Collecting the Data

Individual common weasels can be identified by their unique belly patterns (see Figure 1.12), but only when they are in the hand. The sight of a weasel in the wild is never close enough to see a belly pattern, and certainly not close enough to read an eartag number. Weasels have few regular habits that might offer an observer a sure observation "ambush," and their dens are hard to find in most habitats except with the help of a trained dog.

Snow tracking (Chapter 10) can give a lot of information without disturbing the animals, but it is not possible in summer, or at all in mild climates. More important, one cannot tell which weasel made a given set of tracks. If tracks in snow are found in the same area day after day, it is a fair bet that they were made by the same animal. But since successive residents on the same ground tend to use the same dens and runways (Musgrove 1951), one would prefer the certainty of identification offered by an eartag.

The traditional way to observe weasels year-round in temperate habitats is indirectly, by catching them in live traps, marking them, and then releasing them in the hope of collecting a series of new location records. Unfortunately, live trapping weasels is often unrewarding, at any rate in certain places and in some years. To begin with, while one may fairly assume that where a weasel has been caught it must at least be present, the reverse is certainly not true.

Modeling studies using the frequency distribution of live-trapping data (so many individuals caught once, so many different ones caught twice, so many three times, etc.) can be used to infer the presence of individuals in the population that are never caught, and they could be a substantial proportion of the local population. In one New Zealand study of live-capture data for stoats, the probability of being captured for the first time was only about 17% per day (King et al. 2003a).

These data also showed that even the weasels that have been captured once are less likely to be caught again, and explains why some quickly learn to become downright trap-shy. They confirm the stories told by trappers such as Cahn (1936), who had a battle of wits with a stoat that learned, from a single experience, to avoid a further 32 carefully sited traps. Murphy and Dowding (1995) set a ring of live traps around a den known to contain a litter of young stoats, of which up to four at a time were seen playing around the den entrance, but none was ever caught.

Worse yet, different categories of weasels vary a great deal in the way they react to traps. Resident individuals, which are often older and confident of win ning any encounter with an intruder, tend to be bolder and less shy of traps than the nonresidents, which are usually younger, insecure, and continually on the defensive. In the New Zealand data the probability of recapture for adult males was higher than for adult females, and both were higher than for young of either sex (King et al. 2003a).

Again, these calculations confirm the long experience of field biologists that females are more difficult to catch than males. Many studies of capture-mark-recapture records of weasels report that collecting a series of location records for females is especially difficult (Lockie 1966; Erlinge 1974; King 1975c). On the other hand, common weasels caught in live traps set for rodents are much more often females than males (King 1975a). The difference is not due to any attribute of the traps, but to the fact that traps set for small rodents are always laid out closer together than kill traps set for weasels by gamekeepers. Additional reasons for this bias are that females travel less each day and have smaller home ranges that enclose a smaller number of traps laid out at any given density (Buskirk & Lindstedt 1989).

What proportion of a local population of stoats is caught and marked in a live-trapping survey? Attempts to answer this question are usually frustrated, because, aside from any difference between individuals in reactions to traps, field methods almost always provide more opportunities for some stoats to be captured than others. For example, King and McMillan (1982) reported a simple experiment using regular live trapping to monitor a large number of stoats marked with eartags. Of one group of 21 stoats, tagged on or before January 15 and known to be present in the same area on or after January 25, nine were not caught on any of the 7 days on which the traps were set between January 15 and 25. In other words, a full third of the 21 stoats known to be alive were not recaptured in a whole week's trapping. Since the traps were set in a transect, which could not sample all home ranges equally, it may be that some individuals were simply not close to a trap during those days.

These data referred to a very short period during a period of high numbers of mice, when the probability of trapping a stoat may be lower than normal (Alterio et al. 1999; King & White 2004), but its general conclusion was later confirmed over a longer period and using completely different technology. Dilks and Lawrence (2000) monitored the responses of stoats to bait stations using miniature video cameras placed inside the tunnels. Over a period of 5 weeks of continuous observation, they filmed 45 occasions on which a stoat approached the entrance, but on 8 of these (18%) it did not enter.

Clearly, one has a better chance of catching a resident animal by setting several traps in its home range. Therefore, traps set on a transect line must not be set farther apart than the width of the average home range. Early field experiments in southern New Zealand showed that close-set traps (100 m between sites) catch a higher proportion of the locally resident stoats, but they also cost much more time and effort to operate (King 1980a). Intensive radio-tracking work has confirmed that the widest trap spacing for stoats, that minimizes effort and still puts at least one trap in every home range, is less than 1 km between sites in grassland (Moller & Alterio 1999) and 900 m in podocarp forest (Miller et al. 2001). For smaller stoats or weasels elsewhere in the world, these spacings must be smaller still.

Closer spacings increase both the work and the number of traps per home range, which in turn increase the probability of putting a trap in the right place. To get five traps in every female's range, spacing must be cut down to about 250 m (Moller & Alterio 1999). Yet, some traps constantly catch many more animals than others, although it is hard to define the exact characteristics of the most successful trap sites. Experienced trappers get an eye for good sites, but, whether by judgment or by luck, the way the traps are set out decisively influences the number, sex ratio (see Table 13.2), and proportion of the total population captured. Spacing, baiting, number of days set, and so on are all important.

A live-trapping study will, therefore, not sample the local population very accurately. While the male residents are being trapped day after day and the female residents occasionally, many nonresidents will pass through the area unseen and uncounted. As a further complication, the probability of trapping all classes of weasels may also change with the density of prey (Alterio et al. 1999; King & White 2004). It is hard for the field biologist to balance the size of the study area and the density of traps so as to sample both sexes adequately without dying of boredom from tramping around too many empty traps. Moreover, trapping is not an ideal method of working out the home ranges of any animals, because a resident held in a trap is unable to continue its normal life until it is released, usually many hours later.

So what are the alternatives? The ideal is to observe identifiable live weasels in the field by nonintrusive tracking, and in recent years various methods have appeared or are being developed. Some depend on trapping to start with to equip each weasel with an identifying mark; other methods make use of existing individual characteristics.

Natural Marks

The wanderings of a weasel with a unique footprint could be followed with minimal interference using a large number of tracking tunnels (Jones et al. 2004). For decades, biologists have recorded animal tracks in simple tracking tunnels. Smoked plates or sooted paper (now conveniently made with spray-on "sight black") (E. Rogers and D. Tiller, unpublished) or ink and paper (a pad of "ink" and two papers sprayed with a chemical that reacts with the "ink" to produce an indelible blue dye, or even simple food coloring), have all been used successfully (Mayer 1957; King & Edgar 1977; Zielinski 1995). Weasels readily enter well-set tunnels, especially if they are baited, and this method could give many records for each marked animal in a short time.

The problem is how to give each weasel a unique footprint. Field biologists in the United States have long marked small rodents by surgically removing one or two toes (Powell & Proulx 2003), but extension of this method to the larger toes of weasels, even under anesthetic, probably would not and should not be permitted by any Animal Ethics Committee. Instead, it may be possible to take advantage of computer technology to recognize natural differences between footprints.

For example, Herzog (2003) found that the footpads of fishers are unique, much as human fingerprints are. Herzog digitized the patterns on imprints of the pads left on tracking papers by captive and free-ranging fishers, analyzed them, and showed that individual fishers can be identified from their tracks alone. The tracks of weasels are smaller, and obtaining very clear and detailed tracks from them is often difficult, but the idea is intriguing. The two-component dye technique described by King and Edgar (1977) was adapted for field use from the method used by police to take finely detailed fingerprints, and could in principle produce animal prints clear enough so long as a suitable glossy white paper is used.

Another new method of locating animals with even greater promise is to identify individuals from their DNA. In species that regularly use conspicuous latrines, samples of DNA can be collected from feces (Kohn et al. 1999), and that could work for weasels in some habitats, such as the far north where winter dens are easy to find. In other habitats, it would be simpler to collect weasel DNA from "hair traps" (Woods et al. 1999). A network of tunnels, each with a miniature curry comb or a piece of Velcro inside, can grab loose hairs from animals passing through. DNA extracted from the roots and follicles of hairs collected in them can identify every individual, and the practical application of this method is well advanced in New Zealand (Gleeson et al. 2003). One temporary deterrent is that DNA analysis is presently rather expensive, and hair-collecting devices have to be closed after the first sample and cleared manually, but advances in technology are sure to bring the labor and processing costs down rapidly.

Artificial Marks

The main drawback of tracking naturally marked animals is that the weasels will be recorded only where the investigator has set tracking tunnels or hair traps. In addition, almost nothing can be found out about each animal other than its unique track pattern or its DNA profile. What we need is information about each individual, and a means of following it wherever it chooses to go.

Radiotelemetry is now a very sophisticated technology and is the method of choice for many studies. It provides a more accurate picture of the movements and activities of individual animals than can trapping or tracking tunnels or hair traps, and allows researchers to collect extensive data. Telemetry has not been applied much to the study of weasels, however, for several reasons.

First, it is still important to be able to trap the animals to fit them with radiocollars in the first place, and to retrap them whenever the transmitter is lost or the battery runs down. In animal transmitter packages, the batteries use more space and weigh more than all other components combined. A weasel's small size limits how big a transmitter package it can carry and, hence, how long the package will transmit. Weasels are often uncooperative about returning to have their transmitters serviced, which is a problem because battery life can only be short for small transmitters. At present, even the most efficient systems can provide no more than 6 months of battery life (Gehring & Swihart 2000), although that is a distinct improvement over the 30 days expected at best from early transmitters for weasels (Erlinge 1977b).

Second, designing a collar that will stay on a weasel's muscular neck is difficult. A collar that is too tight risks abrading hair or skin, while a collar that is too loose will surely slip off over the weasel's head. Weasels tend to shed their collars quickly, most within a couple of weeks (Murphy & Dowding 1994; Hellstedt & Kallio 2005); on the other hand, a well-fitted collar can remain in place too long. One male common weasel was recaptured, still carrying its useless burden, more than a year after the 21-day battery had expired (Delattre et al. 1985). Implanted transmitters can avoid the collar problem, but they are too large for weasels and have a much smaller transmission range than collars.

Third, the range of any transmitter is usually short (from 50 to a few hundred meters in thick vegetation) compared with a weasel's daily movements. Range can be increased by transmitting stronger signals or using a vehicle-mounted antenna, which can locate a 6-g collar carried by a male longtail up to a mile away in open country (Gehring & Swihart 2000), but the cost is a shorter battery life or a slower pulse rate for the transmitter beeps, making the weasel harder to find. Because all weasels make such good use of cover, even radio-tagged weasels are seldom seen (Sleeman 1990).

Fourth, there is the worry that a collar or implant will limit a weasel's normal activity by affecting the size of hole it can explore. Erlinge (1977b) reported that his stoats disliked their collars and scratched at them at first. The weight (6 to 10 g) and bulk of a transmitter must be maddening for such a lithe animal. In time they seem to accept them, though it is not clear whether they can hunt in all the usual, confined spaces and behave completely normally. Some studies using telemetry have assessed in captivity whether transmitters appeared to affect the weasels, and affirmed that they did not (e.g., Gehring & Swihart 2000). Others have come to the opposite conclusion, even with much lighter (2.5 g) collars (Delattre et al. 1985), so the question has not yet been answered for certain.

Ultimately, any study aiming to obtain adequate data on known, individual weasels must start by catching them in live traps, marking them, releasing them, and then collecting a series of new location records, either through successive captures, relocations using radiotelemetry, or a combination of methods. Smarter live trapping is therefore the key to further progress.

Improving the design of live traps would be a great step forward, especially if they could be equipped with alarms to call for attention when sprung. Finding good live-trapping sites also takes some skill and a lot of luck. Traps must be set in places weasels are likely to visit (in or under thick vegetation, in the roots of old trees, on fallen logs, alongside streams, under stacks of wood or heaps of stones, in stone walls and hedgerows), and one soon gets an eye for suitable sites (Figure 8.1).

Traps must also be protected from direct sun, excessive rain, potential flooding, and other dangers of the environment. Wooden traps offer the best protection, which is why we both prefer to use them, even though they are heavy and inconvenient to handle. Weasels are highly strung animals, and they really need the darkness, the privacy, and the insulation afforded by a wooden trap provided with adequate bedding. To avoid catching mice and voles, which are much more numerous than weasels and as fond of exploring holes, the mechanism can be weighted so that only animals heavier than, say, 30 g can be caught. This is worth doing because unselective wooden traps will often be blocked against weasels and damaged by gnawing. A trap suitable for stoats, incorporating a separate, insulated nest box that can double as an anesthetizing chamber, can be hand made from published plans (King & Edgar 1977).

By contrast, a night spent in a metal or wire-mesh trap (Belant 1992) can be a cruel experience for a weasel. The sizes and placement of treadles on ready-made, welded wire traps leave too little space for adequate bedding to keep weasels warm during their protracted stay, and it is difficult to protect open-mesh traps from drafts, rain, or excess heat. In addition, captured weasels bite and pull at the wire, often breaking their teeth and skinning their noses. Solid metal traps such as the Sherman can become ovens or freezers, sometimes even both on the same day, unless a detachable wooden nest box is added. We strongly advise against using any kind of metal trap for weasels.

In some habitats, placing traps where weasels must find them is so easy (e.g., in stone walls through open country and along ditches and hedgerows on traditional farms, or along streams or dry stream beds: R. A. Powell unpubl.) that weasels can be trapped without bait, although baited traps tend to catch more (King & Edgar 1977). In forested study sites, attractive trap sites must be carefully constructed and well baited to entice the local weasels to their doors. In some countries, traps can be baited with a live mouse, which both lures the weasel in and provides it with a fresh meal while it is waiting to be released (Erlinge 1974; Lawrence 1999), but providing for the needs of the mouse while it lives adds a lot of extra work.

If traps are left locked open before or between trapping sessions, the weasels get used to going in and out, and sometimes use them as temporary dens or larders. The residents can be counted as soon as the traps are set, and some shyer ones may be caught (at least once) that would otherwise never have ventured into any trap.

Figure 8.1 Three representative trap sites and sets for weasels. Placing a trap in a covered place and making the entrance look like that of a tunnel entices weasels to snoop around and inside the trap.

We consider (others differ) that live-trapped weasels are best handled under general anesthesia, preferably using oral (inhaled) anesthetics. The use of anesthetic ether, pioneered by Lockie and Day (1964), has now given way to more modern drugs such as halothane or ketamine, always after instruction from a vet and under supervision or permit. Present anesthetics are relatively safe for the animal and easy to use; they minimize fright to the animal and the risk of bitten fingers; and they simplify data collection and recording, outfitting with a transmitter, and inserting eartags (Figure 8.2). Tiny PIT (passive integrated transponder) tags implanted under the skin are even better than eartags, because they are less likely to be lost.

Captured animals recover quickly and, if handled carefully, seem not to find the experience traumatic. At least some residents can be recaptured often (King 1975c; King & Edgar 1977; King & McMillan 1982; Gehring & Swihart 2000; Purdey et al. 2004). On the other hand, recovery from anesthesia is never guaranteed; released animals do sometimes die when they get wet or cold; and keeping animals in captivity to recover can affect the social structure of the population. To overcome this problem, Gehring and Swihart (2000) developed a most commendable technique that allowed their long-tailed weasels to recover in a wooden nest box (rationed with a couple of mice) in the field. They opened the nest box remotely after a couple of hours to allow the weasel to leave at will, and removed the box later.

Was this article helpful?

0 0

Post a comment