Interciencia ISSN: 0378-1844 [email protected] Asociación Interciencia Venezuela

Eiris, Gabriela C.; Barreto, Guillermo R. HOME RANGE OF MARSH RATS, sciureus, A PEST IN RICE FIELDS OF VENEZUELA Interciencia, vol. 34, núm. 6, junio, 2009, pp. 400-405 Asociación Interciencia Caracas, Venezuela

Available in: http://www.redalyc.org/articulo.oa?id=33911405005

How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative HOME RANGE OF MARSH RATS, Holochilus sciureus, A RODENT PEST IN RICE FIELDS OF VENEZUELA

Gabriela C. Eiris and Guillermo R. Barreto

SUMMARY

Radio tracking techniques were used to determine the home tours, fixed kernel and adaptive kernel. Home ranges of males range of marsh rats Holochilus sciureus in rice fields of the and females ranged between 0.6 and 1.3ha depending on the Guárico River Irrigation System, Venezuela, in order to study method of estimation. Male-female range overlap and possible movement patterns of tropical and provide information agonistic interactions between males, suggest male territoriality for the management of this agricultural pest. After being cap- and a polygynic mating system. Males showed greater motility tured using Tomahawk traps, ten adult rats were equipped with than females. The results also reveal relationships between these radio-transmitters and tracked during the 2004 dry season be- rodent movements and environmental changes produced within tween 18:00 and 6:00. Home ranges were estimated based on rice plantations. Ecological studies provide information for an four methods: minimum convex polygon, harmonic mean con- integrated pest control program.

ome range is an aspect of 1996; Kenward, 2001; Brown et al., 2003) crops (Agüero, 1979; Cartes, 1979; Agüero ecology that provides an- and this is particularly so in the case of spe- and Castillo, 1981; Cartaya, 1983; Agüero et swers to many biological cies considered as pests in agro-ecosystems. al., 1985; Cartaya and Aguilera, 1985; questions related to population dynamics, so- Rodents have been recognized as important García-Rangel, 2002; Barreto and García- cial interactions, and spacing patterns. Home pests in both agricultural fields and harvest Rangel, 2005). Its average weight is ~130g, range is defined as the area traversed by the storages. Rodent species have also been the but males over 300g have been captured by individual in its normal activities of food focus of many studies of space use and home the authors and by Agüero (1979). This rat gathering, mating and caring for young (Burt, range, mostly in temperate regions. A few feeds mainly on grass stalks, leaving the til- 1943). It varies in size, shape and structure studies, however, have been conducted in lers in mounds on the ground and causing according to the characteristics of the indi- tropical regions (e.g. Endries and Adler, serious damage to the crops mentioned, and vidual, the population and/or the environment 2005) and data is particularly scarce in Latin despair among the farmers (Twigg, 1962, (McNab, 1963; Tew et al., 2000; Saiful et America, where many agro-ecosystems suf- 1965; Martino and Aguilera, 1989). al., 2001; Priotto et al., 2002). The concept fer the outbreaks of rodent populations and Current management of is not synonymous to territory, which refers the concomitant economical consequences. this species in rice fields consists mainly on to an individual’s more or less exclusive area The marsh rat, Holochilus the use of second generation rodenticides and that is defended against the presence of con- sciureus Wagner 1842, is the main rodent bounty schemes. These are applied by the specifics (Murie and Harris, 1978). pest in rice and sugar cane fields in Venezu- growers themselves, with very little planning Spacing patterns are rec- ela and other Neotropical countries. It be- and poor safety precautions. The rodenticides ognized as an important factor for population longs to the Muridae family and is a noctur- are distributed along the banks of the rice regulation (Ostfeld, 1990). Moreover, study- nal semi-aquatic species, possessing mem- tanks, in indiscriminate amounts and with ing home range can prove necessary branes in its hind feet and building its nests questionable results. To make things worse, for the design of management strategies above the water level in flooded fields, mak- in outbreak years, growers become desperate (Harris et al., 1990; Seaman and Powell, ing it a very successful rodent in inundated and choose to use organo-phosphide insecti-

KEYWORDS / Holochilus sp. / Home Range / Marsh Rat / Pest Control / Rodent Pest / Received: 08/04/2008. Modified: 06/08/2009. Accepted: 06/09/2009.

Gabriela C. Eiris. Biologist, Universidad Simón Bolívar (USB), Venezuela. Graduate student, University of Sydney, Australia. e-mail: [email protected] Guillermo R. Barreto. Biologist, Universidad Central de Venezuela. M.Sc. in Biological Sci- ences, USB, Venezuela. D. phil, Oxford University, UK. Professor, USB, Venezuela. Address: Departamento de Biología de Organis- mos. Universidad Simón Bolívar. Caracas 1080-A, Venezuela. e-mail: [email protected]

400 0378-1844/09/06/400-06 $ 3.00/0 JUN 2009, VOL. 34 Nº 6 cides as a “more effective” alternative. The banks surrounding tanks over a 1-2 night pe- by recalculating the area of the minimum main reason for this ill-planned management riod per trapping session between Oct 2003 convex polygon (100% of the fixes) as suc- of H. sciureus is that very little research has and Feb 2004, for a total of 628 trap nights. cessive locations were added. been done on its ecology, with the end result The traps were set and baited using pumpkin that virtually no novel management strategies (Cucurbita maxima) pieces between 16:00 Data analysis have been designed, as it is the case with and 19:00, and were checked and closed be- most tropical rodent pests in South America. tween 6:00 and 10:00 the next day. Rodents Four methods of home Thus, there is the need to study the ecology captured were anesthetized using Halothane range estimation were used: minimum con- of this species with the purpose of offering (Fluothane®), weighed with a dynamometer vex polygons (MCP), harmonic mean con- alternatives regarding its control that are (Pesola AG, Switzerland), measured, and tours (HMC), fixed kernel (FK) and adaptive more cost effective and more environmental- identified to the species level. Those identi- kernel (AK). The percentage of fixes (for ly friendly than those used today. fied as H. sciureus were further classified as MCP), or the utilization distribution in the The home range of H. sci- adults or juveniles according to their weight probabilistic methods (HMC, FK and AK) to ureus in rice fields in Venezuela was previ- (Agüero, 1979). On occasions when trapping be used as estimates of home range were es- ously estimated (±SE) at 0.278 ±0.403ha by efficiency was very low, manual trapping was tablished examining the discontinuities in uti- Cartaya and Aguilera (1984) employing performed taking advantage of the habitat lization plots, which calculate the area of mark-recapture techniques in a 1ha grid and disturbance caused by the plough. Large H. home range cores of different percentages. using minimum convex polygons. They con- sciureus specimens that escaped the plough Home range estimation using the MCP em- sidered this result as an underestimate of the were caught using a net. The rats caught by ployed the established percentage of the loca- real value due to the low capture frequency this method were checked for injuries and tion fixes closest to the center of each ani- and the fact that some individuals were re- then released in a site different from that of mal’s range. We chose the range center as captured 2-3km away from the grid, 1-2 capture because of the lack of vegetation the harmonic mean of the spatial coordinate months after the previous capture. For that cover left by the plough. components to guarantee that it would lay reason, the use of an alternative sampling ap- within an area with high density of location proach, such as radio-tracking, became nec- Tracking program fixes. HMC estimation was performed using essary as well as the use of methods of range a 150×150 matrix without location centering. estimation other than minimum convex poly- A total of 10 adult marsh FK and AK estimations were performed us- gons which have been already proved to be rats, six males and four females, were ing a 40×40 matrix and the smoothing pa- biased (Burgman and Fox, 2003). Therefore, equipped with radiocollars SOM 2380 rameter h was chosen as the median of the this study aims to determine the home range (weight ~10g) or SOM 2070 (weight ~3g) values for each range calculated by least of H. sciureus in rice fields as revealed by from Wildlife Materials, Inc., USA, accord- squares cross validation. To facilitate com- radio-tracking techniques, with the intention ing to each individual’s weight, verifying parisons with other studies, we calculated of improving previous estimations and pro- that the transmitters did not exceed 5% of home range size for each rat using the 100% viding useful information for the develop- the animal’s body mass, percentage that MCP method, the most commonly reported ment of novel management strategies. does not significantly increase their energy method. Areas of intensive use were termed expenditure, following Berteaux et al. core areas and were calculated as 40% of the Study Area (1996). Transmitters for each individual utilization distribution. The ratio of core area were tuned to a different frequency in the to home range (the set percentage for each The study was conducted 219MHz range. Once each rat recovered method) sizes was calculated to determine in rice plantations (plots 152 and 173) located from anesthesia, it was released at the site the relative sizes of core areas within home within the Guárico River Irrigation System of capture, except in cases of manual cap- ranges. Home range and core area contours (SRRG; 8°50’43”N, 67°32’34”W), near the ture, where the animal’s habitat had been were plotted on a map of the area. city of Calabozo, Guárico State, Venezuela. destroyed and they were released along the The median of home range The plots are divided in strips parallel to the nearest canal. All procedures were per- values was used as a measure of central inclination of the terrain, which are in turn formed following guidelines for humane trend because data were not normally distrib- subdivided in rectangles ~3.5ha by earth treatment of . uted. Differences between sexes in home banks, canals and roads creating ‘tanks’ that Radio tracking began as range size, core area and core area to home can be flooded and drained in order to man- the radiocollars were fitted to the rodents range ratios were examined using Mann- age the water level in the field. The climate and was continued until April 2004. In this Whitney tests. Methods of estimation of is seasonal, with a rainy season lasting from way, an entire rice cycle was covered, par- home range size were compared using Fried- May to October and a dry season from De- ticularly the dry season one, which is the man’s ANOVA and were further examined cember to March. April and November are one that suffers the most intense attack by employing Wilcoxon’s pair wise comparisons. transitional months. Mean annual rainfall is rodents. Fixes were obtained by triangula- Independence of fixes was examined by cal- 1328mm. The dominant vegetation consists tion every 30min between 18:00 and 6:00 culating Schoener’s t2/r2 index (Swihart and of rice (Oryza sativa) in different stages of each night using a VHF receptor, a three- Slade, 1985) as implemented by Ranges 6. development. Most abundant rodent species element portable yagi antenna, a compass The distances traversed by individuals in present in the area are Holochilus sciureus, and a GPS for a total of 40 track nights in 30min were calculated and compared be- Sigmodon alstoni, brevicauda six months. One to three animals were tween sexes using a Mann-Whitney test and and sp. (García-Rangel, 2002). tracked each night. UTM coordinates were between individuals using Kruskall-Wallis used throughout the study. tests. Individual differences were further ana- Materials and Methods Incremental area analysis lyzed using Tukey’s multiple contrasts test. and home range estimation were performed Furthermore, maximum distances traversed Trapping program using the software Ranges 6 (Kenward et al., in 30min were identified, as well as maxi- 2003). Incremental area analyses were per- mum distances reached from the point of re- Tomahawk ® live traps (79 formed for each animal in order to establish lease and from the center of the home range ±31 units) were set 10 steps apart along earth if sufficient location fixes had been obtained for each rat.

JUN 2009, VOL. 34 Nº 6 401 data. The range for M3 did not table I stabilize in spite of having enough Median home range (95%) and core (50) location fixes. area (40%) sizes (in ha) of H. sciureus The median home as well as median core area range size using the 100% MCP to home range indexes for each method was 3.90ha. Table I shows method of estimation the median home range (95%) 95% 40% 40% / 95% and core area (40%) sizes ob- tained in this study for H. sci- MCP 0.635 0.166 0.261 ureus, as well as median core HMC 1.288 0.116 0.156 Figure 1. Incremental area analysis of individual area to home range indexes for FK 0.696 0.115 0.283 M12. each method of estimation. No AK 1.098 0.089 0.190 significant differences were de- MCP: minimum convex polygon, HMC: harmonic mean con- Results tected in these variables between tour, FK: fixed kernel, AK: adaptive kernel. The median home males and females (lowest value range size using the 100% MCP method was 3.90ha. A total of 10 animals, six of P obtained was Mann-Whitney males (referred to as M3, M5, M8, M9, M12 Z= -0.71, P=0.48), regardless of (95% MCP of 0.64ha). The next contours, and M14) and four females (F1, F2, F6 and the estimation method used. 95% and 40% MCP, show a smaller size F7), were radio-tracked. Between 7 and 375 The values estimated for difference (Table I). location fixes were obtained per rat. Males home range size vary significantly according Contours corresponding to weighed 224-276g and females 114-212g. to the method employed in obtaining them the HMC, FK and AK methods show more 2 Only animals M3 and M9 were released at (χ = 17.4, d.f.= 3, P=0.00). Further compari- rounded edges than those created by the locations different from the site of capture. In sons showed that estimations obtained by MCP method (Figure 2). This becomes most the field, it was perceived that individuals AK and MCP methods did not differ signifi- pronounced in the case of the HMC method, M14 and F7 occupied two home ranges, each cantly (P=0.5). Differences were detected in where the home ranges in the present study separated in time. all other comparisons (P<0.03), showing that are oval-shaped. The contours generated by Incremental area analyses estimations obtained by FK were significant- the FK method are less uniform than both for the home ranges of individuals M5, M9, ly smaller than those of the other probabilis- the shapes obtained by the HMC and AK M12, M14, F1, F6 and F7 reach asymptotes tic methods whilst those obtained by HMC methods. Furthermore, there is more varia- indicating sampling saturation (Figure 1). were the largest. tion between the different percentage con- The second perceived ranges of animals M14 Figures 2 and 3 show the contours tours presented for the AK method than for and F7 were examined separately and stabili- generated by each method for individuals the FK method. zation was observed, confirming that each of M12 and F7 (first home range). It is worth Home range overlap was these rats had two temporarily separated ar- noting that the area enclosed by the 100% detected between individuals F6 and M14 eas. As was expected, ranges for M8 and F2 MCP method (median= 3.89ha) is much (first home range), and M14 (second home did not reach an asymptote due to lack of larger than the size of the next contour range) and F7 (first home range). M14’s first

Figure 2. Home range and core area contours for M12 according to minimum Figure 3. Home range and core area contours for F7 according to minimum con- convex polygon (MCP), harmonic mean contoun (HMC), fixed kernel (FK) and vex polygon (MCP), harmonic mean contourn (HMC), fixed kernel (FK) and adaptive kernel (AK) methods of estimation. adaptive kernel (AK) methods of estimation.

402 JUN 2009, VOL. 34 Nº 6 home range contains be- Discussion Estimations obtained us- tween 19.7 and 75.4% of ing the AK method, were 1.6 times larger F6’s area, depending on Home range than those of the FK method, mainly be- the method of estimation. estimation methods cause the former allows the smoothing pa- However, overlap of F6’s rameter h to vary within the reference grid. home range on M14’s is The 100% MCP As a result, areas with fewer fixes are as- only 3.6 to 18%. F6’s method estimation ob- signed a larger h than areas of higher den- home range center is lo- tained in this study is sity, causing greater smoothing of these ar- calized inside M14’s home 14 times the figure ob- eas and increasing the importance of the range according to three tained previously by edges of the distribution (Worton, 1989). of the estimation methods, Cartaya and Aguilera By applying the AK method of estimation while F6’s core area is at (1984) for H. sciureus to sets of real and simulated data, Worton least partially contained in (0.278ha) in rice fields (1995) and Seaman and Powell (1996) de- M14’s home range in ev- in the nearby Portu- termined that it produces overestimations ery instance. M14’s second guesa State. The dif- of the home range size, and concluded that home range overlaps 7.2% ference is due to the the FK approach offers a better description on F7’s range, while F7’s fact that these authors of the utilization distribution. Again, the contains 6% of M14’s sec- employed the capture- range of M12 serves as a good example, ond range, when analyzed mark-recapture method since what was observed in the field corre- using the MCP method Figure 4. M3’s route upon M12’s home for obtaining the indi- sponds more accurately to the estimation only. range. vidual locations within obtained by the FK method. Mean distance tra- a 1ha square grid. This versed in 30min by the individuals studied resulted in a small number of fixes per ani- Interactions: territoriality was 37.8 ±52.2m. No significant differences mal (<6) while in the present study calcula- were detected between males and females in tions were performed employing a larger Male and female home this variable (Mann-Whitney U-test z= -0.99, number of fixes (>30). Furthermore, the area ranges showed similar sizes. However, due P=0.32). Mean distance reached from site of studied by Cartaya and Aguilera (1984) was to the apparent separation in classes of release was 469.6 ±375.5m for all individu- limited by the size of the trapping grid. The males (see below), the similarity is proba- als. The mean value of this measurement differences observed show the limitations of bly limited to females and “resident” was 588.5 ±431.8m for males and 271.5 the capture-mark-recapture technique with re- males. ±162.9m for females. The greatest distance spect to radio-tracking techniques for home M3’s encounters with in- traversed from the site of release was range estimations. dividuals M12 and M9 and its subsequent 1153.6m and corresponded to M3 in four Contours generated using travel across more than 100m in 30min nights. This male was released along the the MCP method resulted in similar home suggests events of territory defense by resi- main drainage canal on plot 173 after being range and core area estimations as those of dent rats (M12 and M9) where the intruder caught manually. It started moving along the the probabilistic methods employed. In par- was expelled from their home ranges. In canal and entered M12’s home range core. At ticular, they did not differ significantly both instances, M3 was the smaller rat. this time the location of both rats was the from the AK method estimations. Using Therefore, intrasexual territoriality seems same and subsequently M3 traversed 140m the harmonic instead of the arithmetic to exist among males of H. sciureus, ap- in the next 30min. M3 then went across an mean for determining the home range cen- parently causing males to be separated in adjacent tank, returned to the main drainage ter ensured that its location fell within an two socially distinct classes according to canal and met again with M12. M3 then area with high density of fixes and not in their size: “residents” and “transients”. started moving northeast and came across a place seldom visited by the animal, sug- On the other hand, H. M9 in its home range core; it then traversed gested by Harris et al. (1990) and Kenward sciureus appears to show no intersexual 189m in the next 30min. After that, M3 et al. (2003). territoriality as evidenced by the home moved onto the next plot, reaching its great- Results obtained employ- range overlap between individuals M14 and est distance from the site of release. Finally, ing the HMC method appear to be overes- F6, and their concurrences in space and this animal signal was lost (Figure 4). This timations of the home ranges of the H. sci- time. Overlap was also observed in the animal had not established a distinct home ureus individuals studied when using other field between individuals F6 and F1, sug- range. methods. Furthermore, it is very unlikely gesting no female intrasexual territoriality. M9, a male that was relo- that the home ranges of animals will show Furthermore, workers in the rice fields cated after being manually captured, moved oval shapes like those seen in Figures 2 claim to have found occupied nests only along the main drainage canal and was then and 3. In particular, it was anticipated that 10m apart. repeatedly located in a home range adjacent M12’s home range would have an elongat- Territoriality usually re- to the main road on plot 173. The farthest ed shape aligned with the main drainage sults from the need to defend resources position reached by this animal from the site canal as shown by the other probabilistic that improve an individual’s survival prob- of release was within this area. methods (FK and AK). Furthermore, these ability and reproductive success. Due to In the case of M14, the results agreed with what was perceived di- the large asymmetry in parental care be- greatest distance reached from the site of rectly on the field. Similar results were ob- tween sexes among , reproductive capture corresponded to its second home tained by Seaman and Powell (1996) when success among females is limited by the range. The centers of its home ranges they examined the estimations generated number of young they can successfully were separated by 788m. Similarly, M7’s from real and simulated data using the produce and rear, while in males it is lim- maximum distance reached from the site HMC method. It is important to keep in ited by the amount of females they can of release felt within its second home mind that certain biased or leptokurtic dis- fertilize (Clutton-Brock and Harvey, 1978). range; its home range centers were sepa- tributions will cause errors in the estima- As a result, the existence of territoriality rated by 197m. tions by this method (Harris et al., 1990). among females depends on the spatial dis-

JUN 2009, VOL. 34 Nº 6 403 tribution of food, while in males it depends above the water level within the tanks. and other regions throughout the neotropics. on the spatial distribution of females (Ost- This practice, nonetheless, has not been as- Additional studies including regional trends feld, 1985). The extensive rice field in sessed yet. in population growth, predator-mediated which this study was carried out offers a regulation and experimenting different con- very uniform habitat with an abundant and Migration trol measures will assist in the design of a renewable food supply, in which the cost of cost-effective and environmentally friendly defending this resource by females will be The management of ex- control program for this species. higher than the cost of tolerating intruders, tensive rice fields in the SRRG generates as is suggested by Ostfeld (1990) for small extensive disturbances (tank drainage, har- Acknowledgments mammals that feed on grass. Accordingly, vest, burning of leftover rice plants and no intrasexual territoriality is observed soil preparation) in the habitat of H. sci- The authors thank the among females. The resulting spatial distri- ureus, causing surviving individuals to mi- Cevallos family for giving us access to bution of females appears to allow males grate. This situation was exemplified by their farm and for facilities at their house, to include the home ranges of several fe- individuals M8, M14 and F7. M8 survived Rodrigo Rueda and Ernesto Pannier for males within their own, while not tolerat- the harvest by remaining hidden under- their assistance in the field, and Emilio ing other males, resulting in a pattern sim- ground, as was assessed by radio-tracking, Herrera for his comments on an original ilar to that reported for Microtus xanthog- and then began moving away from the site. version of the manuscript. This study was nathus by Wolff and Lidicker (1980). Rat F7 shifted its home range after the funded by FONACIT, Venezuela (S1 harvest, while M14 changed its home range 2001000690). Mobility when the tank on which it was located was drained. However, after each of the men- References The mean distance tra- tioned disruptions, the vegetative cover is versed by the studied individuals in 30min reestablished, due to the sprouting of rice Agüero D (1979) Análisis reproductivo de una po- suggests that they could potentially travel plants previously harvested or to the begin- blación de (Rodentia- ) en cultivos de arroz del Estado more than 900m in one night (12h) if they ning of a new cultivation cycle, again pro- Portuguesa. Agron. Trop. 28: 101-116. moved in one direction. Indeed, M3 was viding favorable conditions for rodents. As Agüero D, Castillo P (1981) Recomendaciones so- found 781m from the last fix recorded 23h a result, these areas become available to bre el control de ratas en el cultivo de arroz. before. The results indicate no significant animals that had been forced to migrate. CIARCO 3: 1-4. difference in this enormous potential mo- This could explain the variations in the Agüero D, Quevedo J, Tovar I, Flores A, Martínez bility between males and females. Howev- population of H. sciureus observed by V, Espinosa G (1985) Estimación de daños y er, differences among the sexes in maxi- García-Rangel (2002) according to the observaciones sobre la rata arrocera (Holochi- lus venezuelae) en caña de azúcar. C. azúcar mum distance traveled from the site of re- phase of cultivation. It is worth mentioning 3: 63-70. lease were recorded. As a result, it can be that sowing in the plots within the SRRG Alvizu PE, Aguilera M (1998) Uso del espacio de seen that males move across larger distanc- is done asynchronously, which is why there Marmosa robinsoni (Didelphidae: Marsupia- es than females in absolute terms. The are always tanks with rice in differing lia) en una zona xerófila de los Andes Vene- greater absolute mobility observed in males stages of development. This promotes the zolanos. Ecotrópicos 11: 81-92. could explain the deviations from the ex- existence of small populations of rats that Barreto GR, García-Rangel S (2005) Holochilus pected 1:1 sex ratio reported in a study by continuously rotate between tanks. Intra- sciureus. . Sp. 780: 1-5. García-Rangel (2002) at the same site. In sexual territoriality will then promote mi- Berteaux D, Masseboeuf F, Bonzom JM, Bergeron this study, the authors set the traps in the gration of males across greater distances JM, Thomas DW, Lapierre H (1996) Effect of carrying a radiocollar on expenditure of ener- periphery of the cultivation tanks, possibly than in the case of non-territorial females, gy by meadow voles. J. Mammal. 77: 359- causing the capture of a sub-population which will find a new home range with 363. conformed mainly of males and reflecting greater ease, as was observed in this study. Brown PR, Leung LK, Sudarmaji P, Singleton GR the differences in absolute mobility among Because of these disruptions, both male (2003) Movements of the ricefield rat, Rattus the genders. Consequently, it may be time and female rats will be forced to move argentiventer, near a trap-barrier system in to reconsider the common practice of dis- across rodenticide-stocked earth banks. rice crops in West Java, Indonesia. Int. J. Pest tributing rodenticides on the earth banks However, the large number of rats on the Manag. 49: 123-129. that surround the rice tanks. It is important move will deplete the rodenticide that is Brunton CFA, Macdonald DW, Buckle AP (1993) Behavioural resistance towards poison baits in to keep in mind that the habitat of these available and create a breach in the perim- brown rats, Rattus norvegicus. Appl. Anim. rats is covered by rice, and that it is very eter that will allow other rats to move into Behav. Sci. 38: 123-129. unlikely that they will especially go to the the rice tanks without getting poisoned. Burgman MA, Fox JC (2003) Bias in species earth banks to eat the bait. The poor bait Once these healthy rats get settled (females range estimates from minimum convex poly- acceptance by rodents inhabiting areas and “resident” males), it is very unlikely gons: implications for conservation and op- with abundant food resources has been that they will expose themselves on the tions for improved planning. Anim. Cons. 6: noted before (see Leung and Clark, 2005). earth banks again. 19-28. Moreover, behavioral traits may prevent Features of the ecology of Burt WH (1943) Territoriality and home range concepts as applied to mammals. J. Mammal. rats to ingest lethal amounts of poison, as marsh rats have been shown that help un- 24: 346-352. it was demonstrated for brown rats by derstand why control measures are not ef- Cartaya E (1983) Estudio de la Comunidad de Brunton et al. (1993). Our results suggest fective enough to prevent crop damage. Roedores Plaga Asociados a un Cultivo de that if some rats are going to be affected Moreover, these control measures produce Arroz (Oryza sativa L.) a lo largo de su ciclo by rodenticides, they will be mostly “tran- environmental consequences that affect both de vida, en el Estado Portuguesa. Thesis. sient” males, but their death will not re- wildlife and human health. An integrated Universidad Simón Bolívar. Sartenejas, Vene- zuela. 89 pp. duce the birth rate of the population. A pest control program including biological Cartaya E, Aguilera M (1984) Área de acción de plausible solution that has been implement- control, trapping and a rational use of ro- Holochilus venezuelae Allen, 1904 (Rodentia, ed by one of the farmers consists on set- denticides based on knowledge of the ecol- Cricetidae) en un cultivo de arroz. Acta Cient. ting the rodenticide on plastic containers ogy of the pest is much necessary in this Venez. 35: 162-163.

404 JUN 2009, VOL. 34 Nº 6 Cartaya E, Aguilera M (1985) Estudio de la comu- Kenward RE, South AB, Walls SS (2003) Ranges Squirrel Species Inhabiting a Lowland Dip- nidad de roedores plaga en un cultivo de ar- 6 v1.2: For the Analysis of Tracking and Lo- terocarp Forest in Malaysia. Biotropica 33: roz. Acta Cient. Venez. 36: 250-257. cation Data. Anatrack Ltd, Wareham, UK. 346-351. Cartes A (1979) Biología de la rata arrocera (Ho- Leung LKP, Clark NM (2005) Bait avoidance and Seaman DE, Powell RA (1996) An evaluation of lochilus sp., Sigmodon sp., Sigmomys sp. Ory- habitat use by the roof rat, Rattus rattus, in a the accuracy of Kernel density estimators for zomys sp.). I: Morfometría y densidad pobla- piggery. Int. Biodet. Biodeg. 55: 77-84. home range analysis. Ecology 77: 2075-2085. cional. Acta Cient. Venez. 30(Supl. 1): 145. Martino AMG, Aguilera M (1989) Food habits of Swihart RK, Slade NA (1985) Testing for indepen- Clutton-Brock TH, Harvey PH (1978) Mammals, Holochilus venezuelae in rice fields. Mamma- dence in animal movements. Ecology 67: 255- resources and reproductive strategies. Nature lia 53: 545-561. 258. 273: 191-195. McNab BK (1963) Bioenergetics and the determina- Tew TE, Todd IA, Macdonald DW (2000) Arable Endries MJ, Adler GH (2005) Spacing patterns of tion of home range size. Am. Nat. 97: 133-140. habitat use by wood mice (Apodemus sylvati- a tropical forest rodent, the spiny rat cus). 2. Microhabitat. J. Zool.(Lond) 250: Murie JO, Harris MA (1978) Territoriality and (Proechimys semispinosus), in Panamá. J. 305-311. dominance in male Columbian ground squir- Zool. (Lond) 265: 147-155. Twigg GI (1962) Notes on Holochilus sciureus in rels (Spermophilus columbianus). Can. J. British Guiana. J. Mammal. 43: 369-374. García-Rangel S (2002) Estudio de la comunidad Zool. 56: 2402-2412. de roedores asociada al cultivo de arroz (Ory- Twigg GI (1965) Studies on Holochilus sciureus za sativa) en el Estado Guárico. Thesis. Uni- Ostfeld RS (1985) Limiting resources and territori- barbirensis, a cricetine rodent from the coast- versidad Simón Bolívar, Sartenejas, Venezuela. ality in microtine rodents. Am. Nat. 126: al region of British Guiana. Proc. Zool. Soc. 37 pp. 1-15. (Lond) 145: 263-283. Harris S, Cresswell WJ, Forde PG, Trewhella WJ, Ostfeld RS (1990) The ecology of territoriality in Wolff JO, Lidicker WZ (1980) Population ecology Woollard T, Wray S (1990) Home range anal- small mammals. Trends Ecol. Evol. 5: 411-415. of the taiga vole, Microtus xanthognathus, in ysis using radio-tracking data - a review of Priotto J, Steinmann A, Polop J (2002) Factors af- interior Alaska. Can. J. Zool. 58: 1800-1812. problems and techniques particularly as ap- fecting home range size and overlap in Calo- Worton BJ (1989) Kernel methods for estimating plied to the study of mammals. Mammal. Rev. mys venustus (Muridae: ) in the utilization distribution in home-range stud- 20: 97-123. Argentine agro-ecosystems. Mammal. Biol. ies. Ecology 70: 164-168. 67: 97-104. Kenward RE (2001) A Manual for Wildlife Radio Worton BJ (1995) Using Monte Carlo simulation Tagging. Academic Press. London, UK 311 Saiful AA, Idris AH, Rashid YN, Tamura N, to evaluate kernel-based home range estima- pp. Hayashi F (2001) Home Range of Sympatric tors. J. Wildl. Manag. 59: 794-800.

ÁREA DE ACCIÓN DE LA RATA ARROCERA, Holochilus sciureus, UN ROEDOR PLAGA EN CAMPOS DE ARROZ DE VENEZUELA Gabriela C. Eiris y Guillermo R. Barreto RESUMEN

Se determinó el área de vivienda de ratas arroceras, Holo- fijo y kernel adaptado. El área de vivienda de hembras y machos chilus sciureus, en el Sistema de Riego Río Guárico, Venezuela, se estimó en 0,6 a 1,3ha, dependiendo del método empleado. Las mediante técnicas de seguimiento por radio con el fin de contri- áreas de machos y hembras se solaparon y posibles interaccio- buir al conocimiento de los patrones de movimiento de roedo- nes agonísticas observadas entre machos sugieren territorialidad res tropicales y proveer de información útil para el manejo de y un sistema de apareamiento poligínico. Los machos mostraron esta plaga agrícola. Diez ratas adultas fueron capturadas usan- mayor movilidad que las hembras. Estos resultados a su vez re- do trampas Tomahawk, se les colocó un radio-transmisor y se lacionan los patrones de movimiento de estos roedores causados siguieron durante la estación de sequía de 2004 desde las 18:00 por cambios ambientales en campos de arroz. Estudios ecológi- hasta las 06:00. El área de vivienda fue estimada con cuatro cos proveen información para la formulación de programas de métodos: polígonos mínimos convexos, medias armónicas, kernel control integral de plagas.

Área de aÇão dO ratO AQUATICO da espécie Holochilus sciureus, um roedor praga nos PLANTAÇÕES de arroz da Venezuela Gabriela C. Eiris e Guillermo R. Barreto RESUMO

Determinou-se a área de moradia de ratos aquáticos Holochi- fixo e kernel adaptado. A área de moradia de fêmeas e machos lus sciureus, no Sistema de Irrigação Río Guárico, Venezuela, foi estimada em 0,6 a 1,3ha, dependendo do método emprega- mediante técnicas de acompanhamento por radio com o fim de do. As áreas de machos e fêmeas se sobrepuseram e possíveis contribuir ao conhecimento dos padrões de movimento de roedo- interações agonísticas observadas entre machos sugerem territo- res tropicais e prover informação útil para a manipulação des- rialidade e um sistema de acasalamento poligínico. Os machos ta praga agrícola. Dez ratos adultos foram capturados usando mostraram maior mobilidade que as fêmeas. Estes resultados armadilhas Tomahawk, e foram monitorados, mediante o uso de por sua vez relacionam os padrões de movimento destes roedo- radio transmissores, durante a estação de seca de 2004 desde às res causados por mudanças ambientais em plantações de arroz. 18:00 até as 06:00. A área de moradia foi estimada com quatro Estudos ecológicos provêm informação para a formulação de métodos: polígonos mínimos convexos, medias armônicas, kernel programas de controle integral de plagas.

JUN 2009, VOL. 34 Nº 6 405