Jacobs syndrome

The Proximate Basis of the Oak Dispersal Syndrome: Detection of Seed Dormancy by Rodents1

SYNOPSIS. Previously we have shown how a range of physical and chemical characteristics of acorns influences the behavioral decisions of food-hoarding rodents which in turn affects the dispersal, establishment and spatial arrangement of oaks. One such behavior involves the selective caching of acorns of red oaks (subgenus: Erythrobalanus) over those of white oaks (Quercus) because of reduced perishability that results from delayed germination of acorns in the red oak group. In this study, we sought to identify the specific proximate cues (visual and olfactory) that eastern gray squirrels (Sciurus carolinensis) use when making these decisions. In two series of field experiments, we presented individual, free-ranging animals with pairs of experimentally altered acorns (that differed with respect to a single chemical or visual characteristic) and recorded their feeding and caching responses. Squirrels cached artificial acorns with pericarps (shells) of red oak acorns and ate those with shells of white oak regardless of the internal chemical composition of either type of acorn. Only when the shells of artificial acorns were first soaked in acetone (to remove potential chemical odors) did animals eat artificial acorns made with the shells of red oak acorns. Squirrels also ate one-year old red oak acorns that had broken dormancy, even when they exhibited no signs of germination. We argue that a chemical cue in the shell of acorns is important in the detection of seed dormancy and the decision to cache acorns, and that such a cue might ultimately contribute to the differential dispersal of red and white oaks by rodents.

AMER. ZOOL., 41:852-864 (2001)

INTRODUCTION

Much of the theory underlying the study of plant-animal interactions, at least in the terrestrial environment, has been derived largely from insect-plant systems (e.g., Price et al. [1991] includes only 3 of 27 chapters that consider interactions involving vertebrates). While plant-insect interactions are no doubt important, we suggest that this bias may potentially direct ecologists away from a number of systems, interactions, and processes that would otherwise result in a broader, more accurate perspective.

The study of mutualisms between fruitand seed-bearing plants and their consumers, however, is one especially promising area of investigation that has demonstrated an important role of vertebrates in the dispersal, establishment and population biology of plants (Howe, 1986; Fleming, 1991). Because of their size, energetic demands, and vagility, mammals and birds in particular exert a significant impact on the distribution and structure of many plant communities (Howe and Smallwood, 1982; Smith and Reichman, 1984; Howe, 1989; Stiles, 1989; Vander Wall, 1990). Indeed, there is now a mounting body of evidence that the habits of food-hoarding animals, for example, contribute significantly to the distribution of plants (Price and Jenkins, 1986; Stiles, 1989; Vander Wall, 1990).

Despite arguments that the mutualism between fruit and frugivores is the product of a loose or diffuse coevolution between broad taxonomic groups (Janzen, 1983; Howe, 1984; Herrera, 1985, 1986; Fleming, 1991), a number of specific plant and animal traits are thought to have resulted from the fruit-frugivore relationship (see review by Fleming, 1991). And, while much of the research to date on seed dispersal by mammals and birds has focused primarily on descriptive accounts of seed/fruit removal and dissemination (Price and Jenkins, 1986; Vander Wall, 1990), a growing number of more detailed studies now point to some unexpected relationships in some seed-dispersal systems (e.g., Vander Wall, 1992, 1994). Here, we introduce such a system involving the dispersal of oaks by rodents in which there is now evidence to argue for the existence of a strong mutualism.

The oak dispersal syndrome

Dispersal syndromes are constellations of fruit or seed characteristics (e.g., adaptations) that increase the probability of dispersal and establishment of seeds by specific groups of dispersal agents (Howe and Westley, 1988). The dispersal syndrome of oak, previously considered to involve accidental movement of a fruit (acorn) that exhibits few specific adaptations for dispersal, is now known to involve a far more complex interaction between the oaks and their mammalian and avian dispersal agents (Vander Wall, 1990; Steele and Smallwood, 2001). The specific nature of this relationship is linked most closely to the differences in the fruit of the two major groups of oaks found in North America (Kaul, 1985): the white oak group (WO, section Quercus) and the red oak group (RO, section Erythrobalanus). White oak species generally produce acorns with lower tannin (

These seed characteristics are now known to lead to important differences in the potential dispersal of RO and WO acorns (Smallwood and Peters, 1986). For example, at least five species of scatterhoarding rodents over a relatively broad geographic region (central Mexico to eastern and central US) selectively consume early germinating WO acorns but disperse and store those of RO some distance from their source (Smallwood, 1992; Hadj-Chikh et al., 1996; Steele and Smallwood, 1996; Smallwood et al., 1998; Steele et al., 2001), often failing to recover some of these cached seeds (Steele et al., 2001). In those instances in which they do cache WO acorns, at least two species of tree squirrels excise the embryo of these seeds preventing germination and rapid transfer of energy into an indigestible tap root by WOs (Fox, 1982; Pigott et al., 1991; Steele et al., 2001)-a probable adaptation to rodent seed predators (Fox, 1982; Vander Wall, 1990). Moreover, several physical and chemical characteristics of at least four species of RO (e.g., acorn geometry, lipid and tannin gradients) are suggested to promote partial consumption of these acorns (but not those of WO species) and subsequent dispersal and survival of these damaged seeds (Steele et al., 1993; Steele et al., 1998).

Collectively, these observations point to the strong potential for the differential dispersal of RO and WOs, with WO seedlings being clumped around parent trees and those of RO being dispersed some distance from their parental source (Smallwood et al., 1998). And, while preliminary investigations support this Differential Dispersal Hypothesis and a range of other potential consequences for the oaks (Smallwood et al., 1998; Steele and Smallwood, 2001), an equally important set of questions concerns the proximate basis for the behavioral decisions by food-hoarding mammals. Here we present the outcome of a series of experiments aimed at uncovering the specific cues that the animals use in these decisions, thereby providing an even stronger case for the mutualism between the oaks and their seed dispersers.

Seed perishability and food-hoarding decisions

Species that rely on stored food for extended periods of food shortage must select items that do not spoil during storage. It is therefore not surprising that the relative perishability of food stores has emerged as an important proximate factor influencing the caching decisions of many food-hoarding animals (Reichman, 1988; Post and Reichman, 1991; Gendron and Reichman, 1995; Hadj-Chikh et al., 1996; Post et al., 1998). This one factor appears to be an especially important determinant of autumn caching decisions of the eastern gray squirrels (Sciurus carolinensis) in North America. Indeed, this species' behavior of selective dispersal and caching of dormant RO acorns, consumption of WO acorns or embryo excision of WO seeds, and selective storing of sound RO acorns over those infested with insect larvae (Steele et al., 1996) all strongly support this contention.

Recently Hadj-Chikh et al. (1996) used a series of controlled field experiments with gray squirrels to test the relative effects of acorn perishability against other factors such as handling time, previously hypothesized by Jacobs (1992) to influence caching decisions in this species. By presenting free-ranging animals with pairs of acorns that varied in either perishability (due to differences in germination schedules) and handling time (due to differences in acorn sizes), Hadj-Chikh et al. (1996) found that the animals consistently stored those acorns most likely to remain dormant in the cache, regardless of handling time or other factors likely to explain caching responses. Moreover, Smallwood et al. (2001) tested and rejected other alternative hypotheses to explain these caching decisions.

Although these results strongly suggest that gray squirrels store dormant acorns of red oak over those of white oak due primarily to the differences in germination patterns (and therefore perishability), they say little about the specific cues on which the animals rely to make such decisions. Identification of these specific proximate cues, however, is a critical step towards better understanding the mechanism underlying caching decisions, the ecological effects of such decisions on the the differential dispersal of red and white oaks (Smallwood, 1992; Steele and Smallwood, 1994), and the potential evolutionary relationship between the squirrel and the oaks.

The primary objective of this study was therefore to identify the cues used by gray squirrels to discriminate between the two seed types. We hypothesized that such responses are based on either 1, visual cues associated with the outer pericarp of the acorn 2, olfactory cues emanating from the cotyledon of the seeds, or 3, some chemical odor from the pericarp (shell) of the acorn.

That gray squirrels are likely to rely on visual information when foraging and caching follows from those studies reporting keen visual acuity in this species as well as an ability to detect colors (Jacobs, 1981). However, an equally plausible explanation-supported by Steele et al. (1996) in which gray squirrels were found to discriminate between weevil-infested and non-infested red oak acorns even when no visual signs of such infestation were available-is that, like most mammals, squirrels depend on olfactory cues when making foraging and caching decisions. A fourth (4) hypothesis, first introduced by Smallwood and Peters (1986), was also tested. This hypothesis proposes that higher tannin levels might be used as a cue for the caching of red oak acorns.

METHODS

We conducted two sets of experiments with free-ranging gray squirrels similar in format to those performed by Hadj-Chikh et al. (1996). The first was designed to distinguish between the three alternative explanations that hereafter we refer to as the visual, cotyledon-odor, and pericarp-odor hypotheses. The second related set of experiments specifically addressed the tannin hypothesis.

Study site

We conducted all behavioral experiments in Kirby Park, a 48.5-ha urban park in Wilkes-Barre, Pennsylvania, following a format similar to that used in several of our previous studies (Smallwood and Peters, 1986; Hadj-Chikh et al., 1996; Steele et al., 1996). The site is ideal for experimental research on the behavior of gray squirrels because animals there exhibit all natural tendencies of wild squirrels (e.g., sensitivity to predation risks and caching of acorns), but are easily approached and observed due to their habituation to humans. The portion of the park in which we conducted all experiments is dominated by an open mature stand of silver maples (Acer saccharinum), red oaks (Quercus rubra) and pin oaks (Q. palustris) and is surrounded by an extensive urban neighborhood on one side and a heavily wooded area on the other. Regular movement of squirrels between the park and the forest ensures a more natural population, showing behavior patterns typical of wild squirrels.

Experimental design (visual vs. olfactory cues)

To identify the specific cues used by the squirrels when caching, we performed two series of experiments. In this first experimental series, we performed 9 experiments (Fig. 1). In experiment 1, we presented squirrels with a pair of whole, intact red oak and white oak acorns in order to verify the animals' normal caching and feeding responses to the two acorn types. In experiment 2, we presented a pair of artificial acorns both containing ground cotyledon of white oak acorn: one constructed of the shell of a white oak acorn and the other from the shell of a red oak acorn. We soaked the shells of both in acetone prior to their reattachment in order to remove potential chemical cues. For this experiment, the cotyledon-odor and pericarp-odor hypotheses predict that both acorn types will be eaten, whereas the visual hypothesis predicts that only the acorns constructed of a white oak shell will be eaten (Fig. 1). To test for the effects of artificial acorns on the animals' behavior, we presented acorns of red oak and white oak that were cut transversely and immediately reattached with glue (experiment 3).

In experiment 4, we presented a pair of artificial acorns that were both made from the shells of red oak acorns: one containing ground cotyledon of white oak and the other containing ground cotyledon of red oak. Shells of these acorns were also treated with acetone. In experiment 6, artificial acorns were prepared exactly as those in experiment 4 with the exception that the shells in this latter experiment were not treated with acetone. Predictions for experiments 4 and 6 differed only for the olfactory-odor hypothesis (Fig. 1). A first trial of experiment 6, conducted on an unseasonably cold day (-5 deg C) in which squirrels consumed all acorns, produced results that did not support any of the three hypotheses and is therefore not considered further. However, we immediately repeated this experiment following a second baseline experiment (experiment 5) to establish that the animals were again caching whole intact red oak acorns after the weather had returned to normal. Our second attempt at experiment 6 produced results that allowed elimination of one of the three hypotheses.

In experiment 7, we presented squirrels with a pair of whole red oak acorns, one of which was dormant, and the other germinating. We obtained germinating red oak acorns from a supply of seeds we had refrigerated (4 deg C) in 1995 that were no longer dormant and beginning to show signs of germination (i.e., radicle emergence). Experiment 8 was identical to that of 7, with the exception that the non-dormant seeds showed no visuals signs of germination (i.e., radicle emergence, split pericarp). We dissected a sample of these non-dormant acorns to verify their soundness prior to experiment 8. These two experiments are also used by Smallwood et al. (2001) to evaluate the ultimate advantage of squirrel caching preferences. In experiment 9, we presented each squirrel with a pair of whole, intact acorns of red oak, in which one of the pair was treated with acetone. This allowed us to test the effects of the acetone treatment on caching.

Experimental design (effects of tannin on caching)

The second series of experiments were designed to address the fourth hypothesis that the higher tannin concentrations typical of red oak acorns may be used as a cue when caching these seeds. We first conducted two preliminary trials in autumn 1995 and then a more formal series of six experiments in spring 1997. The first trial conducted in 1995 (2 October 1995) involved the presentation of a pair of white oak shells, one containing ground white oak cotyledon and the other ground white oak cotyledon to which 6%70 (by mass) tannic acid (Sigma Chemical) had been added. In the second (5 November 1995), we presented a pair of red oak shells one with white oak cotyledon and the other white oak cotyledon with the 6% tannin supplement.

The formal series of six experiments (numbered 10-15 to distinguish from those performed in the autumn) were conducted between 27 March and 21 April 1997 (Fig. 2). Experiment 10 was a baseline study to again establish that squirrels consistently stored red oak acorns and ate those of white oak. In experiment 11, we presented each squirrel with a pair of artificial acorns both constructed of the shell of red oak acorns. One of these contained ground cotyledon of white oak acorns and the other ground cotyledon of white oak with the tannin supplement. In this experiment, we predicted that if tannin was involved in the caching decision then only those artificial seeds containing the tannin additive should be cached. Similar predictions follow for experiment 12 in which we presented squirrels with a pair of artificial acorns, both constructed from the shell of white oaks: one containing only white oak cotyledon and the other white oak cotyledon and tannin (6%). Experiments 13, 14, and 15 were identical replicates of experiments 10, 11, and 12, respectively (Fig. 2).

Preparation of artificial acorns

Because gray squirrels are highly sensitive to the perishability of acorns it was critical to design artificial seeds that were recognized by the animals as potentially storable items. In preliminary, informal tests in two previous field seasons (with various cutting and gluing techniques) we found that gray squirrels frequently opened acorns at the glued seams, suggesting that they recognized these acorns as perishable items. However, the specific techniques that we finally adopted resulted in acorns that were consistently consumed from the basal end in a manner identical to that used for whole, intact seeds. In all of our experiments, we observed animals to open artificial acorns at the seam on only two occasions, thus suggesting that the procedure did not alter squirrel behavior.

To prepare artificial acorns, we cut acorns transversely, carefully removed the cotyledon, ground the cotyledon to a fine powder (with a laboratory blender), and sealed and stored the material until acorn shells were treated (

Immediately following their preparation, we weighed all acorns ( +/- 0.01 g) and within each trial we paired individual acorns of similar size to ensure minimal differences in mass and handling time. This resulted in mean differences in mass between acorn pairs as low as 0% (experiment 6) and no higher than 16% (experiment 2), thus allowing us to discount acorn size as a possible variable in our experiments (Table 1). We prepared all acorns

Field methods

For each experiment, we presented each of 20-24 individual squirrels with a single pair of acorns and recorded their responses. All squirrels within an experiment received the same acorn pair, but we altered order of presentation of the two acorns in a pair for each successive animal. We presented acorn pairs sequentially rather than simultaneously (see Hadj-Chikh et al., 1996) because we felt the former approach better approximated the true manner in which squirrels encounter different acorns. We presented each acorn following the methods outlined by Steele et al. (1996) and recorded whether the acorn was cached or eaten, the caching or eating time ( +/- 0.05 sec) and the distance it was dispersed prior to caching (+/- 0.5 m).

We maintained statistical independence between observations within (but not between) trials by presenting each individual with only a single pair of acorns. This was done by relying on individual marks previously placed on the animals with nianzol fur dye, and by following a linear transect through the park and moving to a new location following each trial with an unmarked individual (Smallwood and Peters, 1986).

Data analysis

We used Mann-Whitney U-tests to compare eating times, caching times and dispersal distances within each experiment. We compared caching frequencies for each trial with a G-test of independence for 2 X 2 contingency data and caching frequencies for each acorn type within an experiment with a log-liklihood ratio against a null hypothesis of random caching (1:1 ratio for caching and eating), followed by Bonferroni corrections for multiple comparisons. For the first series of experiments (n = 6), for which predictions differed between two or more of the hypotheses tested, we compared the overall number of correct and incorrect caching/eating decisions (based on predictions by each hypothesis) with a Wilcoxon signed rank test.

To conclude whether squirrels were making a decision to cache or eat a particular acorn type we used three criteria similar to those used by Hadj-Chikh et al. (1996). First, we considered that the squirrels had made a decision to eat rather than cache a particular acorn type when 70% or more of the observations within an experiment involved consumption of the acorns. This estimate was based on previous observations over several years in which we consistently observed that frequency of consumption of white oak acorns was between 70% and 100% of the observations within an experiment (Hadj-Chikh et al., 1996; Steele et al., 1996). In contrast, caching frequencies for red oak acorns consistently vary between 30 and 60% and rarely exceed 65%. Secondly, where they were significant, G-- tests of independence allowed us to determine those experiments in which caching frequencies were statistically different. And, finally we used the log-liklihood tests to determine those individual treatments within an experiment in which caching frequencies were statically lower than chance (1:1).

RESULTS

Effects of visual and olfactory cues on caching

In our experiments, acorn sizes only differed between 0 and 16% (Table 1, X(overscore) +/- SE = 5.7 +/- 2.2%). Similarly, eating times within experiments differed by no more than 31.2% (Table 1, X(overscore) +/- -- SE = 15.8 +/3.8%) and were not statistically different for any of the nine experiments (MannWhitney U-test: all P > 0.1). These observations indicate that we were able to successfully control for acorn size and handling time as a potential variable in these experiments (Table 1).

Squirrels cached significantly more intact red oak acorns than those of white oak in both experiment 1 (G-test: G 16.40, P

The pericarp-odor hypothesis was supported in all six experimental treatments (2,4,6,7,8,9), whereas the visual and the cotyledon-odor hypotheses were rejected in 5 and 4 of the experiments, respectively (Fig. 1; Table 2). For those experiments (n = 6) in which predictions differed between two or three of the hypotheses tested, there was no significant difference between the number of correct and incorrect caching and eating decisions for both the visual hypothesis (Z = -1.1, P = 0.249) and cotyledon-odor hypothesis (Z = 1.6, P = 0.106). In contrast, significantly more correct decisions were predicted by the pericarp-odor hypothesis (Z = 2.0, P = 0.046). Overall the three hypotheses successfully explained 71% (n = 174), 61% (n = 150), and 40% (n = 100) of the caching decisions, respectively. The relatively high success for the cotyledon-odor hypothesis is due to the several experiments in which identical outcomes were predicted for the two olfactory hypotheses. However, among the three experiments predicting different results (experiments, 4, 6, and 9), the pericarp-odor hypothesis successfully predicted 35% more of the caching decisions than did the cotyledon-odor hypothesis (n = 84 vs. 62).

In experiments 2 and 4, the pericarps of all artificial acorns were previously treated with acetone and squirrels consumed significantly higher numbers of all three seed types (85-91%) regardless of the shell type or their internal composition (cotyledon type). Similar effects of the acetone treatment were observed in experiment 9, in which squirrels cached significantly more whole, intact red oak acorns, than those treated with acetone (G = 8.06, P

In both experiments 4 and 6, squirrels were presented with a pair of artificial acorns constructed from the shell of red oak acorns, one containing cotyledon of a red oak acorn and the other the cotyledon of white oak acorns. In experiment 4, the shells were first treated with acetone and squirrels consumed nearly all of these seeds (>85%, Table 2). However, in experiment 6, in which neither of the acorn shells received the acetone treatment, both acorns were cached significantly more often than those of experiment 4 (G = 11.49, P

Although experiments 7 and 8 do not allow further discrimination of the two olfactory hypotheses, they do indicate that squirrels can detect and respond to seed dormancy when caching. When presented with pairs of acorns in which one was still dormant and the other not, squirrels stored significantly more of the dormant seeds (G = 9.10, P

Effects of tannin on caching

The initial trials involving manipulation of tannin content resulted in significantly higher rates of caching (G = 20.1, P 4.05, P

DISCUSSION

Three important conclusions follow from this study. First and contrary to hypothesis one, eastern gray squirrels do not rely on either visual cues related to the outer pericarp of the acorn or internal olfactory cues emanating from the cotyledon of the seed. Instead, some cue associated with the acorn pericarp most likely explains the observed results. Secondly, tannin does not appear, as previously conjectured (Smallwood and Peters, 1986), to be the specific cue on which squirrels base their caching decisions. Finally, it is clear that gray squirrels can distinguish between dormant and non-dormant acorns of red oak, even when non-dormant seeds show no visual, external indications of germination.

The most plausible explanation for these collective results is that squirrels rely on some olfactory cue associated with the pericarp in order to detect seed dormancy and determine which acorns to cache. Our results, coupled with those of Hadj-Chikh et al. (1996), support the notion that this cue is likely to serve as an indication of acorn dormancy (and reduced perishability) and that in the absence of such a cue, acorns are perceived as perishable, non-storable items. Alternatively, one might argue that animals rely on an olfactory cue as verification of non-dormancy and germination. However, if animals detected non-dormancy rather than dormancy it is unlikely that the acetone treatments would have resulted in the decline in caching activity.

Although the most popular contemporary view of seed dormancy holds that dormancy in most species is regulated by the interaction of several plant growth regulators (e.g., indoleacetic acid, abscisic acid, and gibberellic acid-Amen, 1968), control of dormancy in Q. rubra and at least four other species of red oaks has been attributed almost entirely to the presence of the pericarp (Bonner, 1968; Peterson, 1983; Hopper et al., 1985; Bonner and Vozzo, 1987). Prior to germination, the pericarp prevents germination by mechanically and chemically inhibiting water imbibition. But, following stratification, the pericarp apparently entraps critical gases that alter its permeability causing it to split, while its mechanical strength remains unaltered (Peterson, 1983). While these observations do not dismiss the possibility of a series of chemical cues originating in the embryo and/or cotyledon, they are consistent with our conclusion that the cues used by squirrels to detect seed dormancy are located primarily in the pericarp.

One important caveat that should be considered, however, is the possibility that the acetone treatments in our study did not remove some critical chemical compound from the shell, but instead compromised the integrity of the shell in some other way. It has been observed previously on a number of occasions that when the shells of acorns are cracked, opened, or removed, squirrels consistently consume these seeds (M. A. Steele, unpublished data). This suggests the possibility of a tactile, rather than an olfactory cue. However, we observed no qualitative changes in the tensile strength or coloration of the pericarps that could have contributed to the observed results. We also note that when first presented with an acorn, squirrels engage in a rolling behavior (up to a min. or more) in which they roll the acorn around just beneath the nose, frequently licking the pericarp. This supports our contention that the cue is chemical (olfactory) rather than tactile. It also should be noted that caching frequencies of artificial acorns were somewhat reduced from those of whole intact seeds, suggesting that the squirrels may perceive the modified seeds as being less attractive for storage. Hence, in those cases in which the squirrels chose to cache, our results may be conservative estimates of the true responses of the animals. The relative consistency of our caching results suggest that the artificial acorns used in our experiments were fully adequate for testing the proposed hypotheses.

The higher tannin content of red oaks represents a potentially confounding factor for most studies considering the influence of these phenolics on the feeding or caching behavior of animals (Smith and Follmer, 1972; Steele et al., 1993; Steele and Smallwood, 2001). Smallwood and Peters (1986) addressed this problem by experimentally manipulating tannin (and fat) content in artificial acorns and found that gray squirrels ate seeds with lower tannin levels, even at a cost to short-term feeding efficiency. They further hypothesized that tannin may be used as a cue for caching in order to maximize long-term rewards from caches. Hadj-Chikh et al. (1996) tested this hypothesis simultaneously with the perishability and handling-time hypotheses but found that predictions for the tannin and perishability hypotheses were the same in five of their six experiments. Nevertheless, they did observe that gray squirrels consistently ate acorns of chestnut oaks (Q. prinus), a white oak species with tannin levels comparable to that of most red oaks. Results of our second series of experiments support this observation and lead us to conclude that tannin levels themselves (at least in the cotyledon of the acorn) do not contribute to the caching decisions of these squirrels.

A final important implication of our study is the potential influence of these caching decisions by gray squirrels on the dispersal and distribution of oaks. Elsewhere, we have shown that gray squirrels selectively disperse and cache acorns of red oak over those of white oak (Steele and Smallwood, 1994; Hadj-Chikh et al., 1996; Steele et al., 1996) and that such decisions are likely to result in differential dispersal of the two types of oak (Smallwood et al., 1998). The ability to detect differences in acorn dormancy by gray squirrels likely represents one more of a suite of behavioral adaptations that has far-reaching effects on the distribution, ecology and the evolution of the oaks. In fact we suggest that the ability to detect seed dormancy may serve as the basis for dispersal syndrome of oaks across much of North America (Steele and Smallwood, 2001). Future studies aimed at identifying the specific chemical cues used to make caching decisions will help to further clarify this complex web of evolution-- ary and ecological interactions between oaks and tree squirrels.

ACKNOWLEDGEMENTS

We thank the City of Wilkes-Barre for permission to work at Kirby Park and the U.S. Fulbright Program and The National Science Foundation (DBI-9978807) for support during the final preparation of the manuscript. Financial support for the project was provided by the Biology Department and Faculty Development Fund of Wilkes University, and the National Science Foundation (DEB-9306641).

REFERENCES

Amen, R. D. 1968. A model of seed dormancy. Bot. Rev. 34:3-31.

Bonner, F.T. 1968. Water uptake and germination of red oak acorns. Bot. Gaz. 129:83-85.

Bonner, E.T. and J. A. Vozzo. 1987. Seed biology and technology of Quercus. U.S. Dept. Agr., So. For. Exp. Sta., Gen. Tech. Rep., SO-66.

Fleming, T.H. 1991. Fruiting plant-frugivore mutualism: The evolutionary theater and the ecological play. In P. W. Price, T W. M. Lewinsohn, G. W. Fernandes and W. W. Benson (eds.), Plant-Animal interactions: Evolutionary ecology in tropical and temperate regions, pp. 119-144. John Wiley & Sons, New York.

Fox, J. E 1974. Coevolution of white oak and its seed predators. Ph.D. Diss., University of Chicago, Illinois.

Fox, J. F 1982. Adaptation of gray squirrel behavior to autumn germination by white oak acorns. Evolution 36:800-809.

Gendron, R. P. and O. J. Reichman. 1995. Food perishability and inventory management: A comparison of three caching strategies. Am. Nat. 145: 948-968.

Hadj-Chikh, L. Z., M. A. Steele, and P.D. Smallwood. 1996. Caching decisions by grey squirrels: A test of the handling-time and perishability hypotheses. Anim. Behav. 52:941-948.

Herrera, C. M. 1985. Determinants of plant-animal coevolution: The case of mutualistic dispersal of seeds by vertebrates. Oikos 44:132-141.

Herrera, C. M. 1986. Vertebrate-dispersed plants: Why don't they behave the way they should? In A. Estrada and T. H. Fleming (eds.), Frugivores and seed dispersal, pp. 5-18. W. Junk Publishers, Dordrecht, Netherlands.

Hopper, G. M., D. W. Smith, and D. J. Parish. 1985. Germination and seedling growth of northern red oak: Effects of stratification and pericarp removal. For. Sci. 31:31-39.

Howe, H. E 1984. Constraints on the evolution of mutualisms. Am. Nat. 123:764-777.

Howe, H. E 1986. Seed dispersal by fruit-eating birds and mammals. In D. R. Murray (ed.), Seed dis

persal, pp. 123-190. Academic Press, Sydney Australia.

Howe, H. E 1989. Scatter- and clump-dispersal and seedling demography: Hypothesis and implications. Oecologia 79:417-426.

Howe, H. F and J. Smallwood. 1982. Ecology of seed dispersal. Ann. Rev. of Ecol. and Sys. 13:201228.

Howe, H. E and L. C. Westley. 1988. Ecological relationships between plants and animals. Oxford University Press, Oxford.

Jacobs, G. H. 1981. Comparative color vision. Academic Press, New York.

Jacobs, L. F 1992. The effects of handling time on the decision to cache by grey squirrels. Anim. Behav. 43:522-524.

Janzen, D. H. 1983. Dispersal of seeds by vertebrate guts. In D. J. Futtuyma and M. Slatkin (eds.), Coevolution, pp. 232-262. Seed dispersal. Sinauer Associates, Sunderland, MA.

Kaul, R. B. 1985. The reproductive morphology of Quercus (Fagaceae). Amer. J. Bot. 72:1962-1977. Ofcarcik, R. P and E. E. Burns. 1971. Physical and

chemical properties of selected acorns. J. Food Sci. 36:576-578.

Peterson, J. K. 1983. Mechanisms involved in delayed germination of Quercus nigra L. seeds. Ann. Bot. 52:81-92.

Pigott, C. D., A. C. Newton, and S. Zammitt. 1991. Predation of acorns and oak seedlings by gray squirrels. Q. J. Forest. 85:173-178.

Post, D. M. and 0. J. Reichman. 1991. Effects of food perishability, distance, and competitors on caching behavior by eastern woodrats. J. Mammal. 72: 513-517.

Post, D. M., M. W. McDonald, and 0. J. Reichman. 1998. Influence of maternal diet and perishability on caching and consumption behavior of juvenile eastern woodrats. J. Mammal. 79:156-162.

Price, M. V. and S. H. Jenkins. 1986. Rodents as seed consumers and dispersers. In D. F Murray (ed.), Seed dispersal, pp. 191-236. Academic Press, Sydney Australia.

Price, P. W., T W. Lewinsohn, G. W. Fernandes, and W. W. Benson. (eds.) 1991. Plant-animal interactions: Evolutionary ecology in tropical and temperate regions. John Willey and Sons, New York.

Reichman, 0. J. 1988. Caching behavior by eastern woodrats Neotoma floridana, in relation to food perishability. Anim. Behav. 36:1525-1532.

Short, H. L. 1976. Composition and squirrel use of acorns of black and white oak groups. J. Wildl. Manage. 40:479-480.

Smallwood, P. D. 1992. Temporal and spatial scales in foraging ecology: Testing hypotheses with spiders and squirrels. Ph.D. Diss., The University of Arizona, Tucson.

Smallwood, P. D. and W. D. Peters. 1986. Grey squirrel

food preferences: The effects of tannin and fat concentration. Ecology 67:168-174.

Smallwood, P. D., M. A. Steele, E. Ribbens, and W. J. McShea. 1998. Detecting the effect of seed boarders on the distribution of seedlings of tree species: Gray squirrels (Sciurus carolinensis) and oaks (Quercus spp.) as a model system. In M. A. Steele, J. E Merritt, and D. A. Zegers (eds.), Ecology and evolutionary biology of tree squirrels, pp. 223-231. Spec. Publ. Virginia Mus. Nat. Hist., 6.

Smallwood, P D., M. A. Steele, and S. H. Faeth. 2001. The ultimate basis of the caching preferences of rodents, and the oak-dispersal syndrome: Tannins, insects, and seed germination. Amer. Zool. 41: Rzl(_R Qan_Qm

Smith, C. C. and D. Follmer. 1972. Food preferences of squirrels. Ecology 53:82-91.

Smith, C. C. and 0. J. Reichman. 1984. The evolution of food caching by birds and mammals. Ann. Rev. Ecol. Syst. 15:329-351.

Steele, M. A., K. Gavel, and W. Bachman, W. 1998. Dispersal of half-eaten acorns by gray squirrels: Effects of physical and chemical seed characteristics. In M. A. Steele, J. E Merritt, and D. A. Zegers (eds.), Ecology and evolutionary biology of tree squirrels, pp. 223-231. Spec. Publ. Virginia Mus. Nat. Hist., 6.

Steele, M. A., L. Z. Hadj-Chikh, and J. Hazeltine. 1996. Caching and feeding decisions by Sciurus carolinensis: Responses to weevil-infested acorns. J. Mammal. 77:305-314.

Steele, M. A., T Knowles, K. Bridle, and E. Simms. 1993. Tannins and partial consumption of acorns: implications for dispersal of oaks by seed predators. Amer. Mid]. Natur. 130:229-238.

Steele, M. A. and P. D. Smallwood. 1994. What are squirrels hiding? Nat. Hist. 103:40-45.

Steele, M. A. and P D. Smallwood. 2001. Acorn dispersal by birds and mammals. In W. J. McShea and W. Healy (eds.), Oak forest ecosystems: Ecology and management for wildlife. Johns Hopkins University Press, Baltimore, MD. (In press)

Steele, M. A., G. Turner, P D. Smallwood, J. 0. Wolff, and J. Radillo. 2001. Cache management by tree squirrels: Experimental evidence for the significance of acorn embryo excision. J. Mammal. 82: 35-42.

Stiles, E. W. 1989. Fruits, seeds, and dispersal agents. In W G. Abrahamson (ed.), Plant-animal interactions, pp. 87-122. McGraw Hill Publishing, New York.

Vander Wall, S. B. 1990. Food hoarding in animals. University Chicago Press, Chicago.

Vander Wall, S. B. 1992. The role of animals in dispersing a "wind-dispersed" pine. Ecology 73: 614-621.

Vander Wall, S. B. 1994. Seed fate pathways of antelope bitterbrush: Dispersal by seed-caching yellow pine chipmunks. Ecology 75:1911-1926.

MICHAEL A. STEELE,2* PETER D. SMALLWOOD,3^ ALBERT SPUNAR,* AND ELISE NELSEN*

*Department of Biology, Wilkes University, Wilkes-Barre, Pennsylvania 18766

^Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104

1 From the Symposium An Integrative Approach to the Study of Terrestrial Plant-Animal Interactions presented at the annual Meeting of the Society for Comparative and Integrative Biology, 5-8, January 2000, at Atlanta, Georgia.

2 E-mail: mail:msteele@wilkes.edu

3 Present address of P.D. Smallwood is Department of Biology, University of Richmond, Richmond VA 23173, U.S.A.

Copyright Society for Integrative and Comparative Biology Sep 2001
Provided by ProQuest Information and Learning Company. All rights Reserved