Field assessment of dog as sentinel animal for plague in endemic foci of Madagascar

The epidemiology of Yersinia pestis, the causative agent of plague, involves vectors and reservoirs in its transmission cycle. The passive plague surveillance in Madagascar targets mainly rodent and fleas. However, carnivores are routinely surveyed as sentinels of local plague activity in some countries. The aim of this study is to assess the use of domestic dog (Canis familiaris) as sentinel animal for field surveillance of plague in a highly endemic area in Madagascar. Cross-sectional surveys of plague antibody prevalence in C. familiaris were conducted in endemic areas with contrasting histories of plague cases in humans, as well as a plague free area. Rodent capture was done in parallel to evaluate evidence for Y. pestis circulation in the primary reservoirs. In 2 sites, dogs were later re-sampled to examine evidence of seroconversion and antibody persistence. Biological samplings were performed between March 2008 and February 2009. Plague antibody detection was assessed using anti-F1 ELISA. Our study showed a significant difference in dog prevalence rates between plague-endemic and plague-free areas, with no seropositive dogs detected in the plague free area. No correlation was found between rodents and dog prevalence rates, with an absence of seropositive rodents in some area where plague circulation was indicated by seropositive dogs. This is consistent with high mortality rates in rodents following infection. Re-sampling dogs identified individuals seropositive on both occasions, indicating high rates of re-exposure and/or persistence of plague antibodies for at least 9 months. Seroconversion or seropositive juvenile dogs indicated recent local plague circulation. In Madagascar, dog surveillance for plague antibody could be useful to identify plague circulation in new areas or quiescent areas within endemic zones. Within active endemic areas, monitoring of dog populations for seroconversion (negative to positive) or seropositive juvenile dogs could be useful for identifying areas at greatest risk of human outbreaks.     

Rodent hosts and flea vectors in Brazilian plague foci: a review

Plague, caused by the Yersinia pestis bacterium, has several foci scattered throughout a large area from the Brazilian territory that ranges from the Northeastern State of Ceará to the Southeastern State of Minas Gerais and another separated area at the State of Rio de Janeiro. This review gathers data from plague control and surveillance programs on the occurrence and geographic distribution of rodent hosts and flea vectors in the Brazilian plague areas during the period of from 1952 to 2019. Furthermore, we discuss how the interaction between Y. pestis and some rodent host species may play a role in the disease dynamics. The absence of human cases nowadays in Brazil does not mean that it was eradicated. The dynamics of plague in Brazil and in other countries where it was introduced during the 3rd pandemic are quite alike, alternating epidemics with decades of quiescence. Hence, it remains an important epidemic disease of global concern. The existence of a large animal reservoir and competent vectors demonstrate a need for continuous surveillance to prevent new outbreaks of this disease in humans.     

From the recent lessons of the Malagasy foci towards a global understanding of the factors involved in plague reemergence

Re-emergence of human cases of plague after decades of silence does not necessarily mean that plague foci are re-emerging. Most often, Yersinia pestis bacteria have been maintained and circulating at low levels in the rodent populations. It seems therefore more appropriate to speak in terms of expansion or regression phases for sylvatic rodent plague foci and to reserve the term re-emergence for human cases. From the analysis of well-documented human plague cases in Madagascar, we underline the causes of re-emergence that can be generalized to most world foci, and can help define environments at risk where the threat of new emergence lurks. In all recent plague outbreaks, usually more than one risk factor was at the origin of the re-emergence. The reduction or discontinuance of surveillance and control, as well as poverty and insalubrity are the main factors in the re-emergence of human cases, allowing increased contacts with infected rodents and fleas. Environment changes (i.e. climatic changes, deforestation, urbanization) induce changes in flea and rodent populations by (i) extension of rodent habitats (for example by replacing forests by steppes or farmlands); (ii) modifications in population dynamics (possible outbreaks due to an increase of available food resources); but also, (iii) emergence of new vectors, reservoirs and new Y. pestis genotypes. Numerous and spontaneous genomic rearrangements occur at high frequencies in Y. pestis, which may confer selective advantages, enhancing the ability of Y. pestis to survive, to be transmitted to new hosts, and to colonize new environments. Therefore, any environmental change should be taken as a warning signal and active surveillance programs should be initiated.     

Surveys of rodent‐borne disease in Thailand with a focus on scrub typhus assessment

The epidemiology of many rodent‐borne diseases in South‐East Asia remains ill‐defined. Scrub typhus and lep‐tospirosis are common and medically significant, while other zoonotic diseases, such as spotted fever group Rickettsiae have been identified, but their overall medical significance is unknown. Rodent surveillance was conducted from June 2002 to July 2004 in 18 provinces from Thailand. Traps were set up for one to three nights. Blood and serum samples and animal tissue samples (liver, spleen, kidney and urinary bladder) were collected. Chiggermites, ticks and fleas were removed from captured rodents. A total of 4536 wild‐caught rodents from 27 species were captured over two years of animal trapping. Rattus rattus was the dominant species, followed by Rattus exulans and Bandicota indica. Almost 43 000 ectoparasites were removed from the captured animals. Approximately 98% of the ectoparasites were chigger‐mites, of which 46% belonged to the genus Leptotrombidium (scrub typhus vector). Other genera included Schoengastia and Blankaartia. Tick and flea specimens together comprised less than 1% of the sample. Among the five species of ticks collected, Haemaphysalis bandicota was the predominant species caught, followed by Ixodes granulatus other Haemaphysalis spp., Rhipicephalus spp. and Dermacentor spp. Only two species of fleas were collected and Xenopsylla cheopis (rat flea) was the predominant species. Using both commercial diagnostic kits and in‐house molecular assays, animal tissue samples were examined and screened for zoonotic diseases. Seven zoonotic diseases were detected: scrub typhus, leptospirosis, murine typhus, tick typhus, bartonella, babesiosis and trypanosomiasis. Most samples were positive for scrub typhus. Other zoonotic diseases still under investigation include borrelosis, ehrlichiosis, the plague, and other rickettsial diseases. Using geographic information systems, global positioning systems and remote sensing technology, epidemiological and environmental data were combined to assess the relative risk in different biotopes within highly endemic areas of scrub typhus in Thailand.     

Parasitic skin diseases in cats

Parasites are important causes of skin disease in cats. In this article 10 species of parasites (Table 1) will be discussed. In the United Kingdom, the flea is the most important single cause of skin disease in cats. The most common flea found is Ctenocephalides felis, with hedgehog, dog and rodent fleas being found occasionally.     

Effects of levonorgestrel‐quinestrol (EP‐1) treatment on Mongolian gerbil wild populations: a case study

Rodent pest population outbreaks occur frequently in grassland ecosystems in northern China. The Mongolian gerbil (Meriones unguiculatus) is a dominant pest rodent which is distributed across the semi‐desert grasslands of Inner Mongolia, China. In 2009, we studied the contraceptive effect of levonorgestrel‐quinestrol (EP‐1), concentration 50 ppm, on a wild Mongolian gerbil population. The one‐off contraceptive treatment was compared with a control group using a semi‐monthly live trapping method in the Ordos Semi‐desert Grassland Region of Inner Mongolia. The results show that juveniles were not recruited in spring in the treatment group. Ratios of juveniles in the control and treatment groups showed significant semi‐monthly differences from spring to summer (one‐way ANOVA, F_{2, 14}= 7.53, P < 0.05). Between both groups, annual fluctuations of juvenile and total population densities were significantly different respectively (F_{2, 14}= 4.64, P < 0.05; F_{2, 18}= 7.72, P < 0.05). The contraceptive EP‐1 delayed the normal reproductive pattern of Mongolian gerbil populations. This suppressed birth rates of gerbil populations, reduced their densities, and changed their age structures. The period of EP‐1 baiting should be extended but it could be an ideal method for controlling Mongolian gerbil populations during each breeding season.     

Fleas parasitizing domestic dogs in Spain

In addition to their importance to veterinary clinical practice as ectoparasites, fleas of domestic dogs are of special concern because they can be vectors of disease, including zoonoses. Flea assemblages parasitizing domestic dogs usually comprise several flea species whose distribution is determined by factors acting at several scales. Knowledge of these factors will aid in assessment of the distribution patterns of flea parasitism, and is an important tool in developing control strategies and in evaluation of flea-borne disease risk in dogs and humans. In this survey we used data from 744 domestic dogs from 79 localities in Spain to explore the associations between the abundance of flea species, host-dependent factors (sex and age), and host habitat factors including abode (farm, house with garden, apartment), location (urban or rural), the presence of other pets, and dog activity (measured as the frequency with which dogs left their abode). We also considered environmental factors including the time of year and mean annual temperature and rainfall. Variations in flea community structure at infracommunity and component community levels were also explored. Four flea species were found parasitizing dogs. Ctenocephalides felis was the most abundant (88.02% of fleas identified), followed by Ctenocephalides canis (10.38%), Pulex irritans (1.47%) and Echidnophaga gallinacea (0.13%). Overall flea abundance was higher on dogs living on farms than in apartments, as was the abundance of Ct. felis, Ct. canis and P. irritans. Ct. felis was more abundant on dogs living in houses than in apartments, but the reverse was found for P. irritans. Overall flea abundance and Ct. canis abundance were highest in rural areas, whereas the presence of other pets sharing the abode was associated with higher overall flea abundance and Ct. felis abundance. Only P. irritans abundance was positively related to the activity of dogs. Ct. canis and P. irritans abundances were higher during the warm period of the year. Mean annual temperature was negatively correlated with overall, Ct. canis and P. irritans abundances, but positively related to Ct. felis abundance. Annual rainfall was negatively correlated with Ct. canis and P. irritans abundances. Variations in the number of flea species found on a dog reflected the abundance distribution patterns for each species and their associations with host habitat and environmental factors. At the component community level, flea species richness was inversely related to annual mean temperature. The structure of flea assemblages on dogs was mainly associated with host habitat and environmental variables, and not with host-dependent variables. However, a large amount of variation in flea abundance remained unexplained, suggesting the effect of other non-controlled factors.     

Zoonotic Helminths of Urban Brown Rats (Rattus norvegicus) in the UK: Neglected Public Health Considerations?

Urban brown rats (Rattus norvegicus) carry microbial human pathogens but their role as reservoir hosts for helminths of public health importance is less well known. In this study, 42 brown rats trapped on Merseyside were subject to thorough combined helminthological and pathohistological post‐mortem examination. Eggs of the rodent‐borne zoonotic nematode Calodium hepaticum were initially detected in histological sections of the livers of 9.5% of rats, but overall diagnostic sensitivity increased to 16.6% when entire liver tissue was disrupted and the resulting filtrates were examined for released eggs. In their rat host, mainly trapped inside the dockland, infections with C. hepaticum were associated with a chronic multifocal pyogranulomatous hepatitis with intralesional eggs and peripheral fibrosis. Mean intensity of hepatic C. hepaticum egg infections was 1041 eggs. This is the first report of C. hepaticum in an urban brown rat population in the UK and provides original data for liver egg burdens in this abundant commensal rodent. The zoonotic cestode Rodentolepis nana had a prevalence of infection of 14.3%. Rodent‐specific, non‐zoonotic helminths found were the spiruroid Mastophorus muris (16.0%) in the stomach, the trichuroid Trichosomoides crassicauda in the urinary bladder (31.0%); the ascarid Heterakis spumosa was the commonest helminth of the large intestine (76.2%). Many millions of brown rats inhabit cities and rural areas of the UK, and the infective stages of the zoonotic worm species, particularly C. hepaticum, are likely to be widely distributed in the environment presenting a threat to public health.     

A survey of rats trapped in the wellington area for ectoparasites and organisms of the salmonella,pasteurella, and leptospiral groups

Extract

It has long been established that rats can play an important part in the transference of infectious diseases to man. With this fact in mind, the late Dr J. H. Blakelock, former Director of the National Health Institute, Wellington, began a survey of the rat population in the Wellington district and of rats brought to the area by overseas ships. This work has been continued, the trapped rats being examined for the presence of Xenopsylla cheopis, a flea known to be the chief vector of plague, for Pasteurella pestis, the causative organism of plague, for organisms of the Salmonella group, and for evidence of leptospiral infection.     

Habitat use and demography of Mus musculus in a rural landscape of Argentina

The main goal of the paper was to determine the habitat distribution of the house mouse (Mus musculus) within a rural landscape of Buenos Aires province, Argentina. We also studied the seasonal variation in abundance and reproductive activity. The habitats studied were poultry farms, human houses in a small village, cropfields, pastures, cropfield and pasture edges, riparian habitats (streams), railway embankments and woodlots. We captured 817 M. musculus and 690 individuals of 5 native rodent species. M. musculus was captured in poultry farms, houses, riparian habitats, cropfield and borders, but it showed a significantly higher abundance in poultry farms compared to the other habitats. Its presence outside poultry farms was significantly related to the distance to streams and poultry farms. The mean trapping success index of M. musculus did not show significant variations between periods, but the proportion of active males was significantly higher in the spring–summer period than in the autumn–winter period. All captures of M. musculus in cropfields, borders and riparian habitats occurred in the spring–summer period. The capture of M. musculus in many types of habitats suggests that it can disperse outside poultry farms, and streams may be used as corridors.     

Detection of Bartonella spp. in wild rodents in Israel using HRM real-time PCR

The prevalence of Bartonella spp. in wild rodents was studied in 19 geographical locations in Israel. One hundred and twelve rodents belonging to five species (Mus musculus, Rattus rattus, Microtus socialis, Acomys cahirinus and Apodemus sylvaticus) were included in the survey. In addition, 156 ectoparasites were collected from the rodents. Spleen sample from each rodent and the ectoparasites were examined for the presence of Bartonella DNA using high resolution melt (HRM) real-time PCR. The method was designed for the simultaneous detection and differentiation of eight Bartonella spp. according to the nucleotide variation in each of two gene fragments (rpoB and gltA) and the 16S–23S intergenic spacer (ITS) locus, using the same PCR protocol which allowed the simultaneous amplification of the three different loci. Bartonella DNA was detected in spleen samples of 19 out of 79 (24%) black rats (R. rattus) and in 1 of 4 (25%) Cairo spiny mice (A. cahirinus). In addition, 15 of 34 (44%) flea pools harbored Bartonella DNA. Only rat flea (Xenopsyla cheopis) pools collected from black rats (R. rattus) were positive for Bartonella DNA. The Bartonella sp. detected in spleen samples from black rats (R. rattus) was closely related to both B. tribocorum and B. elizabethae. The species detected in the Cairo spiny mouse (A. cahirinus) spleen sample was closely related to the zoonotic pathogen, B. elizabethae. These results indicate that Bartonella species are highly prevalent in suburban rodent populations and their ectoparasites in Israel. Further investigation of the prevalence and zoonotic potential of the Bartonella species detected in the black rats and the Cairo spiny mouse is warranted.     

No evidence for enzootic plague within black-tailed prairie dog (Cynomys ludovicianus) populations

Yersinia pestis, causative agent of plague, occurs throughout the western United States in rodent populations and periodically causes epizootics in susceptible species, including black-tailed prairie dogs (Cynomys ludovicianus). How Y. pestis persists long-term in the environment between these epizootics is poorly understood but multiple mechanisms have been proposed, including, among others, a separate enzootic transmission cycle that maintains Y. pestis without involvement of epizootic hosts and persistence of Y. pestis within epizootic host populations without causing high mortality within those populations. We live-trapped and collected fleas from black-tailed prairie dogs and other mammal species from sites with and without black-tailed prairie dogs in 2004 and 2005 and tested all fleas for presence of Y. pestis. Y. pestis was not detected in 2126 fleas collected in 2004 but was detected in 294 fleas collected from multiple sites in 2005, before and during a widespread epizootic that drastically reduced black-tailed prairie dog populations in the affected colonies. Temporal and spatial patterns of Y. pestis occurrence in fleas and genotyping of Y. pestis present in some infected fleas suggest Y. pestis was introduced multiple times from sources outside the study area and once introduced, was dispersed between several sites. We conclude Y. pestis likely was not present in these black-tailed prairie dog colonies prior to epizootic activity in these colonies. Although we did not identify likely enzootic hosts, we found evidence that deer mice (Peromyscus maniculatus) may serve as bridging hosts for Y. pestis between unknown enzootic hosts and black-tailed prairie dogs.     

Habitat evaluation for outbreak of Yangtze voles (Microtus fortis) and management implications

Rodent pests severely damage agricultural crops. Outbreak risk models of rodent pests often do not include sufficient information regarding geographic variation. Habitat plays an important role in rodent–pest outbreak risk, and more information about the relationship between habitat and crop protection is urgently needed. The goal of the present study was to provide an outbreak risk map for the Dongting Lake region and to understand the relationship between rodent–pest outbreak variation and habitat distribution. The main rodent pests in the Dongting Lake region are Yangtze voles (Microtus fortis). These pests cause massive damage in outbreak years, most notably in 2007. Habitat evaluation and ecological details were obtained by analyzing the correlation between habitat suitability and outbreak risk, as indicated by population density and historical events. For the source‐sink population, 96.18% of Yangtze vole disaster regions were covered by a 10‐km buffer zone of suitable habitat in 2007. Historical outbreak frequency and peak population density were significantly correlated with the proportion of land covered by suitable habitat (r = 0.68, P = 0.04 and r = 0.76, P = 0.03, respectively). The Yangtze vole population tends to migrate approximately 10 km in outbreak years. Here, we propose a practical method for habitat evaluation that can be used to create integrated pest management plans for rodent pests when combined with basic information on the biology, ecology and behavior of the target species.     

Histopathology of experimental plague in cats

Formalin-fixed paraffin-embedded archival tissues of seven adult cats of both sexes that died after being experimentally infected with Yersinia pestis were examined light microscopically to characterize the lesions. The cats were exposed in two groups using two routes of infection: ingestion of Y. pestis-infected rodent or a subcutaneous injection of Y. pestis to simulate a flea bite. Immunohistochemistry was performed on tissues from all organ systems from a representative cat from each group to determine the distribution of Y. pestis bacilli during infection. In all seven cats, bubonic plague lesions were seen. The lesions of pneumonic plague were present in two cats. Septicemic plague was confirmed in all seven cats by bacteriologic culture. Aggregations of bacteria were seen in lymphoid tissue in all cats and in lung tissues from the two cats with pneumonic plague. The most consistent histologic finding was necrosuppurative inflammation in the lymph nodes. Invariably, Y. pestis bacteria were present in large numbers at affected sites. Orally infected cats had more numerous lesions in the lymph nodes of the head and neck regions. These experimentally induced cases of feline plague document that cats are unique among carnivores in exhibiting bubonic, pneumonic, and septicemic plague following exposure to Y. pestis. The lesions of the orally infected cats were consistent with those previously described for naturally occurring Y. pestis infections in cats and corroborate the contention that cats most commonly contract plague by eating Y. pestis-infected rodents and not via flea bite. The histopathology of Y. pestis disease in these cats is comparable to that described for human plague.     

From trees to fleas: masting indirectly affects flea abundance on a rodent host

Mast seeding causes strong fluctuations in populations of forest animals. Thus, this phenomenon can be used as a natural experiment to examine how variation in host abundance affects parasite loads. We investigated fleas infesting yellow-necked mice in beech forest after 2 mast and 2 non-mast years. We tested 2 mutually exclusive scenarios: (1) as predicted by classical models of density-dependent transmission, an increase in host density will cause an increase in ectoparasite abundance (defined as the number of parasites per host), versus (2) an increase in host density will cause a decline in flea abundance (“dilution,” which is thought to occur when parasite population growth is slower than that of the host). In addition, we assessed whether masting alters the relationship between host traits (sex and body mass) and flea abundance. We found a hump-shaped relationship between host and flea abundance. Thus, the most basic predictions are too simple to describe ectoparasite dynamics in this system. In addition, masting modified seasonal dynamics of flea abundance, but did not affect the relationship between host traits and flea abundance (individuals with the highest body mass hosted the most fleas; after controlling for body mass, parasite abundance did not vary between sexes). Our results demonstrate that pulses of tree reproduction can indirectly, through changes in host densities, drive patterns of ectoparasite infestation.     

Evaluation of habitat requirements of small rodents and effectiveness of an ecologically‐based management in a hantavirus‐endemic natural protected area in Argentina

Hantavirus pulmonary syndrome is a severe cardio pulmonary disease transmitted to humans by sylvan rodents found in natural and rural environments. Disease transmission is closely linked to the ecology of animal reservoirs and abiotic factors such as habitat characteristics, season or climatic conditions. The main goals of this research were: to determine the biotic and abiotic factors affecting richness and abundance of rodent species at different spatial scales, to evaluate different methodologies for studying population of small rodents, and to describe and analyze an ecologically‐based rodent management experience in a highly touristic area. A 4‐year study of small rodent ecology was conducted between April 2007 and August 2011 in the most relevant habitats of El Palmar National Park, Argentina. Management involved a wide range of control and prevention measures, including poisoning, culling and habitat modification. A total of 172 individuals of 5 species were captured with a trapping effort of 13 860 traps‐nights (1.24 individuals/100 traps‐nights). Five rodent species were captured, including 2 hantavirus‐host species, Oligoryzomys nigripes and Akodon azarae. Oligoryzomys nigripes, host of a hantavirus that is pathogenic in humans, was the most abundant species and the only one found in all the studied habitats. Our results are inconsistent with the dilution effect hypothesis. The present study demonstrates that sylvan rodent species, including the hantavirus‐host species, have distinct local habitat selection and temporal variation patterns in abundance, which may influence the risk of human exposure to hantavirus and may have practical implications for disease transmission as well as for reservoir management.     

Flea-borne pathogens in the cat flea Ctenocephalides felis and their association with mtDNA diversity of the flea host

Flea-borne pathogens were screened from 100 individual cat fleas using a PCR approach, of which 38 % were infected with at least one bacterium. Overall, 28 % of the flea samples were positive for Bartonella as inferred from ITS DNA region. Of these, 25 % (7/28) were identified as Bartonella clarridgeiae, 42.9 % (12/28) as Bartonella henselae consisted of two different strains, and 32.1 % (9/28) as Bartonella koehlerae, which was detected for the first time in Malaysia. Sequencing of gltA amplicons detected Rickettsia DNA in 14 % of cat flea samples, all of them identified as Rickettsia asembonensis (100 %). None of the flea samples were positive for Mycoplasma DNA in 16S rRNA gene detection. Four fleas were co-infected with Bartonella and Rickettsia DNAs. Statistical analyses reveal no significant association between bacterial infection and mtDNA diversity of the cat flea. Nevertheless, in all types of pathogen infections, infected populations demonstrated lower nucleotide and haplotype diversities compared to uninfected populations. Moreover, lower haplotype numbers were observed in infected populations.     

Rodent abundance triggered switch between the relative mutualism and predation in a rodent–seed system of the subtropical island forest

Density-dependent non-monotonic species interactions are important in maintaining ecosystem stability and function, but empirical evidences are still rare. Rodents, as both seed dispersers and seed predators, have dual effects on plant regeneration and may result in non-monotonic rodent-plant interactions. According to the non-monotonic models, the relative positive or negative effects of rodents on seedling establishment can be measured based on the positive or negative association of seedling recruitment rate and rodent abundance. In this study, we investigated the fates of acorns of Quercus serrata by tracking tagged seeds on 21 fragmented subtropical islands in the Thousand Island Lake, China. We found that the proportion of germinated seeds of all released seeds showed a dome-shaped association with rodent abundance per seed. The proportion of removed seeds and cached seeds showed a saturated- and a weak dome-shaped association with rodent abundance per seed, respectively. Our results demonstrated a clear empirical evidence that rodent abundance per seed triggered a switch between the relative mutualism and predation in a rodent–seed system. Our study implied that the observed non-monotonic interactions between plants and animals may play a significant role in maintaining biodiversity and ecosystem function. We appeal for more investigations of the complex non-monotonic interactions in various ecosystems.     

Small mammals and habitat structure along altitudinal gradients in the southern Cape mountains

Small mammals were sampled along altitudinal gradients in two mountain localities, the Swartberg and Baviaanskloof, in the southern Cape in summer 1977–1978. Species composition varied according to altitude and aspect with a total catch of six rodent and two shrew species. Correlations were found between abundance of rodent species and habitat variables, particularly vegetation structure and the cover of rock and bare soil. Rodent species diversity was negatively correlated with the proportion of total foliage at low to mid heights (40 - 80 cm). Factors controlling apparent habitat preferences are discussed together with modes of rodent niche separation. Extrapolation of the results is limited by the single season and restricted geographical coverage of the survey.     

A global evaluation of the associations between long-term dynamics of seed falls and rodents

One classic system of pulsed resource and animal population is mast seeding and population dynamics of seed-eating rodents in forests. However, we still lack an understanding of the global patterns regarding the contributions of seed falls to rodent outbreaks or population dynamics. We analyzed a global dataset of coupled long-term time series of seed abundances and rodent populations from published literature, including 66 and 89 time series (156 rodent–seed pairs from 37 studies) for rodent and seed abundances, respectively. We found only half of the examined rodent populations showed statistically significant coincidence between rodent outbreak and mast-seeding years. Over all the coupled time series, seed abundance was found to positively correlate with rodent abundance with a one-year lag, and the relative importance of seed abundance was much lower than that of density dependence in affecting rodent population growth rates. We also found the relative importance of seed abundance decreased, but that of rodent density dependence increased with the latitude of study. For the first time, our work provides a global pattern on the associations between seed falls and rodent population dynamics mostly in mid- and high-latitude forests, and highlights the necessity of more long-term studies on this subject in more forest ecosystems.