Guidelines to the Use of Wild Birds in Research

VIII. MINOR MANIPULATIVE PROCEDURES

A. Overview

The collection of tissue samples, experimental manipulations using injections and implants of hormones/drugs, playbacks of tape-recorded vocalizations, and presentation of decoys, are fundamental tools for ornithologists. Most if not all of these activities require permits from federal and/or state agencies. [see II]

Clearly, wild birds used in captive studies should be as healthy and free of trauma as possible. Some exceptions to this rule include investigations of the effects of environmental stress. It has been shown that passeriforms require 3-4 weeks to acclimate to captivity before experiments begin. Usually body mass declines after capture, and plasma levels of metabolic and reproductive hormones are often abnormal. After 3-4 weeks body mass returns to that of capture, and hormone levels stabilize (Wingfield et al. 1982). Obviously,

-27-

housing conditions in captivity are important and depend upon the species to be investigated. Even slight over-crowding can delay acclimation to captivity by several weeks (Wingfield et al. 1982).

B. Collection of Blood Samples

Methods for collecting blood samples from birds have been reviewed by Morton et al. (1993) and Campbell (1994). A video demonstrating blood collection techniques is available from NWHC. [see Appendix A] Common techniques include the use of a syringe for obtaining blood from the jugular vein, occipital venous sinus, or heart puncture (see also Dorrestein et al. 1978; Vuillaume 1983). However, many investigators prefer to obtain small amounts blood from the ulnar (wing) vein or from vessels in the tibio-tarsi. Heart puncture by the furcular route may result in severe debilitation or death, especially among smaller species. Utter et al. (1971) found that heart puncture via a sternal route was much less severe, and even free-living birds survived well. However, the potential for debilitation is still marked. Toe-clipping is acceptable in the field for very small birds such as hummingbirds. It is generally necessary to clip only the toenail (Leonard, 1969). (N.B. Although toe-clipping may have the helpful side effect of identifying previously sampled birds, it is not an approved procedure for marking birds.)

In most cases a suitable blood sample can be collected from the ulnar vein or tibiotarsal blood vessels. In larger species a syringe and needle is appropriate. For smaller species (e.g., less than 100 g) it is recommended that the vein be punctured with a 26 gauge or smaller needle and blood collected directly into micro-hematocrit capillary tubes. If the animal is not to be killed or incapacitated as part of the experiment, then the volume of blood to be withdrawn is an important issue (McGuill and Rowan 1989). A general rule of thumb is that no more than 2% of the body weight of the animal be collected in any 14 day period, or no more than 1% at any one time (McGuill and Rowan 1989). For a 10 g bird the maximum would be approximately 2-3 capillary tubes (100 of whole blood) and about I0 capillary tubes (500) from a 50 g bird. These limits apply whether blood is collected for DNA analysis, or whether plasma is harvested for hormone, metabolite studies etc. However, because avian erythrocytes are nucleated, not much blood is needed for most DNA studies. Hence, we recommend quantities of 1/s to ‡ capillary tube for birds <7 g, 1 tube for birds 7-15 g, and 2 tubes for larger birds. Once taken, the blood (and other tissue) samples should be properly preserved for survival under field conditions (Seutin et al. 1991).

In recent years, investigations on the response of adrenocortical hormones to a standardized stressor have been used to study adaptation to environment and to monitor species in potentially disturbed habitats (Wingfield 1994). The procedure may have additional useful application in conservation biology by comparing captive breeding individuals with their free-living counterparts. To do this requires holding the individual for a period of 30-60 rain and collecting a small blood sample at intervals for measurement of hormones. The standard stress is simply capture, handling and restraint - it is assumed that all individuals of all species will regard capture and handling as stressful (Wingfield 1994). Between samples, many birds can be held in cloth bags, which allow adequate ventilation but prevent injury if the bird struggles. These bags should be placed in a secure place in the shade and sheltered from direct effects of weather. Bags are not an appropriate form of confinement or restraint for species with long necks or long bills. [see IX.C] The combined volume of blood taken during a stress series must not exceed the equivalent of 1% of body mass. With care, sequential blood samples may be taken from the same site such as the ulnar vein without creating multiple puncture wounds. Serial collection of blood samples by heart puncture should not be attempted. This stress series protocol provides highly useful information on hormone changes in response to stress and birds are released unharmed. Care should also be taken to ensure that breeding birds are not withheld from their nests for too long. At other times the 30-60 min holding period is not a problem, unless the individual becomes separated from a flock, or could potentially lose a territory. Investigator discretion is required.

In semi-captive or free-living species, collection of blood does not affect survival (Raveling 1970; Bigler et al. 1977; Wingfield and Farner 1976; Gowaty and Karlin 1984; Frederick

-28-

1986). Moreover, normal feeding and brooding activities, molt, and ability to migrate also are not affected (Wingfield and Farner 1976; Frederick 1986). Brown (1995) found that collection of blood samples from the jugular vein of 9-day old Larus delawarensis chicks had no effect on survival rates to 21 days of age. The rate of nest desertion by adults was also unaffected. In captivity, wild birds survive well after repeated blood sampling (even at 3-7 day intervals), and body mass and hematocrits remain normal (Wingfield et al. 1982; Stangel 1986). Lanctot (1994) determined that withdrawal of blood from the jugular vein of Tryngites subruficollis chicks within 24 hr of hatching had no effect on growth or survival to fledging. Up to 0.05 ml of whole blood was collected. Further, the occurrence of hematomas on the jugular in some chicks did not impair survival. Oring (pers. obs.) found little trouble in taking blood from Charadrius vociferus chicks in amounts of 100/10 g. With chicks, the risk of dehydration exceeds that of blood loss, so investigators should take precautions to provide such birds with fluids. In summary, collection of blood samples from wing and leg veins does not impair behavioral patterns, reproduction and survival of wild birds.

C. Collection of Other Tissues

Techniques in modern physiology and genetics often require biopsy of any of several tissues. Those most commonly sampled (in addition to blood) are adipose tissue, muscle, liver, and gonad. Handling time must be minimized, especially with breeding birds. If periods longer than a few minutes are routinely necessary, as they may be if the sampling procedure is complex, then a justification should be included in the ECHOIC protocol.

Various studies (e.g., Baker 1981; Westneat 1986; Westneat et al. 1986, and Frederick 1986) show that biopsy has little effect on body condition or survival in either wintering or breeding birds. After prompt handling and release, the bird often returns to normal foraging and breeding activity. Males often sing within minutes of release, and even nestlings that were biopsied showed no debilitation, begged for food, and were fed normally. It should also be noted that biopsy of the pectoralis major muscle does not hinder flight. Samples taken should involve the minimal amount of tissue necessary for scientific validity. ECHOIC approval is strongly recommended for tissue sampling if anesthesia is not used. [see IX.D]

Feather pulp is also collected for genetic investigations. Plucking a few feathers is usually a relatively innocuous procedure, but care should be taken not to remove so many feathers as to impair flight or other essential functions (this is less of a problem in a captive subject). Note also that removal of growing feathers can result in bleeding, and release should be delayed until this has stopped.

D. Collection of Food Samples and Forced Feeding

Obtaining information on a species' diet in the field is often an important component of ecological and nutritional studies. Use of neck ligatures to obtain food samples from nestlings may occasionally be justified. In such cases, the investigator should be careful to ensure normal blood circulation and tracheal function. Further consideration should be given as to whether the procedure will result in unwarranted food deprivation.

In some instances birds are sacrificed for direct observation of stomach contents. In many cases, however, this is not necessary, and collection of fecal samples and regurgitated pellets can provide most, if not all, of the information needed. However, in some species fecal material is not useful (e.g., frugivores). In other cases, such as marine birds at sea, it is not possible to collect fecal samples, although many will regurgitate stomach contents soon after capture. However, others may not, and the use of palpitation techniques and emetics may become necessary. It should be noted that emetics are potent in their action and some mortality results from choking, severe trauma, and shock (Prys-Jones et al. 1974). Wilson (1984) and Ryan and Jackson (1986) have developed a stomach pump that can be used as an alternative to emetics and sacrifice. They showed that the pump gave qualitative and quantitative results (at least in larger birds) comparable to those obtained from sacrificed birds. Pumping also had no apparent ill effects. However, the potential mortality from use of emetics has been greatly reduced by administering small quantities of the emetic through a small tube inserted down the esophagus into the proventriculus (Gionfriddo et al. 1995).

-29-

Once again, knowledge of the species to be studied is important in assessing whether it may respond adversely. Grebes, for instance, should not be subjected to any regurgitative techniques because of the feathers in their crops (Jehl, pers. comm.)

Nutritional investigations may require force feeding of experimental subjects (usually in captivity). Tube feeding using a soft rubber or atraumatic metal feeding tube of proper size and volumes of food that are appropriate for the size of the bird is safe and effective. Murphy and King (1986) found that force feeding by inserting a tube down the esophagus was injurious in some cases. Food has to be fed as a slurry, and regurgitation can result in choking (especially in small species). Intubation may also injure the esophageal wall. As an alternative, Murphy and King (1986) suggest feeding pelleted food by placing pellets directly into the pharynx with forceps, thereby inducing reflexive swallowing. Mortality is reduced to near zero, and regurgitated pellets do not result in choking, but use of pellets takes much longer than tubal feeding.

E. Cloacal Lavage

Studies of the mode and timing of insemination are important for analysis of population trends, transfer of genetic information, and mating systems. Cloacal lavage, of both males and females, is a technique to acquire information concerning sperm production and transfer (Quay 1984, 1986a,b,1989). The technique is sometimes extended by the implantation of cloacal microspheres (Quay 1988). Like stomach pumping, lavage is not invasive in the sense that it does not require penetration of an epidermal barrier. As with stomach pumping, it should produce only a slight and temporary discomfort when performed by a properly practiced person.

F. Injections and Insertion of Implants

In the United States, all drugs/medications administered to birds must be used according to the specifications of the Animal Medical Drug Use Clarification Act, which requires that extra-label use must take place under the supervision of a veterinarian.

Injection of experimental substances is widespread in research on birds. Subcutaneous and intramuscular injections are simple in the laboratory and cause little trauma. Intravenous injections require some acquired skill. Intraperitoneal injections require justification because some drugs may irritate the viscera and because of possible mechanical or chemical damage of the viscera.

Under field conditions, most injections (especially subcutaneous injections) appear to have no effect (independent of the substance injected) on survival or normal activities (e.g., Reyer 1984). For longer term studies, however, repeated injections are often necessary, requiring multiple capture at frequent intervals. This in itself may cause serious disruption of normal activities. For these reasons, implants in silicone rubber tubes, pellets, or miniosmotic pumps should be used to provide long term administration of the experimental substance (up to several weeks). Whenever possible, such implants should be made sub-cutaneously because intraperitoneal implants are often encapsulated by connective tissue. Implants inserted under the skin of the flank or side of the thorax are most effective and are easy to remove after the experiment is terminated. Implants should not be placed on the back because they frequently rupture the skin, allowing infection. Implants under the skin of the neck are also not advised: they can penetrate the thoracic cavity, resulting in severe respiratory distress. Custom-made, mini-osmotic pumps are available for odd-sized animals or for administering substances for prolonged periods. (Alza Corporation, Palo Alto, CA. The company provides a free training video for the use of these pumps in rodents; it should be applicable to birds.) As with other invasive procedures, the area of operations should be as sterile as possible.

Timing of implants is also important in some cases. Treatment of free-living birds with hormones usually has no debilitating effect, but some treatments, such as the sex steroids, can disrupt the normal temporal progression of reproductive and associated events. Molt and migration can be abolished by implants of sex steroids. Hence, every effort should be made to remove the implant after the experiment. In those species that breed at high latitude or altitude, the short breeding season allows only a short time for molt. If these func-

-30-

tions are disrupted by implants, death may result due to poor plumage and delayed migration. It is recommended that, when possible, all implants be removed from controls and experimentals. However, experimental subjects with control implants, or implants from which all hormone has diffused, do survive over winter at the same rate as unimplanted individuals (Wingfield 1984). Further, the stress of recapture may cause more problems than it solves. A crucial element in assessing appropriate actions in all of the above is whether the risk induced by the experiment applies primarily to individuals or to the population. [see I.A]

G. Determination of Egg Viability

Certain experimental procedures require an estimation of the number of eggs within a clutch that have viable embryos and the age of embryos. A common technique uses transillumination (candling) to detect the presence of an embryo, but many species have eggs with shells too thick or too heavily pigmented for candling to be useful. Breaking open eggs, with obvious deleterious effects on reproductive success, is justified in some cases, but ultrasonography is the technology of choice. Electronic devices such as doppler stethescopes that can detect the embryonic heart beat or movements of the embryo within the shell may also be useful. Some of these techniques are being adapted for use in the field (e.g., Mineau and Pedrosa 1986). Floating the egg is useful for some species. If the egg is more than ten days old and does not float, there is no viable embryo.

H. Playback of Tape-Recorded Vocalizations and the Use of Decoys

Playback of tape-recorded vocalizations to free-living birds causes little disturbance or trauma if the duration of the playback is kept within reasonable bounds (normally less than 30 minutes). More prolonged playback may distract subjects from activities that are essential to reproductive success. Unless required for the experiment, speakers should not be placed close to the nest, etc. Activity while on the territory should also be minimal to avoid destruction of the local habitat (Johnson et al. 1981; Marion et al., 1981; Baptista and Gaunt, 1997).

Live decoys are frequently used in conjunction with playbacks. The same guidelines hold: minimize investigator activity and avoid placing the decoy close to the nest. Live decoys require particular attention in the field. Birds used in this way should be trained for a day or so prior to onset of the experiment. An untrained bird tends to flail around the cage when placed on another's territory. A decoy habituated to housing in a cage under field conditions provides a more appropriate experimental stimulus and is also subjected to less stress. Note also that the decoy must be provided with food and water at all times. Avoid placing the decoy in exposed situations, especially on hot days. Never expose a decoy to full sunlight without some form of shelter. [see IV.C]

Recorded playbacks are often employed in recreational uses of birds. ABA recognizes that such usage, as well as several other practices, e.g., photography, may affect birds adversely, and have addressed that issue. [Appendix B]

I. Artificial Eggs

The use of artificial eggs is invaluable to many ornithological studies, allowing reduced risk during trapping and providing for the development of eggs of special value (e.g., in the maintenance of threatened populations). Artificial eggs composed of a variety of materials, including wood, paper mache, plastic, and clay, have elicited normal nesting responses. However, egg recognition varies widely among species. In some species, individuals recognize the unique patterns of their own eggs. For others, egg recognition mechanisms may be very general. When eggs are used briefly, e.g., during trapping, a general approximation of real eggs will suffice. However, when it is intended that artificial eggs be incubated for days or weeks, extreme care should be given to the mimicry of the original egg shape, size, pattern and weight. Birds uncomfortable sifting on surrogate eggs may desert.

J. Experimental Manipulation of Plumage

Altering the external appearance of a bird by manipulating the size and color of plumes, waffles, etc. has proved to be a powerful experimental tool in behavioral ecology. Imping, a technique for identifying individuals by "transplanting" one or more feathers from a bird of a

-31-

different color, extends back into antiquity. Under captive conditions such manipulations are not traumatic unless, as a result, the experimental subject has difficulty feeding and drinking. Under natural conditions, however, it is important to ensure that such manipulations do not impair flight or other types of locomotion.

(-32-)