Biological and behavioural studies of rodents as a basis for control

Bull. Org. mond. Sante 1972, 47, 281-286 Bull. Wid Hith Org. Biological and behavioural studies of rodents as a basis for control W. B. JACKSON 1 Domestic rodents, particularly those living in urban populations, represent a serious public health problem, and effective control measures are required to deal with this threat to human health. Because of the characteristic interactions between individual animals, certain behaviour patterns occur in rodent populations that are of particular concern to control biologists. The genus Rattus is an extremely diverse group whose ecological require- ments are variable and flexible, while the genus Mus, on account of its small size, limited range, and modest requirements in terms offood and water, is also very difficult to control. For any control operation a knowledge of the growth dynamics of domestic rodent popu- lations is needed; after a period of logarithmic growth, population increases cease when the habitat is fully exploited. Consequently, control operations that merely remove some ani- mals lead only to renewed population growth. Trapping, poisoning, and predation are traditional control measures of this kind. Environmental modification is a more certain, but more difficult, approach. The use of chemosterilants offers some hope of radical control in the future, but at present, although some field trials have been made, these substances are not availablefor general use, one reason being their lack ofspecificity. Anotherproblem connected with the use of chemosterilants is that, on account of the sexual behaviour and physiology of domestic rodents, it would be necessary to reach nearly 100 % of the popu- lation to obtain effective control. JA population of rats is more than a collection of individual animals; certain characteristic interactions result in the appearance of behaviour patterns that are ofparticular concern to control biologists. Urban or rural rat populations consist ofmany small groups of rats that may have little or no contact with one another, but the general principles that apply to one group usually apply to other groups. We can therefore make broad generalizations and consider means of controlling urban rat populations. The number of rats able to exist in given circum- stances is defined by the carrying capacity of the habitat. Liebig's law of the minimum-namely, that the single factor (food, water, or harbourage) in least supply would control population size-was once believed to operate, but Shelford's law of tolerance (Shelford, 1952), which indicates that the interaction of several factors is important, seems to provide a better explanation. Thus, the amount of 1 Professor of Biology, Bowling Green University, Ohio, USA. food or shelter available to the rats is more critical during very cold weather than under less difficult conditions, and a parasite load in a rat under behavioural or physiological stress is more damaging to a subordinate animal than to a dominant one (Davis & Read, 1958). While this concept is diffi- cult to define with precision, it represents a realistic approach to the problem. Rat populations grow in an essentially logistic pattern until the limit of the carrying capacity of the habitat is approached. At this point population growth ceases, the population-controlling factors of death and replacement being, for practical purposes, in balance. During the period of logarithmic popu- lation growth contacts between members of the population are also increasing in number logarith- mically. Feedback phenomena, both behavioural and physiological, may be increasingly important in limiting population growth at higher population densities (Christian et al., 1965). In most situations the rat population displays 2897 -281- W. B. JACKSON a social organization characterized by dominant- subordinate relationships. The dominant animals, usually the larger, more aggressive individuals, have freer access to food, harbourage, and mates than the subordinates do. The subordinate animals are therefore forced to occupy secondary habitats, and these animals are the ones most likely to migrate and most vulnerable to predation. In control operations involving some form of "predation " (e.g., traps or poisons), the population usually reverts to that phase of the logistic growth curve characterized by maximum population growth (Davis, 1952). Thus, one might say that control operations benefit the rats, merely altering the status of the populations. In course of time, it will be necessary to repeat the operations. Modification of the environment in order to reduce its capacity to support rat populations is, in theory, the best approach, since permanent results will be achieved. However, structural changes are often required and may be expensive; they may take a considerable time to carry out and may not show a quick profit. The control operator and the customer are therefore often not interested in making such an effort. Furthermore, environmental management requires a high standard of maintenance for both structure and sanitation. The use of chemosterilants perhaps combines the best features of both approaches (Marsh & Howard, 1970). By rendering the rodent population incapable of reproducing, stressful levels of behaviour are maintained; the best harbourage is occupied by dominant individuals, and without recruitment the population gradually declines. However, before these chemicals become useful in rodent control some serious obstacles must be overcome. The taste of chemosterilants is objectionable to rats and some method of masking the taste, perhaps by micro- encapsulation, is needed. A second problem concerns the sexual behaviour of rats. Female rats in oestrus mate several times; indeed, the copulation frequency for female rats in oestrus may be higher than 90 per hour, many different males in high density populations being in- volved. Pseudopregnancy may not be induced by the initial matings and, for this reason, the chemosterilant will have to reach virtually 100% of the population to be completely effective, particularly if the popu- lation is in the logarithmic phase of growth. Stable (high density) populations would represent a similar control problem. Populations in this state are suffering from behaviourally induced sterility, or at least they are reproducing at a rate well below their potential rate. When rats in the population die or emigrate, the lost animals are replaced by an increase in the rate of reproduction. Chemosterilants may prove to be most effective in tropical environments where rat populations tend to be stable; i.e., the life expectancy for individual animals in the population is longer and mortality rates are lower than in urban populations in tempe- rate climates (Jackson, 1967). Thus, the use of chemosterilants would supplement the behavioural- physiological inhibition of reproduction. Urban populations of rodents are more often kept in the logarithmic phase of growth by changing environ- mental conditions; here, the use of chemosterilants would be more difficult since reproductive and migration rates would tend to be higher. More frequent applications of the chemosterilant would certainly be required. In all cases it would be necessary to apply chemo- sterilants in relatively large areas; application to a single rice paddy or garden, for example, would be unsatisfactory on account of migration from peri- pheral areas. A large proportion of the rodent population must be affected, and steps should be taken to prevent migration. Chemosterilants are of many different kinds and they act in various ways. Induced sterility may be temporary or permanent, and may affect males or females. The lack of species specificity has delayed the operational use of this technique on account of the risk to other animals, including man. How- ever, field trials have been undertaken in Hawaii, California, New York, and probably elsewhere, and the application of chemosterilants in rodent control operations, under careful supervision, may be possible in the near future (Brooks & Bowerman, 1969). Other methods for the biological control of rodents have been tried or suggested. Predators have general- ly failed: mongooses and monitor lizards were found to turn to more easily obtained prey; hawks, owls, snakes, and other predators may take the surplus population but rarely eliminate the breeding stock; pathogens either had little effect (Salmonella) or were highly pathogenic to man (e.g., plague). Cats and dogs have the reputation of being effective predators, and some may be (Elton, 1953). It was found in Baltimore that cats caught only the surplus rats that would have died or emigrated (Jackson, 1951); in the South Pacific area, Storer et al. (1962) 282 BEHAVIOURAL STUDIES OF RODENTS found as many rats in dwellings with cats as in unprotected houses, although the owners often felt more secure in the knowledge that the cat was there. Some surveys have reported the amazing effectiveness of cats in controlling rodents in many parts of the world, but the writer has often wondered whether some of those surveys were influenced by what the respondent thought the answer ought to be. A recent attempt at biological control on several islands in the Pacific area has involved the intro- duction of weasels (Uchida, 1969). Where the island is small and there are no alternative sources of food, rodent populations have been reduced; but the hazards to birds and populations of domestic animals suggest that widespread introduction of these predators should not be made across the whole Pacific area. The concept of the sterile male technique, used so successfully in some insect eradication programmes, has been suggested for use in rat control. A sterile male mutant, complete with morphological marker, whose behaviour pattern is normal is known (Stanley et al., 1968). Preliminary field trials have been star- ted, but the phenomenon of multiple matings by female rats in oestrus suggests that the field appli- cation of this technique may be difficult. Much of our knowledge about rodents has been obtained from trapping programmes. We still know too little about the way in which rats respond to traps and bait stations. Rat trappers often had their own baits or attractants, but few detailed and comprehensive analyses of the acceptability and attractiveness of baits, or of food preferences of rats, have been made (Jackson, 1965). While the house mouse is a highly inquisitive spe- cies that investigates all aspects of its environment, the Norway rat is cautious and prefers the familiar. Rats tend to move close to walls and fences or along established trails. Thus the siting of bait stations adjacent to runs is more important than adding specific attractants to the bait. Rats feed on a wide variety of food and usually select a diet that is nutri- tionally balanced, if it is possible to do so, and tend, initially at least, to select familiar foods rather than foods that would otherwise be preferred. However, very little is known about the development of food preferences. The young rat establishes many behaviour patterns in its early life between the time its ears and eyes open at about 2 weeks of age until it is weaned about 2 weeks later. Initially, fear is absent but a negative response to new objects soon develops and causes the young rat to return quickly to familiar surroundings. It is not known, however, how firmly food preferences are fixed during this period, how variable they are, or how easily they may be modi- fied. What makes a bait palatable? Frequently sugars and oils are added to cereal grains, but fish, meat, fruit, and vegetables may be used if long-term stability is not required. Many kinds of additive are used, and in a recent report Teichner et al. (1970) have provided some systematic information on this complex sub- ject. Baits with a large particle size will be carried or dragged away by the rats and sometimes hoarded, and are thus sometimes involved in secondary poisonings; crushed or ground grains reduce the opportunity for rats to remove the bait, although the author has watched rats carry away ground bait by the mouthful. A knowledge of the movement patterns of rodents is most important in control programmes, and the home ranges of rats in various environments have been determined. In sugarcane fields in Hawaii, for example, roof rats tend to range along bushy edges of the fields while Polynesian rats spread more widely into the fields (Hood, 1968). Studies now in progress in Hawaii and the Philippines using small radio transmitters implanted in rats are pro- viding detailed information on the movements of individual animals (Hood et al., 1970). Sewers provide both harbourage and routes of communication for Norway rats (Barbehenn, 1970; Greaves et al., 1968) and, in some cases, roof rats (Robe, 1966). With food readily available, the rats have no need to come to the surface, but considerable wandering and migration from breeding foci occurs and seasonal variations are evident. Because move- ments into the structure of buildings are facilitated by sewerage connexions, the effective control of these subsurface rat populations is essential in any programme of urban rodent control. Paraffin bait containing either an anticoagulant or acute poison often are hung at intervals in sewers and provide continuous control pressure. Wandering individuals may be unwilling to take baits, and anticoagulants are therefore less suited than acute poisons for controlling these rats. The maintenance of population stability, even at high densities, may be desirable in order to reduce the movements, particularly migrations, of rats. For example, rodent control practices at the present 283 W. B. JACKSON time in the Port of Saigon, a 3-km strip along the river, do not include reductional techniques (A. Fry, USAID, personal communication) since established rats tend to repel invaders and thus reduce the pro- bability of plague-infected fleas being introduced into the population. This approach, combined with DDT dusting, has been effective in preventing the occurrence of plague in this limited area. The genus Rattus is an extremely diverse group, and its ecological requirements are both variable and flexible. The Norway rat is the only species present in some parts of the world. The species is usually found in urban situations, or at least in relation to human habitation, and is not readily separated from the presence of man or from the waste products of human activity. In Hawaii, however, the species inhabits sugarcane fields while in New Zealand it is found on an uninhabited island (Beveridge & Daniel, 1967); elsewhere it has been reported to live successfully without human contact. In some areas the only representative of the genus present is the roof rat. Interspecies competition is an interesting problem that has been most closely studied in Norway and roof rats. A common means by which the differences between the two species are adjusted in cities is vertical stratification. In south-eastern USA, the Norway rat has displaced the roof rat in many agri- cultural areas (Ecke, 1954), and similar displace- ments have occurred in England (Bentley, 1964). In Hawaii, where poisoning operations in the cane- fields were apparently more effective against the Norway rat than against the roof rat and the Polynesian rat, the two latter species have expanded into the hiatus left by the Norway rat. Three species of rat are most prevalent in India- namely, R. rattus, R. norvegicus, and Bandicota bengalensis. In early surveys of Bombay and Cal- cutta made by the Plague Commission a high pre- valence of roof rats was observed. Similarly in Madras and other Indian states this species accounted for more than 90% of the urban rodent population (King & Pandit, 1931). Methods of building construc- tion using thatched or tiled roofs tend to favour this species. More recent studies, however, have indicated that the lesser bandicoot rat, a species said to be more resistant to plague or plague con- trol programmes than roof and Norway rats, is now more common (Seal & Banerjee, 1966; Deoras, 1966; Spillett, 1968). In a study made on the scrubby hillsides of Hong Kong two species of Rattus (R. huang and R. r. sladeni) occupied essentially the same ecological niches (Marshall et al., 1968). Those authors suggested that more rats might have been able to live in this environment if the two species had been active at different times. On account of the similarities of their ecological requirements, the simultaneous control of several species of rat is possible, but the early work of Calhoun (1963) indicated that when many species of small mammal living in the same area are sub- jected to trapping a strong differential trap response will be observed. Some species of mice will be trapped out almost immediately; others will not be found in the traps until the second week of trapping. Wood (1965) more recently found diffe- rential rates of response to applications of poison in populations of North American range rodents. Recent studies on Guam have indicated that a similar phenomenon arises with mixed rodent and shrew populations (Barbehenn, personal communi- cation). House mice were not readily trapped until the shrews and roof and Polynesian rats had been largely removed. The house mouse, perhaps the most abundant species of mammal in the world, second to man, is both preyed upon and held in a submissive condition by rats. The removal of rats may free mouse populations to multiply and disperse. In architectural styles where transit spaces are small enough to exclude rats, mice can move easily. Where food storage facilities are inadequate, as in some of the tall new apartment buildings in Hong Kong, mice become very serious pests. The control of mice is difficult because of their very limited home range and small requirements of food and free water. To determine the effectiveness of a control pro- gramme or the status of an infestation it is necessary to have some means of determining the size of a rodent population. The estimate of one rat per person continues to appear in the parascientific literature but is probably quite incorrect. In most cases the actual population cannot be determined, but an index or relative measure can generally be obtained. A trap or flea index, or the calibration of signs, if carefully made, can provide the information necessary for evaluating the status of rodent popu- lations and the effectiveness of control operations. Censuses of rats in Baltimore and New York City were made two decades ago by calibrating signs and then sampling representative areas of the cities 284 BEHAVIOURAL STUDIES OF RODENTS 285 (Davis, 1950; Davis & Fales, 1950). The estimated population in one case was so small that the health commissioner cancelled an impending rat control programme because there " were not enough rats ". Rats and mice, commensal with man, are very successful species, and an important aspect of their evolutionary history is that they are generalized and highly adaptable animals. These characteristics mean that simple control efforts are not likely to succeed and that effective control programmes must take into account their behavioural and physio- logical complexities. RESUME ETUDE DE LA BIOLOGIE ET DU COMPORTEMENT DES RONGEURS EN TANT QUE FONDEMENT DES MESURES DE LUTTE Les rongeurs domestiques, et en particulier ceux qui infestent les agglomerations urbaines, representent un important probleme de sante publique et une menace pour la sante de l'homme qui exigent I'application de mesures de lutte efficaces. L'existence au sein des popu- lations de rongeurs d'interactions entre les individus qui les composent explique certains aspects de leur comportement qui interessent specialement les biolo- gistes charg6s de les detruire. A l'int6rieur du genre Rattus, on trouve une grande diversite d'especes dont les caracteristiques ecologiques sont variables et adap- tables. Le genre Mus, en raison de la petite taille de ses representants, de leurs deplacements limites et de leurs faibles besoins nutritifs, est difficile a eliminer. Toute operation de lutte contre les rongeurs domes- tiques doit tenir compte de leur dynamique de croissance. Apres une periode de multiplication logarithmique, la population cesse de croitre lorsque l'habitat est com- pletement occupe et exploitd. Si les moyens utilis6s ne detruisent qu'une partie des individus, ils aboutissent a declencher a nouveau le phenommne de pullulation massive. C'est le r6sultat qu'on obtient g6neralement avec les procedes classiques: pose de pieges, emploi d'appats empoisonnes ou d'animaux predateurs. Les effets sont plus certains et plus durables si l'on modifie 1'environnement mais cette methode se r6vele d'appli- cation difficile et couteuse. L'emploi des chimiost6rilisants semble devoir donner des resultats tres favorables, mais il s'agit d'une solution d'avenir car en depit des quelques essais effectues on ne peut actuellement les utiliser a grande echelle par suite de leur manque de specificite. Ils posent aussi un autre probleme: en raison du com- portement sexuel et des caracteristiques physiologiques des rongeurs domestiques, il est indispensable que ces composes atteignent en fait la totalite de la population a detruire si l'on veut obtenir une efficacite optimale. REFERENCES Barbehenn, K. R. (1970) Notes on the ecology of sewer rats in St. Louis. In: Proceedings of the Fourth Verte- brate Pest Conference, West Sacramento, Calif., Cali- fornia Pest Committee, pp. 19-22 Bentley, E. W. (1964) J. anim. Ecol., 33, 371-373 Beveridge, A. E. & Daniel, M. J. (1967) N. Z. J. Sci., 8, 174-189 Brooks, J. E. & Bowerman, A. M. (1969) Soap. chem. Spec., 44, 58, 60, 62, 64, 82 Calhoun, J. B. (1963) The social use of space. In: Mayer, W. V. & VanGelder, R. G., ed., Physiological mammalo- gy, New York, Academic Press, pp. 1-187 Christian, J. J. et al. (1965) Recent Progr. Hormone Res., 21, 501-577 Davis, D. E. (1950) Amer. J. Hyg., 52, 147-152 Davis, D. E. (1952) Publ. Hlth Rep. (Wash.), 67, 888-893 Davis, D. E. & Fales, W. T. (1950) Amer. J. Hyg., 52, 143-146 Davis, D. E. & Read, C. P. (1958) Proc. Soc. exp. Biol. (N. Y.), 99, 269-272 Deoras, P. J. (1966) Significance ofprobable change of rat population in Bombay. In: Proceedings ofa Symposium on Indian Rodents sponsored by the Johns Hopkins University Centerfor Medical Research and Training and the United States Agency for International Develop- ment, Calcutta, pp. 58-68 Ecke, D. H. (1954) J. Mammal., 35, 521-525 Elton, C. S. (1953) Brit. J. Anim. Behav., 1, 151-154 Greaves, J. H. et al. (1968) Ann. appl. Biol., 62, 341-351 Hood, G. A. (1968) Rat control research in Hawaiian sugarcane. In: Proceedings of a Conference on Rodents as Factors in Disease and Economic Loss, Honolulu, Institute for Technical Exchange, pp. 88-94 Hood, G. A. et al. (1970) The rat in Hawaiian sugarcane. In: Proceedings of the Fourth Conference on Vertebrate Pests, West Sacramento, Calif., California Pest Com- mittee, pp. 34-37 Jackson, W. B. (1965) Pest Control, 33, 12-13, 50, 52 Jackson, W. B. (1951) J. Mammal., 32, 458-461 Jackson, W. B. (1967) Micronesica, 3, 5-15 286 W. B. JACKSON King, H. H. & Pandit, C. G. (1931) Indian J. med. Res., 19, 357-392 Marsh, R. E. & Howard, W. E. (1970) Chemosterilants as an approach to rodent control. In: Proceedings of the Fourth Vertebrate Pest Conference West Sacra- mento, Calif., California Pest Committee, pp. 55-63 Marshall, P. M. et al. (1968) Ecological implications of activity patterns of wild hillside rats in Hong Kong. In: Proceedings of a Symposium on Recent Advances in Tropical Ecology, Faridabad, India, pp. 225-236 Rohe, D. L. (1966) Calif. Vector Views, 13, 75-78 Sea], S. C. & Banerji, R. N. (1966) Changing patterns of rodent population in Calcutta and Howrah. In: Proceedings of a Symposium on Indian Rodents spon- sored by the Johns Hopkins University Center for Medical Research and Training and the United States Agency for International Development, Calcutta, pp. 69-83 Shelford, V. E. (1952) Trans. Ill. St. Acad. Sci., 45, 155-160 Spillet, J. J. (1968) The ecology of the lesser bandicoot rat in Calcutta, Calcutta, The Johns Hopkins University Center for Medical Research and Training Stanley, A. J. et al. (1968) A dominant mutant gene producing sterility and restrictions of color pattern in the male rat. In: Proceedings ofthe Third International Congress of Endocrinology, Mexico City, series 157, abstract 286 Storer, T. I. et al., ed. (1962) Bull. Bernice P. Bishop Mus., 225, 1-274 Teichner, W. H. et al. (1969) Mode of action of rodent repellents and attractants. US Army, Natick Labora- tories, Natick, Mass. (Technical Report 70-69-PR) Uchida, T. A. (1969) J. Fac. Agric. Kyushu Univ., 15, 311-329, 355-384 Wood, J. E. (1965) J. Wildl. Mgmt, 29, 425-438