Nine currently available methods for HRP neurohistochemistry have been compared with each other on matching tissue sections from four rats and four rhesus monkeys. The nine methods investigated in this report are the diaminobenzidine (DAB) procedures of LaVail JH and LaVail MM, of Adams JC and of Streit P and Reubi JC, the benzidine dihydrochloride (BDHC) procedures of Mesulam M-M and of De Olmos J and Heimer L; the o-dianisidine (O-D) procedure of De Olmos J; the p-phenylenediamine dihydrochloride and pyrocatechol (PPD-PC) procedure of Hanker JS et al., and tetramethyl benzidine (TMB) procedures of Mesulam M-M and of De Olmos J et al. Quantitative comparisons were based on counts of retrogradely labeled perikarya. The extent of anterograde transport and the size of the injection site were also compared at a more qualitative level. The results indicate that one TMB procedure is distinctly superior to each of the other eight procedures in the number of labeled perikarya that it can demonstrate. Furthermore, these differences are statistically significant at better than the 0.05 level of confidence. Differences in sensitivity are most evident when the perikarya contain small quantities of transported HRP. The same TMB method also demonstrates more anterograde transport and a larger injection site than all other procedures. If less sensitive procedures are employed, afferent or efferent connections that are clearly demonstrated by this TMB procedure are either underestimated or completely overlooked. It is suggested that sensitivity in HRP neurohistochemistry is determined by multiple factors which include the method of fixation, postfixation storage, the choice of chromogen, the incubation parameters, the type of HRP enzyme that is administered, and the postreaction treatment.