Effectiveness of Live Traps and Snap Traps in
Trapping Small Mammals in Kinmen
Ling-Ling Lee
Department of Zoology, National Taiwan University, Taipei, Taiwan 106, ROC
ABSTRACT
Small mammals were trapped at 17 locations with both aluminum live traps and Japanese plastic snap traps between January and May 1993 in Kinmen. Two trap lines, each with 10 trap stations set up at 10-m intervals were established at each location. One live trap and 1 snap trap baited with sweet potato and peanut butter were placed next to each other for 1 night at each trap station. Traps were checked the next morning to record the species caught and their basic measurements. A total of 125 small mammals belonging to 6 species were trapped in 680 trap nights. Live traps caught significantly more small mammals in total, and more Suncus murinus. The species caught by these 2 types of traps were also dif-ferent.
Key words: traps, effectiveness, small mammals, Kinmen
INTRODUCTION
Trap type is a very important variable affecting the results of population studies on small mammals (Wiener and Smith, 1972). Different types of traps are known to have dif-ferent trapping efficacy on difdif-ferent species of small mammals (Cockrum, 1947; Wiener and Smith, 1972; Hansson and Hoffmeyer, 1973; Boonstra and Rodd, 1982; Williams and Braun, 1983; West, 1985; Alexander et al., 1987; Lin and Lin, 1988; Slade et al., 1993). Some com-parative studies on the same types of traps gave very different results as to which type of trap is more effective. For example, Sealander and James (1958) reported that Sherman live traps were more effective in catching small mammals than the Museum Special, a snap trap. Wiener and Smith (1972) found otherwise, and sug-gested that there might have been a sampling bias in the former study. However, many stud-ies indicated that the efficacy of traps may also be affected by different localities (Williams and Braun 1983), seasons (Mengak and Guynn
1987), and other factors.
Several types of live traps and kill traps are used in capturing small mammals in Taiwan. However, only a few studies have been con-ducted to compare the efficiency of these traps. Alexander et al. (1987) compared the efficiency of Sherman live traps and Japanese plastic snap traps in trapping 2 sympatric shrew species at Mt. Ali, and found that Sherman traps caught significantly more shrews than snap traps. Lin and Lin (1988) compared 2 sizes of Sherman traps in catching small mammals at Mt. Ali, and found that larger traps (85 x 105 x 294 mm) caught significantly more Niviventer cul-turatus and female Apodemus semotus, but not male A. semotus. Both studies were conducted in the mountains. The former study used both live and snap traps in catching shrews but not rodents, the latter study tried to capture all small mammal species, but with live traps only. In a recent study conducted on Kinmen (Lee and Lin, 1994), I set up 2 types of traps, alu-minum live traps and plastic snap traps, next to each other at 17 locations to compare the
effectiveness of these traps in capturing both insectivores and rodents in the lowland area.
STUDY SITE
Kinmen is an island about 10 km to the east of Xiamen, Fujian Province, China (Fig. 1). The land area of this island is about 134.25 km2. Its shape is like a bone, wide on both the
east and the west ends, and narrow in the mid-dle. The highest point of the island, Mt. Taiwu, is 253 m in elevation. The topography is gen-erally flat with some mounds. The island has a subtropical monsoon climate, with more fog and rain between April and September, and strong northeast winds the rest of the year. Typhoons bring in more rain in summer and fall. Mean annual temperature is 21.1°C (12.8~28.8°C), and annual precipitation is about 986.9 mm.
The vegetation on the island is mostly
Acacia sp. and Casuarina sp., and cropland.
Agriculture is not intensive. Less disturbed areas are covered by thorny bushes of Scolopia
oldhamii, Maytenus diversifolia, Sageretia thea,
and Litsea glutinosa. There is little industry, and residential construction is in general local-ized. Therefore, a considerable amount of wild land remains.
MATERIALS AND METHODS
Trapping was conducted between January and May 1993. I divided the land area of Kinmen into 76 one-square-kilometer grids, and set traps at the center of 37 grids which covered the longest axes across and along the east and west sides of the island, excluding mil-itary restricted zones (Fig. 2). However, due to the number of traps available, only 17 of the locations were set with 2 types of traps. The 2 types of traps used in this study were alu-minum box live traps (26 x 9 x 9 cm) (San-Kuan Co., Taiwan) and Japanese plastic snap traps (16 x 6 cm) (Easy trap, Japan Trap Co., Japan). Two trap lines, each with 10 trap sta-tions set at 10-m intervals, were established 10 m apart at each location. At each trap station, Figure 1. Location of Kinmen.
1 live trap and 1 snap trap baited with sweet potato and peanut butter were placed next to each other for 1 night. Traps were checked the following morning, and the small mammals caught and their basic measurements were recorded.
Chi-square test was applied to compare the efficacy of the 2 types of traps, and Mann-Whitney U test was applied to compare the measurements of animals caught by the 2 types of trap.
RESULTS
A total of 125 small mammals, belonging to 6 species, were caught in 680 trap nights.
Suncus murinus was the species most often
caught (n = 57), followed by Rattus r.
flavipec-tus (44), and Mus caroli (19). The numbers of M. musculus (3), R. nor veg icus (1), and Bandicota indica (1) caught were too low for
further analysis.
Among the total of small mammals caught, 77 were caught by live traps and 48 by snap
traps. The total number of animals trapped was significantly higher with live traps than snap traps (X2= 6.73, d.f. = 1, P < 0.01).
However, different traps were more effective in capturing different species. More S. murinus were caught by live traps (n = 47) than by snap traps (n = 10) (X2= 24.02, d.f. = 1, P < 0.005).
Snap traps caught more R. r. flavipectus (n = 28) than live traps (n = 16), but the difference was not statistically significant (X2= 3.27, d.f.
= 1, 0.1 < P < 0.05). The number of M. caroli caught by the 2 types of traps was about equal (n = 11 for live traps, and n = 8 for snap traps,
X2= 0.47, d.f. = 1, P < 0.1).
The species caught by the 2 types of traps were different. Both traps caught the more abundant species, such as S. murinus, R. r.
flavipectus, and M. caroli. However, only live
traps caught M. musculus, which made the total number of species caught by this trap 4. Snap traps caught R. norvegicus and B. indica, but not M. musculus, giving the number of species caught by snap traps as 5.
Effectiveness of Traps
Figure 2. Small mammal trapping grids. In hatched grids, both live traps and snap traps
Although the mean body weight, total length, and body length of small mammals caught by these 2 types of traps were different (
P < 0.005, Mann-Whitney U test), such
differ-ences disappeared when data of each species were calculated separately ( P > 0.1) (Fig. 3), which means that the differences were due to the fact that the trapping efficiency of S.
muri-nus and R. r. flavipectus were different with
dif-ferent traps.
DISCUSSION
Previous studies on the efficacy of live traps versus snap traps gave very different results.
Cockrum (1947) reported that live traps (8"× 3"× 3") caught 2 to 3 times more small mam-mals than snap traps (Museum Specials). Sealander and James (1958) also reported that Sherman live traps were more effective than Museum Specials. Hansson and Hoffmeyer (1973) reported that live traps caught more Apodemus flavicollis, A. sylvaticus, and Sorex araneus, but not Microtus agrestis or Cleithrionomys glareolus. Alexander et al. (1987) reported that Sherman live traps caught more shrews than Japanese plastic snap traps. In this study, live traps also caught more small mammals in total than snap traps. However, Figure 3. The mean (A) body weight, (B) total length, and (C) body length of 3 small mammal
live traps caught more S. murinus, an insecti-vore, but fewer R. r. flavipedtus, a rodent, than did snap traps. Weiner and Smith (1972), on the other hand, suggested that Museum Specials were much more effective in catching small rodents than were Sherman live traps. Faust et al. (1971) pointed out, in particular, that Sherman live traps were inferior to Museum Specials in catching Blarina
brevicau-da, an insectivore.
Few hypotheses were proposed in these studies to explain the differences in the results. Here, a few possible factors need to be exam-ined and discussed. First, the types of live traps and snap traps used in these studies were similar in function but differed in size and design. The snap traps used in these studies included Museum specials, Japanese plastic snap traps, and a few other types of snap traps used in Europe (Hansson and Hoffmeyer, 1973). The live traps used in these studies were also different in their design or size. Second, the small mammals captured in these studies were different in species, sizes, and responses to traps. Third, research design and sampling bias were also different in these studies. Some studies removed all animals caught for several consecutive days (Hansson and Hoffmeyer, 1973), some studies marked-and-released ani-mals caught in live traps (Alexander et al., 1987). In the present study, traps were set for only 1 night per site to avoid bias due to mor-tality of the animals caused by snap traps and by cold and wet weather, especially S. murinus. Therefore, it is difficult to draw any general conclusion from these studies before these fac-tors are standardized or fully examined.
The effectiveness of traps may also be affect-ed by factors such as the body mass of the ani-mal. Boonstra and Rodd (1982) found that larger animals were harder to catch with a stan-dard longworth trap, because larger animals might get their back, hind feet, or tail caught under the door when the traps spring, and thus can back out to avoid being trapped. Although individuals of each species caught by live traps were not significantly smaller or lighter than
those of snap traps in this study, all animals caught by snap traps were on average larger and heavier than those caught in live traps, if they were not broken down into different species, because more R. r. flavipectus individu-als, which are larger and heavier than S.
muri-nus, were caught by snap traps than by live
traps. Therefore, we could not completely rule out the possibility that animals larger than a certain size are less trappable with live traps.
Bait may be another factor affecting the capture rates of the 2 traps. With the snap traps, animals need to bite off or move a cer-tain portion of the bait (a small sweet potato cube) to trigger the traps. If too small a piece of bait is eaten or moved, animals will not be caught. S. murinus may be attracted to the smell of peanut butter, but may be less likely to eat or remove a significant portion of this bait than are rodents, and is therefore less likely to be caught by snap traps. However, if entering a live trap, it is more likely to be caught whether it eats the bait or not, as long as it steps on the treadles.
There are at least 20 species of small terres-trial mammals resident in Taiwan (Chen and Yu 1984), and at least 3 types of live traps and 3 types of snap traps commonly used in small mammal studies. How effective these traps are in capturing different species under different environmental conditions is very important for future work, and should be examined further.
ACKNOWLEDGMENTS
I would like to thank Y. J. Lin, H. H. Su, L. C. Mei, S. C. Chiu, S. A. Chuang, S. C. Cheng, Y. L. Yang, and Y. Y. Lee for their assistance in field work, and Dr. G. G. Musser and Dr. H. T. Yu for species identification. This survey could not have happened without the kind help from Dr. C. J. Chen of Kinmen Institute of Aquaculture, Mr. S. J. Chuang of Kinmen High School, and Mr. C. G. Yang and C. C. Chen of Kinmen County Government. This project was supported by the Council of Agriculture, Executive Yuan.
REFERENCES
Alexander, P. S., L. K. Lin and B. M. Huang. (1987) Ecological notes on two sympatric mountain shrews (Anourosorex squampies and Soriculus fumidus) in Taiwan. J. Taiwan Museum 40: 1-7.
Boonstra, R. and F. H. Rodd (1982) Another potential bias in the use of the longworth trap. J. Mamm. 63: 672-675.
Chen, J. T. F. and M. J. Yu (1984) A Synopsis of
the vertebrates of Taiwan. Vol. 3: Shang-Wu
Publications, Taipei. (in Chinese)
Cockrum, E. L. (1947) Effectiveness of live traps versus snap traps. J. Mamm. 28: 186. Faust, B. F., M. H. Smith and W. B. Wray.
1971. Distances moved by small mammals as an apparent function of grid size. Ann. Zool. Fennici 8: 7.
Hansson, L. and I. Hoffmeyer (1973) Snap and live trap efficiency for small mammals. Oikos 24: 477-478.
Lee, L. L. and Y. C. Lin (1994) A survey of the natural resources in Kinmen-Wildlife. National Park Bulletin 5: 1-20. (in Chinese) Lin, L. K. and I. J. Lin (1988) The relative
effi-ciency of two sizes of Sherman live-traps. Yushania 5(2): 21-23. (in Chinese)
Mengak, M. T. and D. G. Guynn Jr. (1987) Pitfalls and snap traps for sampling small mammals and amphibians to two trap types and two trap baits in Pacific Northwest [USA] forest. Amer. Midl. Nat. 118: 284-288.
Sealander, J. A. and D. James (1958) Relative efficiency of different small mammal traps. J. Mamm. 39: 215-223.
Slade, N. A., M. A. Eifler, N. M. Gruenhagen and A. L. Davelos (1993) Differential effec-tiveness of standard and long Sherman live traps in capturing small mammals. J. Mamm. 74: 156-161.
West, S. D. (1985) Differential capture between old and new models of the Museum Special snap trap. J. Mamm. 66: 798-800.
Wiener, J. G. and M. H. Smith (1972) Relative efficiencies of four small mammals traps. J. Mamm. 53: 868-873.
Williams, D. F. and S. E. Braun (1983) Comparison of pitfall and conventional traps for sampling small mammals popula-tions. J. Wildl. Manage. 47: 841-845. (Received Aug. 9, 1996; Accepted Nov. 3, 1996)
