SEARCH

SEARCH BY CITATION

Keywords:

  • Ambystoma tigrinum;
  • Ambystoma californiense;
  • climate;
  • dispersal;
  • hybrid vigour;
  • hybridization;
  • migration

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

1. Anthropogenic environmental changes have the capacity to disrupt natural population dynamics. For amphibians with complex life cycles, it is important to understand how environmental perturbations interact with variation in larval period duration to affect the timing of dispersal and the abiotic conditions under which terrestrial movements occur.

2. Widespread hybridization between California tiger salamanders and introduced tiger salamanders has created a situation in which length of larval development and metamorphic timing varies because of extrinsic variation in abiotic factors such as temperature, and also as a result of individual-specific variation in genomic admixture.

3. We examined how line-cross type (e.g. F1, F2, backcrosses, etc.), morphology and temperature interact to affect the performance of emigrating immature salamanders. We performed endurance trials on a mechanized treadmill to simulate dispersal following metamorphosis. Our study provides insights into the interaction of environmental change and genomic composition on locomotor performance and the spread of the hybrid swarm.

4. We found that temperature had a significant positive impact on endurance, potentially making it an important factor driving dispersal distances in the wild. Line-cross type also affected performance, with F1 hybrids demonstrating the greatest movement capacity. However, we did not observe an interaction between line-cross type and temperature.

5. These results demonstrate that an increase in ambient temperature may enhance the dispersal ability of hybrid individuals and accelerate the spread of the hybrid swarm. An increase in the rate of introgression of non-native tiger salamander alleles into formerly pure native habitats places enhanced importance on the proper management of populations near the boundary of the hybrid swarm and emphasizes the need for management actions that prevent the intentional or accidental translocation of non-native species.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Amphibians with a biphasic life cycle face a basic ecological tradeoff. Prolonged larval periods can lead to larger sizes at metamorphosis, increased survival in the post-metamorphic terrestrial phase, a shorter time to sexual maturity and overall greater fitness (Semlitsch, Scott & Pechmann 1988; Scott 1994; Morey & Reznick 2001). However, long larval periods can also lead to increased risk of desiccation-associated larval mortality as vernal pools dry, and post-metamorphic mortality if metamorphosis occurs during hot, dry conditions when overland movements are difficult or impossible. Interacting with these life history tradeoffs are the landscapes and climatic conditions where species and populations have evolved. For example, the southeastern US tends to have abundant summer rainstorms that promote successful terrestrial migrations of vulnerable young metamorphs, whereas most of California has a Mediterranean climate with virtually none of the summer rains that aid dispersal elsewhere. Predictably, the solutions to these life history tradeoffs faced by many metamorphosing amphibians vary geographically and taxonomically in keeping with local rainfall and climate patterns.

In response to variable pond hydroperiods and rainfall patterns, closely related species of amphibians may display profound variation in the degree to which metamorphic timing is a plastic or relatively fixed life history feature. Within our tiger salamander study system, species display tremendous life history variation in metamorphic timing. The barred tiger salamander (BTS; Ambystoma tigrinum mavortium; Fig. 1) has the ability to coordinate metamorphosis and subsequent emigration with biotic factors (e.g. prey abundance, competitor density) and aspects of the abiotic environment (e.g. pond hydroperiod, temperature, humidity, rainfall) to optimize fitness (Wilbur & Collins 1973). As the BTS is native to the US Great Plains and breeds in both semi-permanent and fishless permanent ponds (Petranka 1998), plasticity in larval period often leads to a delay in metamorphosis and exploitation of aquatic habitats with little risk of catastrophic early pond drying. Summer rains contribute additional yearly variation in the hydroperiods of native BTS breeding habitats and provide a moist terrestrial environment for emigration of newly metamorphosed juveniles periodically throughout the summer and fall.

image

Figure 1.  Left: Adult hybrid tiger salamander (Ambystoma tigrinum × californiense) from Monterey County, California; Right: Adult California tiger salamander (A. californiense) from Solano County, California; photo by J. R. Johnson.

Download figure to PowerPoint

In contrast, the California tiger salamander (CTS; Ambystoma californiense; Fig. 1) exhibits remarkably little plasticity in the timing of metamorphosis. CTS evolved in the Mediterranean climate of California’s Central Valley where natural aquatic habitats (shallow vernal pools) have consistently short hydroperiods and rainfall is essentially absent following the winter breeding season (Trenham et al. 2000). Peak metamorphosis occurs from late June to early July (Shaffer & Trenham 2005) as ponds dry, at which time CTS juveniles must make nighttime, overland movements to upland retreats (usually rodent burrows or soil cracks) without the benefit of rainfall (Loredo, Vanvuren & Morrison 1996). The choice of burrow may be extremely important to emigrating salamanders because drift-fence studies suggest that the initially chosen retreat is continuously occupied for the next several months (Searcy & Shaffer 2008). Individuals that fail to find shelter will perish if exposed to daytime temperatures and dryness (Loredo, Vanvuren & Morrison 1996), and desiccated juveniles are periodically found on the surface (C. Searcy & B. Shaffer, unpublished data). Interestingly, even when pond hydroperiods are unusually long, CTS tend to metamorphose at small sizes and forego the potential for increased larval growth and larger metamorphic size both in the laboratory (J. Johnson, B. Fitzpatrick & B. Shaffer unpublished data) and in the field (C. Searcy & B. Shaffer, unpublished data). In natural California landscapes, the benefits of plasticity in metamorphic timing during rare wet years apparently do not outweigh the importance of early metamorphosis or the costs associated with the maintenance of phenotypic plasticity (Relyea 2002).

Hybridization between CTS and BTS has resulted from the intentional release of BTS from the southwestern US in the 1950s into the Salinas Valley (Riley et al. 2003; Fig. 2). Previous studies have found both hybrid vigour (Fitzpatrick & Shaffer 2007a) and hybrid dysfunction (Fitzpatrick 2008) in the larval stage and have demonstrated that larval CTS are at a competitive disadvantage in experimental mesocosms (Ryan, Johnson & Fitzpatrick 2009). While this hybrid system has been well studied with respect to the geographical patterns of introgression and larval fitness (Fitzpatrick & Shaffer 2004, 2007a, b; Fitzpatrick et al. 2009, 2010), the consequences of admixture have not been investigated for any post-metamorphic fitness components in the lab or the field.

image

Figure 2.  Range of California tiger salamanders outlined by grey shading in California’s Central Valley. The inset displays the Western United States and the black box denoted by (a) represents the likely source of the introduced barred tiger salamanders from New Mexico and Texas. The oval denoted by (b) represents the main extent of the hybrid swarm in the Salinas Valley, Monterey County, California.

Download figure to PowerPoint

It is well established that the novel genetic variation produced through hybridization generates phenotypic variation in the timing of metamorphosis that greatly exceeds that found in native CTS populations under laboratory and mesocosm conditions (Ryan, Johnson & Fitzpatrick 2009, J.R. Johnson, B.M. Fitzpatrick & H.B. Shaffer unpublished).

What remain unclear are the consequences of increased plasticity and longer larval period for post-metamorphic salamanders. One possibility is that delayed metamorphosis may negatively affect the fitness of late-metamorphosing juvenile salamanders given the extreme environmental conditions experienced during late emigration, when daytime temperatures can exceed 40 °C and rain is non-existent. However, elevated ambient temperatures have been shown to improve the locomotor performance of ectotherms (Bennett 1990) including tiger salamanders (Else & Bennett 1987), and extreme daytime highs may have a negligible effect on salamander fitness because movements occur at night when temperatures are consistently much cooler. Therefore, hybrids that delay metamorphosis might experience an increase in survival if increased locomotor capacity translates into a greater chance of finding an appropriate underground refuge during predawn field conditions.

In this study, we examined locomotor performance of both pure and early-generation hybrid juvenile salamanders to identify differences that could impact fitness based on line-cross type (e.g. F1, F2, backcrosses, etc.) or phenotype (e.g. mass). Variation that is attributable to line-cross type is by definition genetically controlled and provides insights into the genetic basis of differences among species. We replicated our experiments at two temperatures to explore the physiology of temperature-dependent locomotor capacity and potential genotype-by-temperature interactions. Our goal was to determine the contributions of temperature and size on locomotor performance of recently metamorphosed salamanders to shed light on how environmental and morphological variation might impact fitness of recently metamorphosed tiger salamanders in the wild. Importantly, an improved understanding of locomotor capacity will help predict the dynamics and spread of the hybrid swarm that currently threatens the remaining populations of pure CTS (Fitzpatrick & Shaffer 2007a,b). Given the endangered status of the California Tiger Salamander and the importance of hybridization as a key threat to the species (US Fish and Wildlife Service 2004), understanding the fitness consequences of hybridization is also a critical element of endangered species management in this system.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Study animals

J.R. Johnson, B.M. Fitzpatrick & H.B. Shaffer (unpublished) generated 1st and 2nd generation hybrids [i.e. F1, F2, and backcrosses (bcCTS & bcBTS)] between Ambystoma californiense and A. tigrinum mavortium, as well as control parental crosses for both species. Larvae were reared in individual containers from hatching through metamorphosis and a random selection of metamorphs was maintained for this study. Line-cross sample sizes were: CTS (N = 6), BTS (N = 5), bcCTS (N = 10), bcBTS (N = 9), F1 (N = 5) and F2 (N = 6). Following metamorphosis, salamanders were maintained individually in the laboratory at 20 °C on a 12 h light to 12 h dark photoperiod prior to experiments.

Salamanders from each cross type were run on a treadmill at each of two different temperatures and were randomly assigned to one of two regimes – hot first or cold first – to account for any potential effect of trial order. All animals were fed 3 days before initiation of trials and all members of a trial regime (hot or cold first) were kept for 3 days at either 15 °C or 25 °C. These trial temperatures were chosen to represent the average field temperatures observed during times of juvenile salamander dispersal as measured via iButtons (Maxim Integrated Products, Sunnyvale, CA, USA) set to record the temperature every 30 min in situ at a nearby breeding site and averaged over 3 years (see Fig. S1 in Supporting Information). Average nighttime ambient temperatures during late June–early July (time period of peak metamorphosis; Shaffer & Trenham 2005) are about 15–20 °C (Fig. S1), although evaporative cooling may produce lower salamander body temperatures than those recorded by the iButtons.

Performance trials

Following acclimation, salamanders were selected at random and were run individually to exhaustion on a mechanized treadmill. The treadmill was constructed within a 65 cm long × 15·3 cm wide × 22 cm deep clear plastic frame (Fig. S2). The treadmill speed was set to match the empirically determined maximum speed of each individual independently. To accomplish this, at the beginning of each trial, the treadmill speed was adjusted to match the apparent maximum initial walking speed of each individual salamander. This speed varied for individual animals, but remained constant during that individual’s trial. As salamanders became tired, they were maintained in constant maximum forward motion by tapping or lightly pinching the tail if needed. Tests for fatigue were performed every 3 min, or when salamanders refused to walk any further. Fatigue was determined by a righting response test – the inability of a salamander to right itself physically within 3 s after being placed on its back (Shaffer, Austin & Huey 1991; Austin & Shaffer 1992). Each trial was terminated at fatigue and each animal was tested only once at each temperature. After a trial was terminated, walking speed, total walking duration and mass were recorded. Distance travelled was calculated as the product of speed and duration walked.

Data analyses

Data analyses were conducted in R (R Core Development Team, 2009). The effects of temperature, line-cross type and mass were evaluated using maximum likelihood estimation of linear mixed-model regression with the R package ‘nlme.’ We used temperature, line-cross type, and mass as fixed effects and individual ID as a random effect in each model. Measures of performance (i.e. distance walked and speed walked) were log-transformed (ln[x + 1]) to improve data normality. We standardized mass across line-cross types by using the difference between each individual’s mass and its line-cross type mean value to remove the dramatic differences in mass among line-cross types (Fig. S3). The use of standardized values allows us to investigate the effect of mass on performance independent of line-cross type. We also parameterized models using untransformed and log-transformed mass, but do not report those results because the same best-fit models were selected in each case. We performed an analysis of variance on the best-fit models and evaluated line-cross-type contrasts with native CTS values as a reference. We were primarily interested in evaluating the hypothesis that hybrid cross types performed differently from native CTS individuals and hence we did not perform all pairwise contrasts among line-cross types.

For ‘speed walked’, we calculated Q10 (for rate-based responses) values, and for ‘distance walked’, we calculated R10 (for quantity-based responses) values, which represent the change in locomotor performance over a ten-degree change in temperature: Q10 = [R2/R1], where R is the rate of the response variable at temperatures 1 and 2, and R10 = [S2/S1], where S is the value of the response variable at temperatures 1 and 2 (Bennett 1984). The temperature coefficient (Q10) and the thermal ratio (R10) are analogous terms that are used to describe the thermal dependence of the response variables and allow qualitative comparison among studies investigating thermal responses.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Sample sizes, mean mass and performance trial results are presented in Table 1. As found in more formal quantitative genetic analyses, CTS are much smaller than BTS, and hybrid categories are intermediate following a roughly additive model (J.R. Johnson, B.M. Fitzpatrick & H.B. Shaffer unpublished). We observed a dramatic increase in locomotor performance of all line-cross types at 25 °C relative to 15 °C. Mean duration of trials (averaged across line-cross types) at 15 °C was 17 min vs. 102 min at 25 °C. This difference in trial duration translates into a difference of more than 1 km travelled or a 30% increase in walking speed on average (Fig. 3).

Table 1.   Means and standard deviations for each locomotor performance variable
Cross typeNMass (g)Distance (m)Speed (cm s−1)
15 °C25 °CR1015 °C25 °CQ10
MeanSDMeanSDMeanSDMeanSDMeanSD
CTS610·82·839·08·91082·11098·427·711·51·318·13·51·6
bcCTS1011·32·895·859·31507·41080·415·712·32·421·17·51·7
F1513·72·7261·8124·71778·2931·16·812·72·621·35·71·7
F2615·36·750·720·3647·1809·512·711·21·612·73·21·1
bcBTS920·77·7148·0115·31234·8726·18·311·82·216·94·81·4
BTS526·59·9160·8190·8682·2543·54·214·83·218·37·41·2
image

Figure 3.  Comparison of ln(x + 1)-transformed endurance times: (a) distance walked and (b) speed walked across all individuals at 15 and 25 °C. Solid black lines denote the median value for each cross type and the open vertical rectangles represent the quartiles. The dotted lines extend to the minimum and maximum values. Statistics are presented in Table 3.

Download figure to PowerPoint

Model selection revealed that while temperature had the greatest effect on performance, line-cross type and mass were also important (Table 2). Overall, the best-fit model for distance walked included temperature, line-cross type, mass and the interaction between line-cross type and mass. The best-fit model for speed walked included additive effects of temperature and line-cross type only (Table 2). The best three-factor model for speed walked was not a significant improvement over the model including only temperature and line-cross type (Table 2, Table S1). We found no evidence of a significant interaction between temperature and line-cross type, suggesting that all cross types displayed a similar pattern of improved performance from cold to hot trials. Qualitatively, however, CTS demonstrated a dramatic increase in distance walked at higher temperatures and BTS showed the least response, with hybrid cross types in the middle of these two extremes (Table 1, Fig. S4). For speed walked, we observed a positive response to increasing temperatures for the smallest cross types (bcCTS, CTS, and F1), while the larger categories maintained more constant speeds across the temperature treatments (Table 1). For all response variables, temperature had the largest effect, followed by additive effects of line-cross type.

Table 2.   Linear mixed model comparison using corrected Akaike’s information criterion (AICc). The best models for each response variable are denoted with an asterisk. All parameters listed in models are included singly in addition to any interactions
Modeld.f.AICc
DistanceSpeed
  1. *Best single-factor model.

  2. †Best two-factor model.

  3. ‡Best three-factor model.

  4. §Best overall model.

Null4316·081545·0968
Temperature5268·5312*17·3479*
Line-cross type9316·386447·0039
Mass5317·693443·8530
Temperature + Line-cross type10259·2560†15·2200†§
Temperature + Mass6270·543417·7644
Line-cross type + Mass10318·024445·4690
Temperature × Line-cross type15267·154718·7666
Temperature × Mass7271·950119·3714
Line-cross type × Mass15315·755848·8171
Temperature + Line-cross type + Mass11261·334215·4387‡
Temperature × Line-cross type + Mass16269·299819·1582
Temperature + Line-cross type × Mass16257·4974‡§21·4468
Temperature × Mass + Line-cross type12262·986016·9398
Temperature × Line-cross type × Mass27275·371629·0199

Backcrosses and F1 individuals walked significantly farther than CTS (Fig. 4a; Table 3). This general pattern of hybrid vigour was also evident for walking speed, but was not statistically significant for backcross and F1 hybrid walking speed contrasts to CTS (Fig. 4b; Table 3). F2 individuals performed poorly relative to other hybrid individuals, but performed similarly to each parental species with respect to distance walked (Fig. 4a; Table 3). However, we observed a decrease in the walking speed of F2 individuals relative to all other cross types (Fig. 4b; Table 3). We found some support for an interaction of relative mass and line-cross type on the distance travelled (Table 3) and there is a significant positive relationship between performance and mass for BTS (F1,3 = 10·2, P = 0·05; Fig. 5). In contrast, CTS and all other line-cross types show no significant relationship between mass and performance after accounting for mass differences between cross types.

image

Figure 4.  Box and whisker plot describing the effect of line-cross type on ln(x + 1)-transformed endurance times across temperatures: (a) distance walked and (b) speed walked. Solid black lines denote the median value for each cross type and the surrounding boxes represent the quartiles. The dotted lines extend to the minimum and maximum values. Statistics are presented in Table 3.

Download figure to PowerPoint

Table 3.   Summary of fixed-effects from regression models with the lowest AIC value for each of the three response variables measured
ResponseLevelValueStd.Errord.f.t-valueP-value
  1. *Significance at α = 0·1.

  2. **Significance at α = 0·05.

  3. ***Significance at α = 0·01.

  4. Line-cross type comparisons are made using the CTS value as a reference sample.

DistanceIntercept8·489840·319256826·59334<0·0001
Temperature [25]***2·040580·23502688·68246<0·0001
Line-cross type [F1]***1·527980·44027683·470530·0009
Line-cross type [F2]−0·273640·4197968−0·651860·5167
Line-cross type [BTS]0·267060·44027680·606570·5462
Line-cross type [bcBTS]**0·812610·38321682·120530·0376
Line-cross type [bcCTS]**0·753910·37547682·007930·0486
Mass−0·118320·1139368−1·038440·3027
Line-cross type [F1]:Mass0·039820·16626680·239490·8114
Line-cross type [F2]:Mass0·029800·12325680·241780·8097
Line-cross type [BTS]:Mass*0·219420·11953681·835640·0708
Line-cross type [bcBTS]:Mass0·125210·11839681·057620·2940
Line-cross type [bcCTS]:Mass0·028330·13694680·206890·8367
SpeedIntercept2·560570·075527433·90533<0·0001
Temperature [25]***0·334760·05397746·20301<0·0001
Line-cross type [F1]0·112750·10462741·077730·2847
Line-cross type [F2]*−0·184330·0997574−1·847860·0686
Line-cross type [BTS]0·095550·10462740·913240·3641
Line-cross type [bcBTS]−0·031140·0910674−0·341940·7334
Line-cross type [bcCTS]0·078490·08922740·879770·3818
image

Figure 5.  Scatter plots of the relationship between relative mass and ln(x + 1)-transformed distance walked. Linear mixed-effect models of the type ‘Response ∼ Temperature + Mass | ID’ for each line-cross type category reveal that only BTS performance is significantly affected by mass (F1,3 = 10·2, P = 0·05). Comparisons of interactions between line-cross types and mass are presented in Table 3.

Download figure to PowerPoint

For distance travelled we recorded R10 values ranging from 4·2 to 27·7 (Table 1) with a mean R10 of 12·6. Thus, on average salamanders walked more than 12 times farther at 25 °C than at 15 °C. This difference in distance travelled across temperatures was the greatest for CTS (1082 vs. 39 m at 25 °C and 15 °C respectively, R10 = 27·7), and the smallest for BTS (682 vs. 161 m, R10 = 4·2). For speed travelled, however, corresponding Q10 values were much lower, ranging from 1·1 for F2 hybrids to 1·7 for F1s (Table 1) and averaging 1·5. Once again, the Q10 for CTS (1·6) was on the high end of the observed range and that of BTS on the low end (1·2), suggesting that the two species differ in their response to temperature.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Effect of temperature on performance

Temperature appears to be the most important factor influencing locomotor performance (Table 2). The performance differences are very large, supporting the general pattern of increasing locomotor performance with increasing temperature that has been previously observed in ectotherms, including salamanders (Bennett 1990). Else & Bennett (1987) demonstrated a significant positive effect of temperature on running endurance (at 10 °C and 20 °C) for the Arizona tiger salamander (A. tigrinum nebulosum), and our data both corroborate their general findings and extend them taxonomically within the tiger salamander complex. However, our R10 values are dramatically higher than those found by Else & Bennett (1987) indicating a more pronounced effect of temperature on endurance in this system, particularly for the native CTS. While some of the differences in R10 values can be attributed to methodological differences (for example, we did not systematically increase the speed of the treadmill during trials), our results indicate a much more profound effect of temperature on distance travelled, particularly for animals with a complete (CTS) or nearly complete (bcCTS) native CTS genome. Given the unique life history of CTS, where newly metamorphosed juveniles emerge and migrate at the onset of the hot, dry California summer, this may suggest that physiological adaptation to warm temperatures is an important attribute of CTS success in the Central Valley environment.

Our data demonstrate that CTS, BTS and early-generation hybrids can all walk longer distances, at greater speeds, at higher temperatures. Assuming that walking greater distances affords new metamorphs an increased opportunity to find a suitable shelter in the few pre-dawn hours available to any given individual, our results imply that delaying metamorphosis as late as possible in the season might constitute an optimal strategy. This would particularly be the case for native CTS, which, although small at metamorphosis, display a nearly 30-fold increase in distance travelled at 25 °C compared to 15 °C. However, selection for metamorphic timing may be also be driven by other factors that select for early emergence, such as rainfall, humidity or competition with conspecifics for terrestrial burrows or food.

The total evidence available from previous studies, which is quite limited, suggests that early metamorphosis at cooler temperatures seems to be the overall preferred strategy for CTS metamorphosis. Based on field (C. Searcy & B. Shaffer, unpublished data), mesocosm (Ryan, Johnson & Fitzpatrick 2009) and laboratory (J.R. Johnson, B.M. Fitzpatrick & H.B. Shaffer unpublished) data, the CTS metamorphosis strategy appears to be: (i) get out as soon as possible or (ii) get out during the first rain after reaching some minimum size threshold. Although from a pure locomotor performance perspective, selection for delayed metamorphosis at a higher temperature seems to be optimal, the short hydroperiods and absence of rainfall typical of CTS breeding sites during spring and summer emergence appears to select for rapid emergence, regardless of cooler temperatures.

Effect of line-cross type on performance

Line-cross type was also important in determining the performance of individuals. Temperature and line-cross type did not significantly interact to affect performance, but we did observe some differences among line-cross types with respect to their response to changes in temperature, as evident from the wide-ranging R10 values (Table 1). Hybrid individuals (with the exception of F2s) consistently walked farther than individuals from either parental species. Specifically, F1 animals walked farther than any other cross type (significantly farther, but not faster than CTS), providing yet another example of hybrid vigour in this system. Critically, these are the first published data demonstrating hybrid vigour from the post-metamorphic salamander life history stage for CTS-BTS hybrids. Similar studies of fishes (Rosenfield et al. 2004) and lizards (Kearney, Wahl & Autumn 2005) have also found hybrid vigour with respect to locomotor performance. Predictably, F2 individuals in our study appear to suffer from hybrid dysfunction following the disruption of coadapted gene complexes during the first generation of recombination (e.g. Lynch 1991).

Our data suggest that large differences in performance based on the genetic composition of an individual’s genome have the potential to affect the fitness of particular genotypic classes (i.e. F1, bcBTS, bcCTS) and enhance the spread of the hybrid swarm. For example, landscape genetics of CTS has identified genetic structure resulting from isolation-by-distance and terrestrial habitat composition (Wang, Savage & Shaffer 2009). If hybrid individuals are capable of travelling farther than native CTS, they have the potential to disrupt natural patterns of population structure and spread non-native alleles deeper into the range of CTS. Ecological theory and empirical data also suggest that long-distance dispersers play a dominant role in determining the rate of spread of invasive species (Neubert & Caswell 2000). F1 hybrids in this system appear to be primed for long-distance dispersal. Very recent genetic data may further support this view; the populations studied by Wang, Savage & Shaffer (2009) have a low level of non-native genes present, suggesting a more efficient movement of non-native genes on the landscape than was previously thought (Fitzpatrick et al. 2010).

Effect of mass on performance

Bennett, Garland & Else (1989) found greater performance capacity of larger A. tigrinum salamanders and we expected to find similar results. However, we found no overall effect of mass on performance. We did identify a positive relationship between mass and performance for the introduced A. t. mavortium, which are genetically similar to the A. t. nebulosum salamanders used by Bennett, Garland & Else (1989). For all other genotypic categories, we did not detect a significant effect and in most cases, the relationship appeared negative (Fig. 5). The lack of significant relationships observed for non-BTS line-cross types may reflect real biological differences between A. californiense and other members of the tiger salamander complex or it may reflect the much smaller range of body mass examined in our study compared with that in Bennett, Garland & Else (1989).

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Overall, our results suggest an advantage to late metamorphosis in terms of temperature-dependent locomotor ability, although other trade-offs, such as increased desiccation risk related to metamorphic timing, remain uninvestigated. We did not predict this advantage given the reliably short times to metamorphosis of native CTS, and it provides the greatest benefit to hybrid and introduced individuals that routinely delay metamorphosis, particularly in artificially enhanced ponds with long hydroperiods. Native, introduced and hybrid tiger salamanders appear to have the capacity to move much farther at higher temperatures, with potential benefits in terms of finding suitable terrestrial refugia.

While we did not find significant interactions between line-cross type and temperature, the performance of F1 hybrids (and backcrosses to a lesser degree) suggests an enhanced capacity for spread of introduced genes relative to the typical dispersal distances of native CTS across a range of temperatures. The primary question for future research are whether an increased capacity for movement results in longer movements of hybrids in natural populations, and how these movements affect population structure and the spread of non-native alleles. Low-levels of non-native genes have already been detected across a large part of the CTS range (Fitzpatrick et al. 2009, 2010), and levels of introgression are particularly high in human-modified, relatively permanent aquatic habitats that allow for late-emerging hybrid metamorphs (Fitzpatrick & Shaffer 2007b). This improved understanding of the effect of abiotic environmental variation on the movements of salamanders will aid in predicting the consequences of global climate change on connectivity and population dynamics, a critical element in managing this biological invasion.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

We thank L. Austin for contributing the neoprene for the treadmill and the Shaffer Lab group for helpful discussion. Additionally, we thank B. Fitzpatrick, J. Piovia-Scott, M. Ryan and three anonymous reviewers for comments on early manuscript drafts. All procedures were approved by the UC Davis Animal Care and Use Committee (protocol #13269). This work was supported by a REU supplement from the National Science Foundation (DEB# 0516475) and the UC Davis Agricultural Experiment Station.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Austin, C.C. & Shaffer, H.B. (1992) Short-, medium-, and long-term repeatability of locomotor performance in the tiger salamander Ambystoma californiense. Functional Ecology, 6, 145153.
  • Bennett, A.F. (1984) Thermal-dependence of muscle function. American Journal of Physiology, 247, 217229.
  • Bennett, A.F. (1990) Thermal-dependence of locomotor capacity. American Journal of Physiology, 259, 253258.
  • Bennett, A.F., Garland, T. & Else, P.L. (1989) Individual correlation of morphology, muscle mechanics, and locomotion in a salamander. American Journal of Physiology Regulatory Integrative and Comparative Physiology, 256, 12001208.
  • Else, P.L. & Bennett, A.F. (1987) The thermal-dependence of locomotor performance and muscle contractile function in the salamander Ambystoma tigrinum nebulosum. Journal of Experimental Biology, 128, 219233.
  • Fitzpatrick, B.M. (2008) Dobzhansky-Muller model of hybrid dysfunction supported by poor burst-speed performance in hybrid tiger salamanders. Journal of Evolutionary Biology, 21, 342351.
  • Fitzpatrick, B.M. & Shaffer, H.B. (2004) Environment-dependent admixture dynamics in a tiger salamander hybrid zone. Evolution, 58, 12821293.
  • Fitzpatrick, B.M. & Shaffer, H.B. (2007a) Hybrid vigor between native and introduced salamanders raises new challenges for conservation. Proceedings of the National Academy of Sciences of the United States of America, 104, 1579315798.
  • Fitzpatrick, B.M. & Shaffer, H.B. (2007b) Introduction history and habitat variation explain the landscape genetics of hybrid tiger salamanders. Ecological Applications, 17, 598608.
  • Fitzpatrick, B.M., Johnson, J.R., Kump, D.K., Shaffer, H.B., Smith, J.J. & Voss, S.R. (2009) Rapid fixation of non-native alleles revealed by genome-wide SNP analysis of hybrid tiger salamanders. BMC Evolutionary Biology, 9, 176.
  • Fitzpatrick, B.M., Johnson, J.R., Smith, J.J., Kump, D.K., Voss, S.R. & Shaffer, H.B. (2010) Rapid spread of invasive genes into a threatened native species. Proceedings of the National Academy of Sciences of the United States of America, 107, 36063610.
  • Kearney, M., Wahl, R. & Autumn, K. (2005) Increased capacity for sustained locomotion at low temperature in parthenogenetic geckos of hybrid origin. Physiological and Biochemical Zoology, 78, 316324.
  • Loredo, I., Vanvuren, D. & Morrison, M.L. (1996) Habitat use and migration behavior of the California tiger salamander. Journal of Herpetology, 30, 282285.
  • Lynch, M. (1991) The genetic interpretation of inbreeding depression and outbreeding depression. Evolution, 45, 622629.
  • Morey, S. & Reznick, D. (2001) Effects of larval density on post-metamorphic spadefoot toads (Spea hammondii). Ecology, 82, 510522.
  • Neubert, M.G. & Caswell, H. (2000) Demography and dispersal: calculation and sensitivity analysis of invasion speed for structured populations. Ecology, 81, 16131628.
  • Petranka, J.W. (1998) Salamanders of the United States and Canada. Smithsonian Institution Press, Washington D.C.
  • Relyea, R.A. (2002) Costs of phenotypic plasticity. The American Naturalist, 159, 272282.
  • Riley, S.P.D., Shaffer, H.B., Voss, S.R. & Fitzpatrick, B.M. (2003) Hybridization between a rare, native tiger salamander (Ambystoma californiense) and its introduced congener. Ecological Applications, 13, 12631275.
  • Rosenfield, J.A., Nolasco, S., Lindauer, S., Sandoval, C. & Kodric-Brown, A. (2004) The role of hybrid vigor in the replacement of Pecos pupfish by its hybrids with sheepshead minnow. Conservation Biology, 18, 15891598.
  • Ryan, M.E., Johnson, J.R. & Fitzpatrick, B.M. (2009) Invasive hybrid tiger salamander genotypes impact native amphibians. Proceedings of the National Academy of Sciences of the United States of America, 106, 1116611171.
  • Scott, D.E. (1994) The effect of larvaldensity on adult demographic traits in Ambystoma opacum. Ecology, 75, 13831396.
  • Searcy, C.A. & Shaffer, H.B. (2008) Calculating biologically accurate mitigation credits: insights from the California tiger salamander. Conservation Biology, 22, 9971005.
  • Semlitsch, R.D., Scott, D.E. & Pechmann, J.H.K. (1988) Time and size at metamorphosis related to adult fitness in Ambystoma talpoideum. Ecology, 69, 184192.
  • Shaffer, H.B., Austin, C.C. & Huey, R.B. (1991) The consequences of metamorphosis on salamander (Ambystoma) locomotor performance. Physiological Zoology, 64, 212231.
  • Shaffer, H.B. & Trenham, P.C. (2005) Ambystoma californiense (Gray, 1853) California tiger salamander. Amphibian Declines. The Conservation Status of United States Species (ed. M.J.Lannoo). pp. 605608, University of California Press, Berkeley, California.
  • Trenham, P.C., Shaffer, H.B., Koenig, W.D. & Stromberg, M.R. (2000) Life history and demographic variation in the California tiger salamander (Ambystoma californiense). Copeia, 2000, 365377.
  • US Fish and Wildlife Science (2004) Endangered and Threatened Wildlife and Plants; Determination of Threatened Status for the California Tiger Salamander; and Special Rule Exemption for Existing Routine Ranching Activities 69 Federal Register, 149, 4721147248.
  • Wang, I.J., Savage, W.K. & Shaffer, H.B. (2009) Landscape genetics and least-cost path analysis reveal unexpected dispersal routes in the California tiger salamander (Ambystoma californiense). Molecular Ecology, 18, 13651374.
  • Wilbur, H.M. & Collins, J.P. (1973) Ecological aspects of amphibian metamorphosis. Science, 182, 13051314.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Fig. S1. Average daily temperatures.

Fig. S2. Treadmill design specifications.

Fig. S3. Line-cross type mass boxplots.

Fig. S4. Temperature interaction plots.

Table S1 Log-likelihood model comparisons.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
FEC_1723_sm_FS1.pdf452KSupporting info item
FEC_1723_sm_FS2.pdf929KSupporting info item
FEC_1723_sm_FS3.pdf172KSupporting info item
FEC_1723_sm_FS4.pdf335KSupporting info item
FEC_1723_sm_TS1.pdf28KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.