The analysis presented here suggests that the organization of agonistic behavior is remarkably similar across all species of muroid rodents. This conclusion contradicts an impression given in many textbooks that the social behavior of mammals is characterized by "species-specific behaviors." Instead, it emphasizes the underlying unity of mammalian social behavior and its neural substrate.
The differences in agonistic behavior among muroid rodents is often greater as a function of developmental differences than phylogenetic differences. This is particularly evident for the behaviors that are altered by bringing animals from a wild state into a laboratory environment. Thus, many of the motivating stimuli for defense and submission are not effective in laboratory animals because their activity has been restricted to an environment in which escape has not been possible. Also, the relative familiarity of the odors of all of the animals in a laboratory colony may exert a profound effect upon agonistic behaviors; as a result, patrol/marking and offense in laboratory rats and mice have become dependent upon pheromone-dependent odors for motivating stimuli rather than unfamiliar conspecific odors, and defense has become replaced completely by submission.
Keeping in mind that differences may reflect developmental as well as phylogenetic processes, there are four categories of behavioral differences across species that may be treated in some detail. The first category includes motor patterns of scent-marking that occur in a bewildering variety of forms often more different in closely related species than in distantly related species. The second category includes submissive motor patterns and their effectiveness in inhibiting or precluding offense. The third category concerns motor patterns of offense and submission that serve as threats. And the fourth category concerns alarm signals.
The great variability of the motor patterns of scent-marking parallels an equally great variability in the anatomical distribution of the glands that presumably secrete the pheromones broadcast by the scent-marking. Thus, the sebaceous glands of the oral lips and angle that presumably provide pheromones for the scent-marking patterns of facial grooming and cheek-rubbing, are so varied among genera and species that Quay  has proposed a "thesis of plasticity and reversibility of sebaceous gland evolution." The preputial gland, which may provide pheromones for urine-marking or perineal drag scent-marking, is also quite variable; it is present in at least six of the nine rodent families analyzed by Brown and Williams  , but in none of these families with more than one analyzed genus are the preputial glands universally present. One may conclude in this case that the differences in anatomical distribution of the gland become greater rather than less as one proceeds with the analysis to finer and finer taxonomic units. Similarly, the midventral sebaceous glands, which may provide secretions for the ventral-rub scent-marking, differ greatly among closely related species of Peromyscus [Doty and Kart, 1972] and Rattus [Rudd, 1966]. Finally the posterolateral sebaceous glands, which would provide secretions for flank rub or sand-bathing, differ greatly among various genera and sometimes even species of the Microtinae [Quay, 1968]. Although there is also a great deal of variability in glands of the male reproductive tract in muroid rodents [Arata, 1964; Linzey and Layne, 1969], it has been proposed that this is due to the diverse roles that these glands play in the production of a copulatory plug [Hartung and Dewsbury, 1978].