Horizontal transmission of the entomopathogenic fungus, Metarhizium anisopliae (Metchnikoff) Sorokin, and hydramethylnon toxicant among individuals of the German cockroach, Blattella germanica (L.), was evaluated in this study. Transmission of hydramethylnon occurred through the feces. Contaminated feces were toxic to other cockroaches when mixed with standard laboratory diet at different ratios. Lethal time (LT50) of the nymphs increased as the proportion of contaminated feces in the healthy laboratory diet was decreased. When cockroaches were fed a diet consisting of hydramethylnon-contaminated feces and a laboratory diet at ratios of 1:0, 1:1, and 1:5, the mortality reached 100% at days 9, 12, and 17, respectively. The mortality was reduced to approximately 80% at a ratio of 1:10. Fifth-stage nymphs exposed to the conidia of M. anisopliae or hydramethylnon toxicant for 6, 12, 24, or 48 h transferred the fungal conidia or the toxicant to healthy nymphs. Rate of mortality increased significantly by increasing the ratio of infected to unexposed cockroaches (i.e., 1:1 ratio >1:10 ratio), and by increasing the exposure time for the infected cockroaches (48 h versus 6 h) to both M. anisopliae and hydramethylnon. Cockroaches killed by M. anisopliae or hydramethylnon before being presented to healthy cockroaches were less effective in spreading the fungus or toxicant than were live infected cockroaches. When live infected nymphs were mixed with healthy cockroaches, M. anisopliae initially killed cockroaches slightly faster (LT50[95% CL] = 10.1 [9.2–11.0] d) than hydramethylnon (LT50[95% CL] = 12.5 [11.4 - 13.7] d). However, cumulative mortality reached 100% at day 26 for both M. anisopliae and hydramethylnon treatments. After mixing healthy nymphs with dead infected cockroaches, the rate of mortality was slower for M. anisopliae (LT50 = 20 [19.1 - 20.9] d) and hydramethylnon (LT50 = 14 [13.4 - 14.7] d) than those recorded in the previous test (that is, mixing live-infected nymphs with healthy cockroaches). However, the cumulative mortality 28 d after exposure to M. anisopliae and hydramethylnon reached 77 and 67%, respectively. Fungal growth was observed on all body parts of dead infected nymphs within 14 to 16 d of exposure to M. anisopliae. Dead infected nymphs were not cannibalized suggesting avoidance behavior by healthy nymphs. Consequently, the fungal conidia were not spread as effectively by dead nymphs as with live infected nymphs. Factors affecting performance in light of horizontal transmission of M. anisopliae are discussed.