A NONINVASIVE TECHNIQUE FOR BLOCKING VOMERONASAL CHEMORECEPTION IN RATTLESNAKES

：芭芭拉

A NONINVASIVE TECHNIQUE FOR BLOCKING VOMERONASAL CHEMORECEPTION IN RATTLESNAKES
C Patrick Stark, David Chiszar, Hobart M Smith. The Psychological Record. Gambier: Fall 2006. Vol. 56, Iss. 4; pg. 471, 17 pgs

Abstract (Summary)
To examine the effects of vomeronasal deprivation on strike-induced chemosensory searching (SICS) in rattlesnakes we used a newly developed technique to anesthetize the vomeronasal organs. We compared rate of tongue flicking after striking prey in avomic rattlesnakes to vomic controls. Avomic rattlesnakes exhibited significantly fewer tongue flicks after striking prey than did vomic controls. However, avomic subjects still exhibited a significantly higher rate of tongue flicking (RTF) than did subjects that were presented with prey but not allowed to strike indicating that the increase in RTF seen in avomic snakes reflects a component of predatory behavior that is released by the strike and not dependent upon further vomeronasal activation. A second experiment was conducted to examine the role of the vomeronasal system (VNS) in location of envenomed prey. Disruption of the VNS chemoreception significantly reduced the number of poststrike tongue flicks, significantly increased the latency to locate the prey item and significantly increased the latency to commence swallowing the prey item, indicating that normal expression of these behaviors is dependent upon a functioning vomeronasal system.

：芭芭拉

Full Text (6663  words)

Copyright The Psychological Record Fall 2006

[Headnote]
To examine the effects of vomeronasal deprivation on strike-induced chemosensory searching (SICS) in rattlesnakes we used a newly developed technique to anesthetize the vomeronasal organs. We compared rate of tongue flicking after striking prey in avomic rattlesnakes to vomic controls. Avomic rattlesnakes exhibited significantly fewer tongue flicks after striking prey than did vomic controls. However, avomic subjects still exhibited a significantly higher rate of tongue flicking (RTF) than did subjects that were presented with prey but not allowed to strike indicating that the increase in RTF seen in avomic snakes reflects a component of predatory behavior that is released by the strike and not dependent upon further vomeronasal activation. A second experiment was conducted to examine the role of the vomeronasal system (VNS) in location of envenomed prey. Disruption of the VNS chemoreception significantly reduced the number of poststrike tongue flicks, significantly increased the latency to locate the prey item and significantly increased the latency to commence swallowing the prey item, indicating that normal expression of these behaviors is dependent upon a functioning vomeronasal system.


Rattlesnakes (e.g., prairie rattlesnakes, Crotalus viridis viridis) that feed on rodent prey typically release prey immediately following a successful predatory strike, allowing the rodent to flee from the site of envenomation (Estep, Poole, Radcliffe, O'Connell, & Chiszar, 1981; Gans, 1966; Kardong, 1986; O'Connnell, Greenlee, Bacon, & Chiszar, 1982; Radcliffe, Chiszar, & O'Connell, 1980). Rattlesnakes then exhibit a sustained high rate of tongue flicking (RTF) and commence highly stereotyped side to side searching movements with the head (Dullemeijer, 1961). This sequence of behavioral events has been labeled strikeinduced chemosensory searching (SICS; Chiszar, Radcliffe, & Scudder, 1977; Chiszar, Radcliffe, & Smith, 1978) and facilitates the identification and following of trails left by envenomated prey (Dullemeijer, 1961 ; Golan, Radcliffe, Miller, O'Connell, & Chiszar, 1982).

High RTF seen during SICS indicates increased utilization of the vomeronasal system (VNS) (Burghardt, 1970, 1980; Graves & Halpern, 1-9-8-9, 1990; Halpern, 1992; Halpern & Kubie, 1980, 1984; Kahmann, 1932, 1934). As the oral ducts leading to the vomeronasal organs lay within the direct pathway of the retracted tongue, it is generally assumed that the tongue serves as the normal means by which chemical cues are delivered to the VNS in snakes (Kahmann, 1932, 1934; reviewed in Young, 1993). Indeed, partial or complete removal of the bifid tongue produces a marked reduction in behaviors believed to be mediated by the VNS such as aggregation in garter snakes (T. sirtalis; Burghardt 1983), and prey-trailing ability in rattlesnakes (Dullemeijer, 1961).

Although the high RTF exhibited during SICS has been linked to utilization of the VNS (Burghardt, 1970, 1980; Graves & Halpern, 1-9-8-9, 1990; Halpern & Kubie, 1980, 1984), it has been shown that sustained elevation in RTF occurs in rattlesnakes after striking prey regardless of whether or not prey-related chemical cues are present in the snakes' environment (Golan et al., 1982; Scudder, Chiszar, & Smith, 1983). That is, after a successful strike, the snakes search for a trail that would normally be available and do so systematically even in circumstances when the trail is not present, as in certain laboratory experiments investigating SICS behavior. Thus, the elevation in RTF seen during SICS is believed to be dependent upon the delivery of a predatory strike, and not on taxic feedback from chemical cues (Chiszar, 1986, Chiszar et al., 1977; Golan et al., 1982).

In studies examining the ability of T. sirtalis to follow trails made with earthworm extract, however, snakes followed concentrated trails more efficiently than more dilute trails and emitted a higher RTF while doing so. Thus, for T. sirtalis there appears to be a feedback system between tongue flick mediated stimulation of the VNS and the motor mechanisms that produce tongue flicking (Halpern & Frumin, 1979; Halpern & Kubie, 1983). This raises the possibility that the magnitude of RTF seen during SICS in rattlesnakes may also be influenced by an effect of chemical cues on the chemoreceptive systems that mediate tongue flick behavior. If feedback exists in rattlesnakes between the vomeronasal system and the motor systems controlling the tongue, we hypothesized that disruption of VNS chemoreception during SICS should reduce the magnitude of poststrike RTF.

Past attempts to disrupt VNS chemoreception in rattlesnakes have relied upon various surgical manipulations. These techniques have revealed a marked reduction in normal predatory strike responses. The degree of this reduction has ranged from about 50% to 100% of vomeronasally deprived subjects failing to strike when presented with prey (Alving & Kardong, 1996; Graves & Duvall, 1985). While these techniques have provided valuable information concerning the role of the VNS in prestrike and strike behaviors, the resultant disruption of normal predatory strikes makes it difficult to examine the effects of loss of vomeronasal chemoreception on strike dependent behaviors, such as SICS. Therefore, one purpose of the experiments presented here was to develop a technique that would disrupt O p e r ation of the VNS without simultaneously eliminating strike behavior.

Using the methodology of Andren (1982), we attempted to apply anesthetic Xylocaine tape to the vomeronasal ducts in several species of rattlesnakes. We also manufactured form-fitting plastic inserts that could be temporarily sealed with adhesive to the anterior roof of the snakes' mouths, thus blocking access to the vomeronasal ducts. A third method involved gluing the vomeronasal ducts closed using veterinary skin adhesive.

Although each technique possessed its relative merits, all of these methods required extensive handling of the subjects. Rattlesnakes handled in this way became defensive (Chiszar, Radcliffe, O'Connell, & Smith, 1981), such that using them immediately thereafter in feeding experiments was impossible. In fact, defensiveness often persisted for several hours. Hence, it was necessary to develop a method of limiting the sensory responsiveness of the vomeronasal epithelium that did not require handling of the subjects. Experiment 1 introduces a new method that accomplished this goal. Snakes were presented with mice that had been covered with a thin coating of Xylocaine ointment. Upon striking the prey item, the Xylocaine ointment was transferred into the snakes' mouths (due to the location of the fangs, the upper anterior portion of the mouth received the majority of the ointment) where it was presumably transferred into the vomeronasal ducts and the sensory epithelium. Xylocaine (5% lidocaine ointment in a petroleum-based vehicle, similar in consistency to Vaseline) is an effective topical anesthetic used in both medical and veterinary practice to anesthetize mucous membranes. Its effects are fast acting with an onset latency in the adder (Vipera berus) demonstrated to be approximately 2-3 min with a duration of about 2 hr (Andren, 1982).

Surgical techniques used in previous research, although highly effective in eliminating vomeronasal input, produce difficulties beyond those associated with eliminating strike behavior. Vomeronasal nerve transection is irreversible and results in the eventual death of the subject. There is also the possibility that complete lack of afferent activity, over the extended time period required for recovery from surgery and complete testing, could produce degeneration of nerve tissue in vomeronasal projection sites, resulting in perturbations of surrounding systems connected to the behavior under examination (e.g., Kubie & Halpern, 1979). Likewise, suturing techniques, although reversible, produce mechanical trauma within the mouth of the snake that could produce changes in behavior outside of those produced by loss of VNS input. The new method presented in the following experiments effectively disrupts VNS chemoreception and carries the additional benefits of being a reversible procedure, not imposing mechanical limitations upon the subject, not causing undue suffering or tissue damage, not causing defensive behaviors, and not remaining in effect for extended periods of time.

：芭芭拉

☆Experiment 1

Method

Subjects

Subjects were 12 rattlesnakes (3 Crotalus atrox, 3 Crotalus oreganus oreganus, 3 Crotalus oreganus lutosus, and 3 Crotalus oreganus helleri). They were captured as adults and had been in captivity for at least 3 years. These snakes were housed individually in glass terraria (50 x 27.5 x 30 cm) and, prior to testing, were accepting rodent prey (Mus musculus or Rattus norvegicus) on a fortnightly schedule. The room was kept at 26 to 28 °C during photophase (0700-1900) and at 22 to 24 °C during scotophase. These snakes had been used in previous experiments involving presentation of chemical stimuli; however, no surgical or pharmacological procedures took place. Therefore, these snakes were representative of long-term captive rattlesnakes. All procedures were approved by the local IACUC and were in accord with the guidelines of the Society for Neuroscience.

Procedure

Snakes remained within their home cages during all phases of the experiment. One week prior to the first week of testing, snakes were fed one juvenile R. norvegicus (100 g). To control for the effects of satiety, snakes were not fed during the 4 weeks of testing.

To establish that the subjects were in a state of quiescence prior to experimental testing, each trial began by the observer recording the initial rate of tongue flicking with a handheld counter for 10 min. Following this initial recording, the subject was presented with one of the following experimental conditions: (a) a target mouse coated in Vaseline, (b) a target mouse coated in Xylocaine ointment, (c) a nontreated target mouse (control for the effect the petroleum vehicle), and (d) a nontreated mouse held just out of striking range (control for the effect of striking).

Preparation of Target Mice

Mouse carcasses used as target mice were previously killed by cervical dislocation and stored frozen at O °C. On the day they were to be used, mice were thawed to room temperature. Just prior to use in an experimental trial, mice were warmed to approximately 35 °C on a stainless steel warming tray placed in front of an electric space heater. Final temperatures were measured by a thermometer placed within the ear. Mice were then treated in the following manner. Xylocaine ointment (5% Lidocaine in a petroleum-based jelly, Astra Pharmaceutical Products, Inc., Westborough, MA 01581) was applied to the dorsal su**ce of a warm carcass with a wooden tongue depressor such that the carcass was completely covered from the head to the base of the tail in a thin coating of ointment. The carcass was then returned to the warming tray, ventral su**ce down, until body temperature returned to approximately 35 °C. This allowed the Xylocaine ointment to melt slightly, ensuring complete coverage of the dorsal su**ce. Care was taken in the preparation of target mice to ensure that a similar amount of Xylocaine jelly was applied to each carcass so as to limit variance in the amount of Xylocaine jelly available for delivery to a subject's mouth during a strike. Additional mice were treated in the same manner save that Vaseline ointment was used in the place of Xylocaine. Untreated mice were thawed and warmed as above but were not coated with either Vaseline or Xylocaine ointment.

In Conditions 1 through 3, mouse carcasses were suspended via long forceps approximately 10 cm in front of and slightly elevated above the subject's head (this is the same procedure that is used for presenting mice during routine feeding). Carcasses were presented with the dorsal su**ce facing the snake such that the roof of the snake's mouth would contact the part of the carcass containing the experimental ointment during the strike. All snakes oriented towards the target mice, struck within 3 to 5 s, and immediately released the prey. The mouse was then immediately withdrawn. In Condition 4, the mouse was held approximately 20 cm from the snake's head for 3 s and then removed before the snake could strike. At no time during the presentation of the mouse, in any of the experimental conditions, was it allowed to touch the floor or sides of the subjects' cages.

Immediately following removal of the target mouse, the observer began recording the number of tongue flicks emitted per minute for 30 min. Presentation of experimental conditions was randomized, with snakes tested once per week over a 4-week period. Baseline RTF and post-mouse-presentation RTF for each condition were analyzed by separate repeated measures ANOVAs. Post-hoc comparisons between conditions were performed using protected f tests.

Results

Initial RTF per minute for the four conditions were low (X = 1.04, 0.86, 0.98, 0.81, respectively) and did not differ significantly, indicating that snakes were quiescent prior to introduction of prey, F(3, 33) = 0.30, p > 0.05.

Figure 1 presents the mean RTF for the four conditions, and ANOVA revealed a significant difference between these values, F(3, 33) = 40.43, p < 0.0001, with an effect size (η^sup 2^) of .77 and an observed power (γ) of .99. Post-hoc comparisons indicated that the Vaseline condition (X = 44.2, SEM = 2.5) did not differ significantly from the untreated control condition (X= 45.8, SEM = 4.3), f(11) = 0.41, p > 0.05. However, the Xylocaine condition (X = 34.5, SEM = 3.7) was significantly lower than both the Vaseline and control conditions, f(11) = 4.69, f(11) = 2.36, p < 0.05, respectively. Although the mean RTF for the Xylocaine condition was significantly lower than the means of the Vaseline and control conditions, it was significantly higher than the mean of the no-strike condition (X = 8.9, SEM = 2.6), f(11) = 7.32, p < 0.05.

Specimens of four taxa of rattlesnakes were used in this study to assess the generality of the effects of VNS disruption. All taxa responded in an identical fashion to the experimental conditions. There was no significant main effect of taxa nor any significant interaction involving taxa. Accordingly, this factor will not be mentioned further.

Discussion

Striking a Xylocaine-coated prey item produced a significantly lower mean RTF than did striking a Vaseline-coated prey item or an untreated prey item. A plausible explanation for this finding is that the magnitude of SICS was reduced by the topical anesthetic's effects upon the vomeronasal sensory epithelium. This suggests that the magnitude of poststrike RTF is partially dependent upon the ability of the VNS to respond to stimulation subsequent to the strike. Such findings replicate those reported by Alving and Kardong (1996) who found that snakes with bilateral vomeronasal nerve transections exhibited a significantly lower RTF after striking rodents compared to intact controls. At first, such a view appears to run counter to those studies reporting that the magnitude of SICS is independent of the presence or absence of chemical cues in the poststrike environment (Chiszar et al., 1977; Golan et al., 1982). However, it is possible that high poststrike RTF is not dependent upon the presence of specific chemical cues that may be considered relevant to prédation (e.g., integumentary compounds or alarm pheromones from envenomed prey), but rather that it is dependent upon the VNS being fully functional and capable of receiving stimulation.

Animals in the Xylocaine condition produced RTF values that, while lower than the Vaseline and control conditions, were still significantly higher than those produced during the no-strike trials. This suggests that a portion of the poststrike elevation in RTF is not dependent upon the functional state of the VNS, but is a behavioral response activated by striking that is independent of VNS stimulation. Again, these results replicate those reported by Alving and Kardong (1996) where snakes with vomeronasal nerve transections exhibited poststrike RTF values that were significantly higher than snakes that were presented with rodent prey but prevented from striking. Our data suggest that 71% of the magnitude of SICS is dependent upon delivery of a successful predatory strike, whereas 29% is dependent upon having a patent VNS.

An alternative explanation for the reduction in poststrike RTF exhibited by snakes in the Xylocaine condition is that the anesthetic properties of the Xylocaine ointment impaired the motor systems mediating lingual movement such that normal high poststrike RTF could not be achieved in these animals. Examination of RTF data for the Xylocaine condition showed, however, that all snakes did at times achieve an RTF comparable to those reached during the Vaseline trials. This "spiking" in RTF occurred episodically during the 30-min test period, indicating that snakes in the Xylocaine condition were capable of exhibiting a normal poststrike RTF but did not maintain these rates consistently during the observation period. Thus, we find indirect support for the conclusion that the reduced poststrike RTF seen in the Xylocaine condition was not caused by muscular impairment, but derived from sensory blockade. This conclusion is consistent with the results reported by Andren (1982) where application of Xylocaine tape to areas of the mouth, other than at the entrance to the vomeronasal organs, in adders (Vipera berus) had no effect on reproductive behaviors dependent upon tongue flicking such as the detection of the s-e-x of a conspecific. Furthermore, in human applications Xylocaine disrupts sensory processes, but not motor processes (AIi, Laundl, Wallace, Shaw, Decarle, & Cook, 1994; Mansson & Sandberg, 1974), suggesting again that the lower RTF in the Xylocaine condition was not a result of lingual impairment.

The purpose of Experiment 2 was to investigate the effects of Xylocaine on the performance of a task believed to be dependent upon the VNS, namely, the location of an envenomed prey item (Duvall, Scudder, & Chiszar, 1980).

：芭芭拉

☆Experiment 2

Method

Subjects

Subjects were 10 rattlesnakes (4 Crotalus oreganus oreganus, 1 Crotalus oreganus lutosus, and 5 Crotalus viridis viridis). They were captured as adults and had been in captivity for at least 3 years. These snakes were housed individually in glass terraria (50 x 27.5 x 30 cm) under conditions identical to those described in Experiment 1.

Procedure

All snakes were observed in their home cages. A trial began with the subject being presented with either a target mouse carcass coated in Xylocaine (experimental condition) or a target mouse carcass coated in Vaseline (control condition). Targets were suspended by forceps approximately 10 cm in front of and slightly above the snakes' heads such that the dorsal su**ce of the target was facing the snakes. Within 1 to 3 s, snakes oriented to these carcasses, struck, and then released them. Following the strike, a second envenomed mouse carcass (the "prey item," see below) was placed on the floor at the far end of the cage from the snake. At no time during the presentation of the target mouse carcass was it allowed to touch any su**ce of the snake's cage. Each snake received both conditions 1 week apart, with half of the snakes getting the Xylocaine treatment before the Vaseline treatment, while the remaining snakes received the reverse order.

Mouse carcasses used for target mice and for prey items were previously euthanized by cervical dislocation and stored frozen at 0 °C. Mice were prepared for use following the procedures of Experiment 1.

Rattlesnakes can discriminate between rodents based upon chemical cues obtained during the strike (Melcer & Chiszar, 1-9-8-9). To control for the possibility that odor or some other factor related to the Xylocaine or Vaseline ointment may have an effect on the snake's ability to locate the envenomated rodent; the coated target mouse was not used as a prey item. Using the methods of Duvall et al. (1980), a second mouse carcass, of similar size and weight as the target mouse, was envenomed by a conspecific rattlesnake at approximately the same time as the target mouse was struck. This mouse (hereafter referred to as the prey item) was then placed within the snake's cage and recording of the following dependent variables commenced: (a) latency to locate (i.e., touch the lips to the) prey item, (b) the number of tongue flicks emitted while the snake searched for the prey item and, (c) latency to commence swallowing the prey item recorded from when it was first located. Each snake was observed for 20 min or until the snake had completely ingested the envenomed mouse carcass. Dependent variables were analyzed by separate repeated measures ANOVAs.

Results

Figure 2 displays the mean RTF per minute emitted while the subjects were searching for prey items. ANOVA revealed that the mean RTF during the Vaseline condition was significantly higher than the mean RTF during the Xylocaine condition, replicating the findings reported in Experiment 1, F(1, 9) = 10.58, p < 0.01, η^sup 2^ = .34, γ = .84. As RTF was calculated by dividing the total number of tongue flicks emitted while searching for the prey item by the latency (in min) to locate the prey item, this effect could be due to differences in latency scores (i.e., very short latencies produce artificially high RTF scores). Therefore, comparison of the number of tongue flicks emitted during the last minute prior to locating the prey item was made. ANOVA revealed that the number of tongue flicks emitted during this time period in the Xylocaine condition (X = 20.3, SEM = 2.7) was significantly less than those in the Vaseline condition (X= 38.5, SEM = 5.3), F(1, 9) = 11.34, p < 0.01, η^sup 2^ = .56, γ = .86, indicating that the effect illustrated in Figure 2 was not an artifact of latency differences. All 10 snakes located the prey item within the 20-min test period regardless of experimental condition. However, Figure 3 illustrates that snakes with anesthetized VNS took significantly longer than intact snakes to locate the prey item (X= 412.4, SEM = 74.4 and X= 197.3, SEM = 37.5), F(1, 9) = 9.42, p < 0.01, η^sup 2^ = .51, γ =.78.

All snakes swallowed prey during the Vaseline condition (100%). All but one snake swallowed the prey item during the Xylocaine condition (90%). This subject's latency to commence swallowing score was dropped from subsequent analysis. Figure 4 illustrates that snakes with intact VNS_attempted to consume the prey item in less than 30 s after locating it (X= 28.0, SEM = 9.9). However, snakes with anesthetized VNS took a significantly greater amount of time to investigate the prey item prior to attempting to consume it (X= 82.8, SEM = 21.3), F(1, 8) = 6.29, p < 0.05, η^sup 2^ = .44, γ = .60. Once head-first ingestion began, it continued to completion_and the two groups did not differ on the latency to complete swallowing (X= 690, SEM = 116.69 and X = 610, SEM = 96.00), F(1, 5) = .39, p > 0.05, η^sup 2^ = .07, γ= .08, indicating no impairment in buccal motor processes caused by Xylocaine.

Discussion

Rattlesnakes can locate envenomated prey in the absence of a chemical trail if the available search area is relatively confined (Dullemeijer, 1961). Data presented here provide additional verification of this ability. Snakes with intact VNS located and began ingesting prey within 4 min of striking. As the prey items were placed in the cage without touching any part of the cage other than the floor directly underneath there were no nonvolatile prey-derived chemical trails available to the snakes to guide their searching efforts.

It is possible that visual cues expedited location of the prey item. As the length of the home cage determined the maximum distance from the snake to a prey item (approximately 50 cm), all prey items were within visual range during testing. However, although initial detection of prey and delivery of predatory strikes depend upon information mediated through the visual and infrared systems (De Cock Bunning, 1983; Dullemeijer, 1961; Kardong, 1986, 1992; Newman & Hartline, 1981, 1982), it has been shown that rattlesnakes cease attending to visual/thermal cues after striking, focusing attention upon chemical information (Chiszar et al., 1981; Dullemeijer, 1961). This shift in attention from visual/thermal cues to chemosensory information after striking was evident in the rattlesnake's behavior during the current study. When the prey item was being placed within the subject's cage, the head of the snake would orient to the carcass and track its movement from the lid of the cage to where the carcass was finally placed. One would expect that the snake would focus its subsequent SICS efforts in the direction of the mouse carcass visible upon the floor of the cage. However, when the subject began SICS, tongue flicks were invariably directed towards the area where the strike occurred. Only after thoroughly investigating the cage around the strike area did the snake widen its search. The head of the searching rattlesnake often came within inches of the envenomed prey item, but the snake failed to orient to the prey until mak-ing direct lingual contact. This suggests that visual cues are not attended to during this phase of predatory behavior and that lingual contact is crucial to the acquisition of chemical information relevant to locating prey (see Kubie & Halpern, 1979, for a similar finding in T. sirtalis).

Snakes in the Xylocaine condition did not display any marked differences in the outward appearance of their search behavior compared to controls. These snakes commenced and continued SICS behavior after striking in a manner indistinguishable from snakes with patent VNS. However, Xylocaine significantly increased the amount of time necessary for these snakes to locate envenomed prey. A plausible explanation for this finding is that, if the ability of the VNS to detect chemical cues was impaired by the anesthetic, subjects were incapable of using this system as a means of eliminating previously searched areas that did not yield the prey item. Indeed, although the pattern of searching was similar to controls, snakes with anesthetized VNS were seen returning to areas of the cage that had been previously searched at a greater rate than controls, suggesting that snakes with intact VNS were better able to narrow their search efforts to areas of the cage that had not yet been investigated.

Application of Xylocaine also contributed to an increased latency to commence swallowing prey once located. One possible explanation for this is that snakes in the Xylocaine condition had difficulty locating the head region of the prey item. Upon locating envenomated prey, intact snakes direct tongue flicks in a sweeping motion from the nasal-oral region to the anal-genital region (Chiszar & Radcliffe, 1976; Duvall et al., 1980) in an effort to locate the head of the carcass, as swallowing of adult rodent prey is facilitated by doing so head first (Chiszar & Radcliffe, 1976). The difficulty encountered by snakes in the Xylocaine condition in locating the head region of the prey item was evident in that 4 snakes in this condition (40%) first attempted to swallow the prey item by grasping the thorax or posterior end of the carcass.

A related explanation for the delay in ingestive behavior is that in the Xylocaine condition, snakes were unable to detect the appropriate cues that normally initiate swallowing behavior. For example, during the Xylocaine condition snake C.o.oreganus #3 contacted the envenomated mouse in 427 seconds. However, after mak-ing a few tongue flicks it made no further attempt to investigate the carcass and resumed searching. After another 166 seconds it recontacted the carcass and after 30 s of investigative tongue flicking it began swallowing the mouse posterior first. Similar behavior was observed in C.o.oreganus #2. This snake contacted the prey item in 294 s, but then continued to search the adjacent area without attempting to consume the carcass. Upon "relocating" the prey item 182 s later it attempted to bite the head of the carcass, but then spit it out and resumed searching. This pattern was repeated twice before the snake coiled up on top of the carcass and remained stationary for the duration of the test period. These data suggest that application of Xylocaine disrupted the sensory system(s) necessary for the detection of appropriate chemical cues that signal the reinforcing value of a prey stimulus and/or release ingestive behavior. A similar response to VNS deprivation has been demonstrated in T. sirtalis. Garter snakes with complete transection of the vomeronasal nerves gradually ceased predatory attack responses to earthworm bits postO p e r atively and spit out the earthworm bits when they did attack (Kubie & Halpern, 1979).

：芭芭拉

General Discussion

Previous attempts to examine the role of the VNS in SICS behavior in rattlesnakes have relied upon surgical disruption of the VNS prior to presentation of prey. These methods have produced a serious decline in predatory strike responses (Alving & Kardong, 1996; Graves & Duvall, 1985) suggesting that the VNS is involved, to some degree, in the mediation of strike behavior. Our data were gathered utilizing a new method for disrupting the VNS that did not interfere with striking prey and did not affect VNS chemoreception until after the strike. As such, our investigation was focused solely upon the effects of vomeronasal deprivation on poststrike behavior.

Although snakes in the Xylocaine condition in both Experiments 1 and 2 showed a significant decline in poststrike RTF relative to the Vaseline condition, these values were still significantly higher than RTF values for the no-strike and baseline conditions. Examination of the ratio of mean RTF in the Xylocaine condition to mean RTF in the Vaseline and control conditions combined in Experiment 1 suggests that 71 % of the increase in RTF seen in intact rattlesnakes is not dependent upon VNS stimulation poststrike. This rapid tongue flicking in the absence of stimulation by environmental cues lends support to the conclusion that a major portion of SICS is a centrally coordinated motor pattern that is released by the strike (Barlow, 1977; Chiszar, 1986; Chiszar et al., 1977; Lorenz, 1981; Tinbergen, 1951). The remaining 29% that is necessary to raise RTF to normal levels may reflect a component of SICS that is dependent upon a patent VNS.

Although the methods presented here offered many advantages over previous surgical techniques, there were several drawbacks that should be addressed. The extent to which sensory transduction in the vomeronasal organs was eliminated was not confirmed using electrophysiological or other techniques. Nor was it possible to control the amount of Xylocaine delivered to the vomeronasal ducts. Therefore, an empirical means of verifying the elimination of vomeronasal chemoreception by Xylocaine is necessary. One option would be to record multiunit neural responses in the vomeronasal nerves during presentation of chemical stimuli to the vomeronasal organs before and after application of Xylocaine to the vomeronasal ducts. A similar procedure has been used with success to investigate trigeminal responses to thermal stimulation before and after anesthetization of the facial pit organs in rattlesnakes (Dickman, Colton, Chiszar, & Colton, 1987). Such a study would provide the evidence necessary to reach a definitive conclusion concerning the transmissionblocking efficacy of prey delivered VNS anesthetization.

However, there is a good deal of indirect evidence indicating the effectiveness of Xylocaine application in disrupting the VNS. The effects of VNS anesthetization on poststrike RTF reported here are comparable to those reported by Alving and Kardong (1996) in a nerve transection experiment. Having established that Xylocaine applied to the vomeronasal ducts prevented the occurrence of normal reproductive behavior in male adders (V. berus), Andren (1982) demonstrated that Xylocaine applied to areas of the mouth other than the vomeronasal ducts did not produce any changes in behavior, indicating that Xylocaine was an effective means of blocking VNS chemoreception without interfering with motor processes. The results of Experiment 2 provide further convergent evidence that Xylocaine application interferes with behaviors believed to be dependent upon the VNS and thus is an effective means of investigating the role of the VNS in poststrike predatory behaviors.

One could argue that the effects of Xylocaine reported here did not arise from blocking sensory transduction in the VNS but were the consequence of (a) interference with motor control of tongue flicking, (b) change in motivational and/or emotional state, or (c) interference with movement of prey-derived chemicals along the oral mucosa because of viscous material adhering to the roof of the mouth. However, neither tongue flicking nor any aspect of ingestion was disrupted in the Vaseline condition. Therefore, the last of these possibilities can be eliminated. All but one snake ingested the prey item in the Xylocaine condition and swallowing (once a head-first orientation was achieved) appeared normal. There was no difference between snakes in the Xylocaine condition and control conditions in the time taken to swallow the mouse carcass once swallowing commenced, nor was there any difference in locomotion, nor in the pattern of search behavior. Snakes in the Xylocaine condition also did not display any signs characteristic of defensive behavior. Therefore, it is unlikely that Xylocaine interfered with motivational and/or emotional systems. There is the possibility that Xylocaine may have interfered with the mechanics of tongue flicking. However, the topography of tongue flicks appeared indistinguishable in the Xylocaine, Vaseline, and control conditions. Furthermore, all snakes exhibited maximal RTFs episodically during the Xylocaine condition, indicating that Xylocaine treatment did not eliminate the ability to produce high RTFs, but may have impaired the ability to maintain a high RTF. Such a conclusion implies that there is a rate at which chemical information must be delivered to the VNS in order for the system to function properly. However, snakes have been shown to identify stimuli based upon only one or two tongue flicks (Shine, Phillips, Waye, LeMaster, & Mason, 2003). Therefore, the RTFs seen in the Xylocaine condition, although lower than those seen in the Vaseline and control conditions, would have been sufficient for identification of prey cues if the VNS was O p e r ational. For these reasons, we suggest that the deficits observed in Experiments 1 and 2 were based on blockade of the VNS rather than on motor disturbance and we look forward to further confirmation of this hypothesis in future studies.

[Reference]
References
ALI, C. N., LAUNDL, T. M., WALLACE, K. L, SHAW, D. W., DECARLE, D. J., & COOK, I. J. (1994). Influence of mucosal receptors on deglutitive regulation of pharyngeal and upper esophageal sphincter function. American Journal of Physiology-Gastrointestinal and Liver Physiology, 267, 4, 644-649.
ALVING, W. R., & KARDONG, K. V. (1996). The role of the vomeronasal organ in rattlesnake (Crotalus viridis oreganus) predatory behavior. Brain Behavior & Evolution, 48, 165-172.
ANDREN, C. (1982). The role of the vomeronasal organs in the reproductive behavior of the adder Vipera berus. Cope/a, 1, 148-157.
BARLOW, G. W. (1977). Modal action patterns. In J. Sebeok (Ed.), How animals communicate (pp. 98-134). Bloomington: Indiana University Press.
BURGHARDT, G. M. (1970). Chemical perception in reptiles. In J. W. Johnston, D. G. Moulton, & A. Turk (Eds.), Advances in chemoreception. I. Communication by chemical signals (pp. 241-308). New York: AppletonCentury-Crofts.
BURGHARDT, G. M. (1980). Behavioral and stimulus correlates of vomeronasal functioning in reptiles: Feeding, grouping, s-e-x, and tongue use. In D. Duvall, D. Muller-Schwarz, & R. M. Silverstein (Eds.), Chemical signals of vertebrates and aquatic invertebrates (pp. 275-301). New York: Plenum Press.
BURGHARDT, G. M. (1983). Aggregation and species discrimination in newborn snakes. Zeitschrift fur Tierpsychology, 61, 89-101.
CHISZAR, D. (1986). Motor patterns dedicated to sensory functions. In D. Duvall, D. Muller-Schwarz, & R. M. Silverstein (Eds.), Chemical signals in vertebrates 4 (pp. 37-44). New York: Plenum Press.
CHISZAR, D., & RADCLIFFE, C. W. (1976). Rate of tongue flicking by rattlesnakes during successive stages of feeding on rodent prey. Bulletin of the Psychonomic Society, 7, 485-486.
CHISZAR, D., RADCLIFFE, C. W., & SCUDDER, K. M. (1977). Analysis of the behavioral sequence emitted by rattlesnakes during feeding episodes. Behavioral Biology, 21, 418-425.
CHISZAR, D., RADCLIFFE, C. W., & SMITH, H. M. (1978). Chemosensory searching for wounded prey by rattlesnakes is released by striking: A replication report. Herpetological Review, 9, 54-56
CHISZAR, D., RADCLIFFE, C. W., O1CONNELL, B., & SMITH, H. M. (1981 ). Strikeinduced chemosensory searching in rattlesnakes (C. viridis) as a function of disturbance prior to presentation of rodent prey. The Psychological Record, 31,57-62.
DE COCK BUNNING, T. (1983). Thermal sensitivity as a specialization for prey capture and feeding in snakes. American Zoology, 23, 363-375.
DICKMAN, D. J., COLTON, J. S., CHISZAR, D., & COLTON, C. A. (1987). Trigeminal responses to thermal stimulation of the oral cavity in rattlesnakes (Crotalus viridis) before and after bilateral anesthetization of the facial pit organs. Brain Research, 400, 365-370.
DULLEMEIJER, P. (1961). Some remarks on the feeding behavior of rattlesnakes. Koninklijke Nederlandische Akademie von Wetenschappen, Series C, 64, 383-396.
DUVALL, D., SCUDDER, K. M., & CHISZAR, D. (1980). Rattlesnake predatory behavior: Mediation of prey discrimination and release of swallowing by odors associated with envenomated mice. Animal Behaviour, 28, 674-683.
ESTEP, K., POOLE, T., RADCLIFFE, C. W., O'CONNELL, B., & CHISZAR, D. (1981). Distance traveled by mice (Mus musculus) after envenomation by prairie rattlesnakes (Crotalus viridis viridis). Bulletin Psychonomic Society, 18, 108-110.
GANS, C. (1966). The biting behavior of solenoglyph snakes-Its bearing on the pattern of envenomation. Proceedings of the International Symposium on Venomous Animals. Sao Paulo, Brazil: lnstituto Butantan.
GOLAN, L., RADCLIFFE, C., MILLER, T., O'CONNELL, B., & CHISZAR, D. (1982). Trailing behavior in prairie rattlesnakes (Crotalus viridis). Journal of Herpetology, 16, 287-293.
GRAVES, B. M., & DUVALL, D. (1985). Avomic prairie rattlesnakes (Crotalus viridis) fail to attack rodent prey. Zeitschrift fur Tierpsychology, 67, 161-166.
GRAVES, B. M., & HALPERN, M. (1-9-8-9). Chemical access to the vomeronasal organs of the lizard Chalcides ocellsatus. Journal of Experimental Zoology, 249, 150-157.
GRAVES, B. M., & HALPERN, M. (1990). Roles of vomeronasal organ chemoreception in tongue flicking, exploratory and feeding behavior of the lizard Chalcides ocellsatus. Animal Behavior, 39, 692-698.
HALPERN, M. (1992). Nasal chemical senses in reptiles. In C. Cans & D. Crews (Eds.), Hormones, brain and behavior: Biology of the Reptilia Vol. 18 (pp. 423-523). Chicago: University of Chicago Press.
HALPERN, M., & FRUMIN, N. (1979). Roles of the vomeronasal and olfactory organs in prey attack and feeding in adult garter snakes. Physiology & Behavior, 22, 1183-1189.
HALPERN, M., & KUBIE, J. L. (1980). Chemical access to the vomeronasal organs of garter snakes. Physiology & Behavior, 24, 367-371.
HALPERN, M., & KUBIE, J. L. (1983). Snake tongue flicking behavior: Clues to vomeronasal system functions. In R. M. Silverstein & D. Muller-Schwarze (Eds.), Chemical signals in vertebrates 3 (pp. 45-72). New York: Plenum Publishing.
HALPERN, M., & KUBIE, J. L. (1984). The role of the ophidian vomeronasal system in species-typical behavior. Trends in Neuroscience, 7, 472-477.
KAHMANN, H. (1932). Sinnesphysiologische studien an reptilien: I. Experimentalle unterswuchungen uber das jakobsonche organ der eideschen und schlangen. Zool. Jhb. Zoological Physiology, 51, 173-238.
KAHMANN, H. (1934). Zur Chemorezeption der Schlangen (ein Nachtag). Zool. Anz., 107,249-263.
KARDONG, K. (1986). Predatory strike behavior of the rattlesnake (Crotalus v. oreganus). Journal of Comparative Psychology, 100, 304-314.
KARDONG, K. V. (1992). Proximate factors affecting guidance of the rattlesnake strike. Zool. Jb. Anatated., 122, 233-244.
KUBIE, J. L., & HALPERN, M., (1979). Chemical senses involved in garter snake prey trailing. Journal of Comparative Physiological Psychology, 93, 648-667.
LORENZ, K. Z. (1981). The foundations of ethology. New York: Springer-Verlag.
MANSSON, I., & SANDBERG, N. (1974). Effects of su**ce anesthesia on deglutition in man. Laryngoscope, 84, 3, 427-437
MELCER, T., & CHISZAR, D. (1-9-8-9) Striking prey creates a specific chemical search image in rattlesnakes. Animal Behavior, 37, 477-486.
NEWMAN, E., & HARTLINE, P. (1981). Integration of visual and thermal information in bimodal neurons of the rattlesnake optic tectum. Science, 213,789-791.
NEWMAN, E., & HARTLINE, P. (1982). The infrared "vision" of snakes. Scientific American, 246, 116-127.
O'CONNELL, B. R., GREENLEE, J., BACON, J., & CHISZAR, D. (1982). Strikeinduced chemosensory searching in old world vipers and new world pit vipers at San Diego Zoo. Zoo Biology, 1, 287-294
RADCLIFFE, C. W., CHISZAR, D., & O'CONNELL, B. (1980). Effects of prey size on post strike behavior in rattlesnakes. Bulletin of the Psychonomic Society, 16, 449-450.
SCUDDER, K. M., CHISZAR, D., & SMITH, H. M. (1983). Effect of environmental odors on strike induced chemosensory searching by rattlesnakes. Copeia, 1983 519-522
SHINE, R., PHILLIPS, B., WAYE, H., LEMASTER, M., & MASON, R. T. (2003). Chemosensory cues allow courting male garter snakes to assess body length and body condition of potential mates. Behavioral Ecology and Sociobiology, 54, 162-166
TINBERGEN, N. (1951). The study of instinct. Oxford: Oxford University Press.
YOUNG, B. A. (1993). Evaluating hypotheses for the transfer of stimulus particles to Jacobson's organ in snakes. Brain Behavior & Evolution, 41, 203-209.

[Author Affiliation]
C. PATRICK STARK
Western State College
DAVID CHISZAR and HOBART M. SMITH
University of Colorado, Boulder

[Author Affiliation]
We thank Katherine Stiles for her assistance in data collection. Correspondence concerning this article should be addressed to Charles Patrick Stark, Department of Psychology, Western State College, 207a Kelley Hall, Gunnison, CO 81231. (E-mail: pstark@western.edu).