a percentage of neurons in the GCL. The Method column refers to the general method used to identify retinal ganglion cells in either the mouse of rat retina. Details of the methods used in each study are: ^aTotal non-vascular cells were counted from Nissl-stained retinal whole mounts. RGCs were identified by Horse Radish Peroxidase (HRP) retrograde labeling, where the optic track on one side was macerated at the level of the Lateral Geniculate Nucleus and 1 mg of HRP was applied to the lesion. A total of 12 mice were examined (12 retinas with HRP and the contralateral eye for Nissl). Of note, 69% of the cells with Nissl-substance were also HRP labeled. The non-HRP Nissl+ labeled cells were on average 21% smaller in area. Sample sizes were estimated at 6.1% of the retina. The source of the mice was not given. ^bTotal neurons were counted in 9 retinal whole mounts by labeling nuclei with an ethidium homodimer. Neurons were classified by morphological criteria of the nuclear stain (round to oval nucleus and prominent nucleoli). Ganglion cell numbers were estimated from axon counts of 3 optic nerves, which were evaluated by transmission electron miccroscopy (osmium tetroxide post-fix and uranyl acetate stain). The source of mice was Charles River. ^cControl mouse retinas were subjected to in situ hybridization for Thy1 mRNA using a block tissue staining protocol. After staining, the retina blocks were embedded in JB-4 Plus plastic and sectioned. Sections were counter stained with DAPI to identify nuclei. A total of 8 control retinas were examined, with 12–15 sections per retina being analyzed (cells counted along a 400 µm length per section). Of note, after optic nerve crush, these studies noted a loss of 57.8±8.1% of the neurons in the GCL after 3 weeks post optic nerve crush. Total neurons and RGCs were not estimated. The source of mice was an inbred colony of the investigator. ^dSame in situ hybridization protocol as above (c) except that eyes were from rats with ocular hypertension. Control eyes (n=4) had the higher percentage, while eyes with minimal damage (based on semiquantitative evaluation of paraphenaline diamine-stained sections of optic nerves, n=5) had the lower percentage. Data was not significantly different. The source of rats was not given. ^eTotal neurons were counted from Nissl-stained retinal whole mounts prepared from ~30 mice each individual strain (~10% retinal area surveyed per retina). Ganglion cell numbers were estimated from silver impregnated sections of 3 sections each of 5 separate nerves from each individual strain (9 1000×fields per nerve). The source of mice was Jackson Laboratories. ^fCells in the GCL were double labeled using a retrograde approach by dipping the optic nerve stump of enucleated eyes into a mixture of Lucifer Yellow (LY, stays in ganglion cells) and neurobiotin (NB, leaks out of ganglion cells). LY and NB double-labeled cells were considered ganglion cells. TO-PRO-3 was used as a nuclear counter stain. Sample size estimated at ~10% of the retina and a total of 9 mice were analyzed. The source of mice was Jackson Laboratories. ^gIn the first method reported in this study, total neurons (per field) were identified in 4 retinal whole mounts based on nuclear morphology after DAPI counter stain. Ganglion cells were identified by immunostaining the wholemount for SNCG protein using a commercial antibody. No indication was given of the numbers of fields counted for the immunostained samples. The total neurons and retinal ganglion cells were not estimated using this method. ^hIn the second method reported in this study, values for total neurons and total retinal ganglion cells were extrapolated from data presented for wild-type mice in this report. They were not used to calculate the percentage of ganglion cells by the authors. Total ganglion cell numbers were estimated from osmium and toluidine blue stained sections of 10 optic nerves. Total neurons were extrapolated from DAPI-stained whole mounts of wild-type mice (Table 1, 6596 neurons were counted, which was estimated as 7% of the total retina).