Printer Friendly

Is the Coyote (Canis latrans) a potential seed disperser for the American Persimmon (Diospyros virginiana)?


Authors have proposed that several large fruited plants are anachronistic (Barlow, 2000), in that there is no obvious extant seed disperser, despite the production of large, apparendy edible fruit. Presumably, the coevolved mutualistic animal disperser is recently extinct. A classic example includes the Tombalacoque Tree (Calvaria major) of Mauritious (Temple, 1977), which may have coevolved with the Dodo Bird (Raphus cucullatus), although this example has come under intense scrutiny in recent years. (Barlow, 2000; Witmer and Cheke, 1991). North America may have an especially rich variety of anachronistic fruits, due to the late Pleistocene megafaunal extinction that occurred about 10,000 years ago. The American Persimmon (Diospyros virginiana) produces a large true berry which has been described as either a strong (Janzen and Martin, 1982) or moderate (Barlow, 2000) candidate as an anachronistic fruit.

The American Persimmon ranges throughout the eastern half of the United States from the Great Plains to the Atlantic Coast, north to the 40th parallel (Halls, 1981). The fossil record of Diospyros extends from the mid Cretaceous Period and Diospyros virginiana was present and presumably abundant in North America during the Pleistocene (Skallerup, 1953). Although it is unknown if there was a now-extinct megafaunal seed disperser for the American Persimmon (Mastadons have been suggested, Barlow, 2000), it is too large to be consumed by typical temperate North American dispersers such as birds (Skallerup, 1953). However, with its fleshy pericarp, it appears to be a large potential food reward (Chambers and MacMahon, 1994). The fruit is consumed by several native carnivores which may represent an example of an atypical seed disperser. Species from the Order Carnivora have not typically been considered good candidates for coevolution with plants as seed dispersers, since they feed predominately on other animals. However, many carnivores frequently include plant material in their diet (Herrera, 1989; Willson, 1993) especially fruits with high caloric value (Traba et al., 2006; Alves-Costa and Eterovick, 2007; Guitian and Munilla, 2010), and it has been suggested that carnivores contribute to seed dispersal, particularly in fleshy fruited temperate North American plants (Willson, 1993). Carnivores may even provide advantages such as larger foraging ranges and longer seed passage times (Zhou et al., 2008).

One candidate seed disperser for the Diospyros virginiana is the Coyote (Canis latrans) which is known to frequently include fruit in its diet. The seeds of various plants are known to survive gut passage in Coyotes and in some cases appear to benefit from the interaction (Silverstein, 2005). Coyotes consume persimmon fruits in large quantities (Chavez-Ramirez and Slack, 1993; Cypher and Cypher, 1999). An Illinois study showed that Persimmon seeds could survive gut passage through Coyotes, but the authors argued against coevolution because seeds had higher mortality when eaten (Cypher and Cypher, 1999).

Here we report the effect of Coyote ingestion on germination success, seedling emergence time, and growth after emergence. There have been several experimental design issues raised about seed dispersal experiments, especially in terms of proper controls. For instance, many researchers compare ingested seeds with seeds dissected from fruits. Samuels and Levey (2005) and Robertson et al. (2006) have stressed the importance of including planted whole fruits which is more representative of an unconsumed fruit in nature. We therefore included different experimental groups, some mimicking natural conditions and some representing common experimental procedures from the literature to determine how design affects results.


All collections were performed in Faulkner and Conway Counties, in central Arkansas. Freshly fallen persimmon fruits were collected from Nov. 5-12, 2010 and twelve Coyote scats were collected from Nov. 5-6, 2010 along trails and gravel roads within the two counties (seeds per scat: mean [+ or -] SE = 16.0 [+ or -] 3.72). Scats and fruits from trees were collected in the same area, although we could not determine from what tree the scat seeds originated. We first dissected 35 persimmon fruits to determine the average number of seeds per fruit (mean [+ or -] SE = 2.68 [+ or -] 0.34), a necessary calculation for the experimental design.

We established four experimental groups. The first experimental group consisted of whole fruit (WF) planted intact and unmanipulated. We did not determine number of seeds per fruit a prior since this process would have damaged the fruit. Seeds that were collected from Coyote scat were planted three to a pot (Coyote Multiple: CM), which is the average number of seeds found per fruit (mean = 2.68). The last two experimental groups were seeds dissected from fruits, one group consisted of a single planted seed (Dissected Single: DS) and the other group consisted of three planted seeds (Dissected Multiple: DM) in order to match the CM experimental group. Seeds from the CM, DM, and DS groups were randomly assigned to pots. Dissected seeds had the entire pericarp removed so that only the seed and seed coat remained. We consider the first two experimental groups (WF vs. CM) to be most natural, while the latter two experimental groups less natural, although these types of experimental treatments have been used in many published studies (Samuels and Levey, 2005).

On Nov. 11-12, 2010, seeds and fruits were planted 1 cm deep in 200 ml plastic pots filled with GardenPlus [TM] all-purpose potting soil. Number of pots per treatment groups (i.e., replicates) were 50 (WF), 50 (DS), 36 (CM), and 36 (DM). Total number of seeds for DS, CM, and DM were 50, 108, and 108 respectively. Number of seeds in the WF treatment was 104 and determined at the end of the experiment when pots were carefully searched for any seeds that failed to germinate. We placed the pots in a refrigerator at 4 C for 68 d to cold-stratify the seeds, which is a requirement for the American Persimmon to germinate (Halls, 1981). On Jan. 19, we placed the pots in a greenhouse, watered and observed every other day for 95 d, after which the experiment was terminated. Germination rate and emergence time was determined by the first appearance of the seedling above the ground. At the end of the experiment, seedlings were cut at the ground level, dried at 50 C for 24 h and weighed to determine above ground biomass. We also examined all seeds that had not produced a seedling to determine if germination had begun (in no cases had this occurred).

We analyzed germination rate by Chi-square analysis followed by pairwise comparisons. Since the measure was germination success (positive or negative) and some treatments have multiple seeds per plot, there is inevitable pseudoreplication in the Chi-square analysis. For seedling emergence times and final mass, we calculated the mean values per pot before analysis to avoid pseudoreplication. Seedling emergence time and final seedling mass were analyzed by One-way ANOVA followed by a Tukey post-hoc test if significance was found. All analyses were performed on IBM SPSS statistical package.


Experimental treatment had significantly different seedling emergence proportions (Chi square = 46.56, df = 3, P < 0.001). Seeds that had passed through Coyotes (CM) had almost the same percentage of seeds germinate (45.2%) as seeds from the whole fruit (WF) experimental group (45.4%). The other two experimental groups had much higher germination success (dissected single seeds (DS) = 84.0%, dissected multiple seeds (DM) = 78.5%). Pairwise comparisons indicated that the proportion of seeds that germinated were not significantly different for WF and CM treatments (Chi-square = 0.001, df = 1, P = 0.98) but they differed significantly from DM and DS treatments (WF vs. DM, Chi square = 24.86, df = 1, P < 0.001; WF vs. DS, Chi square = 20.85, df = 1, P < 0.001; CM vs. DM, Chi square = 25.01, df = 1, P < 0.001; CM vs. DS, Chi-square = 20.89, df = 1, P < 0.001) while DM and DS treatments were not significantly different (Chi square = 0.65, df = 1, P = 0.42).

There was a significant effect of experimental treatment (One-way ANOVA, [F.sub.3,121] = 16.51, P < 0.001) on number of days to emerge with whole fruits taking the longest time (mean = 54 [+ or -] 2.8 days). Dissected single seeds had the shortest emergence time (mean = 30 [+ or -] 2.0 days) while dissected multiple seeds had an intermediate emergence time (mean = 40 [+ or -] 2.2 days). Seeds that had been consumed by Coyotes had a significantly shorter emergence time (mean = 35 [+ or -] 2.5 days) when compared to only whole fruits, but were not different from the two dissected seed experimental groups.

The mass of the persimmon seedlings at the end of the experiment was significantly different between experimental groups (One-way ANOVA, [F.sub.3,121] = 2.95, P = 0.036). Seeds that passed through Coyotes produced seedlings with significantly smaller masses, approximately 29% smaller than the whole fruit experimental group. The seedling masses in the two dissected seed groups and whole fruit group did not differ significantly at the end of the experiment (Fig. 1). Both the whole fruit group and the dissected single group had significantly higher seedling masses compared to seedlings from seeds that passed through Coyotes.

There appeared to be a qualitative difference in seedlings among the experimental groups. The seed coat often remained attached to seedlings in coyote ingested seeds, which appeared to damage the young plants and affect post-emergence growth. Seedlings resulting from seeds contained in whole fruits almost always lacked the seed coat.


Clearly the seeds of Diospyros virginiana can survive gut passage through Coyotes which agrees with the results of Cypher and Cypher (1999). In fact, the germination rate for Coyote ingested seeds was almost exactly the same as those planted in whole fruits, our most natural comparison group. Interestingly, our two dissected seed groups had much higher germination success compared to our two more natural experimental groups, showing the importance of experimental design in these types of studies (Samuels and Levey, 2005). Our experiment demonstrates the necessity of including whole fruit experimental groups in seed dispersal studies. If we had not incorporated this treatment in our experiment and only compared Coyote ingested seeds to those artificially removed from fruit, we would have concluded that Coyote passage has a substantial negative effect on seed germination.

Coyote gut passage and artificial removal of seeds decreased the time required for emergence when compared to whole fruits. It is well known that some fruits contain growth inhibitors which delay seed germination (Robertson et al., 2006), presumably as an adaptation to ensure animal consumption before germination. Although persimmon fruit delayed emergence, maintaining the intact fruit did not affect final germination percentages. It has been suggested that fruit inhibition is a strategy to distribute germination over a longer period of time (Kelly et al., 2004) increasing the probability that at least some seedlings survive. Indeed, the range for the Coyote ingested was 70 days compared to 91 days for whole fruit seeds.

Seedlings that resulted from seed that had passed through Coyotes had significantly smaller mass by the end of the experiment than the WF and DS treatments, but not the DM seed treatment. This result was surprising considering that Coyote seedlings also emerged in less time, giving them more potential time to grow. Therefore, we suggest that Coyote gut passage of seeds retarded early seedling growth. Seeds that had passed through Coyotes (98%) and those that were artificially dissected from fruits (multiple seed pots = 89%, single seed pots = 93%), tended to have the seed coat remaining on the seedling to the point where growth was inhibited and obvious leaf and stem damage occurred. Seeds contained in intact fruits tended to produce seedlings without the seed coat attached (23% attached) which apparently allowed them to grow rapidly after germination. The effect was probably caused by damage to the ruminate endosperm which affected adherence to the seed coat (Bayer and Appel, 1996).

More experiments (field experiments) will be necessary to determine if Diospyros virginiana seeds benefit from being consumed by Coyotes. Considering the reduction in seedling quality in Coyote ingested seeds and similar germination rates between those seedlings from intact fruit and Coyote ingested seeds, we agree with Cypher and Cypher (1999) that there is no evidence of coevolution between these two species. Indeed it appears from this experiment that seedlings from Coyote scat will start their development substantially behind those that have not experience gut passage. This result is not necessarily surprising since Coyotes only recently expanded eastward in the eastern part of North America where the D. virginiana is abundant. Whether there is a net positive effect on persimmons will be dependent on the dispersal advantage outweighing the reduction in seedling quality shown in this experiment and will require field studies to determine. Cypher and Cypher (1999) showed that the Raccoon (Procyon lotor) may increase the germination percentage for persimmons and is a more likely coevolution candidate. Coyotes, with their recent range expansion into eastern North America, substantial population increase (Gompper, 2002), and heavy consumption of Persimmon fruit may therefore be exerting a negative effect on persimmon recruitment.

Acknowledgments.--We wish to thank M. Raney for help in the field and A. Willyard for advice on experimental design, reading an earlier draft of the manuscript, and help in the laboratory. R. Brown assisted with the greenhouse portion of the study. J. Penner assisted in the statistical analysis. Roger Anderson and two anonymous reviewers made valuable suggestions to improve the manuscript.


ALVES-COSTA, C. P. AND P. C. ETEROVICK. 2007. Seed dispersal services by coatis (Nasua nasua, Procyonidae) and their redundancy with other frugivores in Southeastern Brazil. Acta Oecol., 32:77-92.

BARLOW, C. 2000. The Ghosts of Evolution Past. Basic Books, New York, New York. 304 p.

BAYER, C. AND O. APPEL. 1996. Occurrence and taxonomic significance of ruminate endosperm. The Bot. Rev., 62:301-310.

CHAMBER, JEANNE. C. AND J. A. MACMAHON. 1994. A day in the life of a seed: Movements and fates of seeds and their implications for natural and managed systems. Annu. Rev. Ecol. Syst., 25:263-292.

CHAVEZ-RAMIREZ, F. AND R. D. SLACK. 1993. Carnivore fruit use and seed dispersal of two selected plant species of the Edwards Plateau, Texas. Southwest. Nat., 38:141-145.

CYPHER, B. L. AND E. A. CYPHER. 1999. Germination Rates of Tree Seeds Ingested by Coyotes and Raccoons. Am. Midl. Nat., 142(1):71-76.

GOMPPER, M. E. 2002. The ecology of northeast coyotes. Wildlife Conser. Soc., 17:7-42.

GUITIAN, J. AND I. MUNILLA. 2010. Responses of mammal dispersers to fruit availability: Rowan (Sorbus aucuparia) and carnivores in mountain habitats of northern Spain. Acta Oecol., 36 (2):242-247.

HALLS, L. K. 1981. Diospyros virginiana L. Common Persimmon. Silvics of N. Amer. Hardwoods. Agric. Handb., 654(2):294-298.

HERRERA, C. M. 1989. Frugivory and seed dispersal by carnivorous mammals, and associated fruit characteristics, in undisturbed Mediterranean habitats. Oikos, 55:250-262.

JANZEN, D. H. AND P. S. MARTIN. 1982. Neotropical anachronisms: the fruits the Gomphotheres ate. Science, 215:19-27.

KELLY, D.,J.J. LADLEY, AND A. W. ROBERTSON. 2004. Is dispersal easier than pollination? Two tests in new Zealand Loranthaceae. New Zeal. J. Bot., 42:89-103.

ROBERTSON, A. W., A. TRASS, J. J. LADLEY, AND D. KELLY. 2006. Assessing the benefits of frugivory for seed germination: the importance of the deinhibition effect. Funct. Ecol., 20:58-66.

SAMUELS, I. A. AND D. J. LEVEY. 2005. Effects of gut passage on seed germination: do experiments answer the questions they ask? Funct. Ecol., 19:365-368.

SILVERSTEIN, R. P. 2005. Germination of native and exotic plant seeds dispersed by coyotes (Canis latrans). Southwest. Nat., 50:472-478.

SKALLERUP, HARRY. R. 1953. The distribution of Diospyros virginiana L. Ann. Mo. Bot. Gar., 40(3):211-225.

TEMPLE, S. A. 1977. Plant-animal mutualisms: coevolution with Dodo leads to near extinction of plant. Science., 197:885-886.

TRABA, J., S. ARPIETA, J. HERRANZ, AND C. M. CLAMAGIRAND. 2006. Red fox (Vulpes vulpes L.) favour seed dispersal, germination and seedling survival of Mediterranean Hackberry (Celtis australis L.). Acta Oecol., 30(1):39-45.

WILLSON, M. F. 1993. Mammals as seed-dispersal mutualists in North America. Oikos., 67:159-176.

WITMER, M. C. AND A. S. CHEKE. 1991. The Dodo and the Tambalacoque Tree: An Obligate Mutualism Reconsidered. Oikos, 61:133-137.

ZHOU, Y., L. ZHANG, Y. KANEKO, C. NEWMAN, AND X. WANG. 2008. Frugivory and seed dispersal by a small carnivore, the Chinese ferret-badger, Melogale moschata, in a fragmented subtropical forest of central China. For. Ecol. Manag., 255(5-6):1595-1603.




Department of Biology, Hendrix College, 1600 Washington Avenue, Conway, AR 72032

(l) Corresponding author:
COPYRIGHT 2013 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Roehm, Katherine; Moran, Matthew D.
Publication:The American Midland Naturalist
Article Type:Report
Geographic Code:1USA
Date:Apr 1, 2013
Previous Article:Age-related morphometrics and sex ratios in the eastern mole (Scalopus aquaticus).
Next Article:Prairie dog aboveground aggressive behavior towards black-footed ferrets.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters