Seasonal trophic ecology of the white-ankled mouse, Peromyscus pectoralis (Rodentia: muridae) in central Texas.
Species of Peromyscus have been characterized as granivores with omnivorous tendencies that reflect opportunism in feeding habits (Cogshall 1928; Baker 1971; Montgomery 1989). Most dietary studies have compared two or more closely related species in sympatry and generally found Peromyscus are efficient foragers with broad diets consisting of seeds, fruits, green vegetation, and arthropods (Cogshall 1928; Hamilton 1941; Jameson 1952; Williams 1959; M'Closkey & Fieldwick 1975; Grant 1978; Knuth & Barrett 1984). Before interspecific interactions between potentially competing congeners can be fully determined, studies of each species in the absence of potential competitors are needed to ascertain the breadth of the trophic niche. In addition, climatic and phenological events may cause notable changes in community structure that affect the quantity and quality of food resources and shape the trophic niche of a species (Waser 1978a; 1978b). Seasonality in diet is evident in populations of most northern species of Peromysus. Because of the paucity of seasonal data for southern populations, a nonexistent or limited dietary cycle in these populations has been suggested (Montgomery 1989).
The white-ankled mouse, Peromyscus pectoralis Osgood, occurs in a variety of habitats over the central Mexican Plateau and Sierra Madre Oriental in Mexico northward to southeastern New Mexico and western and central Texas into southern Oklahoma. The species has a propensity for rocky environments, especially rock outcrops (Kilpatrick & Caire 1973; Schmidly 1972; 1974; Baccus & Horton 1984; Etheredge et al. 1989). Although substantial information exists about the habitat affinities of this species, the trophic niche of this mouse is poorly known. Alvarez (1963) observed the species eating fruits of nopal (prickly pear, Opuntia lindheimeri) cactus in Tamaulipas, Mexico. In Texas, Davis (1974) reported consumption of juniper berries, hackberry seeds, and acorns in central Texas. Schmidly (2004) mentioned the lack of a detailed food habits study for the species and speculated the diet consisted of seeds, cactus fruits, lichen, fungi, and insects. Here, the first detailed analysis and description of the seasonal trophic niche of P. pectoralis in central Texas is presented. No other sympatric congeners (i.e., P. attwateri) inhabited the study site (Mullican & Baccus 1990); thus, no accounting for interspecific competition or differential use of resources was necessary in the analysis.
White-ankled mice were collected 5 km W. San Marcos, Hays County Texas (29[degrees]47' N, 97[degrees]58' W) in an abandoned limestone quarry at the eastern periphery of its distribution. The landscape consists of large to medium size, strewn boulders and truncated limestone outcrops. Ashe juniper trees (Juniperus ashei) dominate the woody vegetation with trees and shrubs of green sumac (Rhus virens), Texas persimmon (Diospyros texana), sugar hackberry (Celtis laevigata), plateau live oak (Quercus fusiformis), agarito (Mahonia trifoliolata), and Roosevelt weed (Baccharis neglecta) also present. Herbaeceous vegetation consists of Johnsongrass (Sorghum halepense). King Ranch bluestem (Bothriochola ischaemum), little bluestem (Schizachyrium scoparium), Texas wintergrass (Stipa leucotricha), threeawn grass (Aristida sp.), prickly pear, frostweed (Verbesina virginica), Lindheimer senna (Cassia lindheimeriana), knotted hedgeparsley (Torilis nodosa), prairie bluet (Heydotis nigricans), Drummond skullcap (Scutellaria drumondii), white sweetclover (Melilotus albus), oneseed croton (Croton monanthogynus), Indianmallow (Abutilon incanum), common sorrel (Rumex acetosella), sensitivebriar (Schrankia roemeriana), pepperweed (Lepidium virginicum), and Texas bluebonnet (Lupinus texensis). Plant taxonomy followed Hatch et al. (1990).
Thornthwaite (1948) classified the climate of Hays County as [C.sub.1][B.sub.4] (dry subhumid, mesothermal) with a mean annual potential evapotranspiration of 99.7 to 114 cm. The mean monthly maximum temperature is 35[degrees]C (July) with a mean monthly minimum of 5[degrees] C (January). Mean annual precipitation is 85.7 cm. During this study, annual precipitation was below normal (76 cm).
Monthly collections between October 1976 and September 1977 were accomplished by Museum Special snap-traps set along 100 trap-station transects (spacing 15 m) during 12 consecutive days each month (total for study = 14,400 trap-days). Age (Schmidly 1972), standard external measurements, gender, and reproductive condition were recorded for each mouse. White-ankled mice were collected under Texas Scientific Permit SPR-0890-234.
Quantitative evaluations of stomach contents from 149 (67 females and 82 males) P. pectoralis followed relative-occurrence evaluation and histomicroscopic methods (Sparks & Malechek 1968; Free et al. 1970; Hansson 1970; Westoby et al. 1976; Ellis et al. 1977; Dawson & Ellis 1979; Cockburn 1980; Copley & Robinson 1983; Mclnnis et al. 1983; Copson 1986; Carron et al. 1990). Stomach contents were washed through a 0.125-mm sieve, homogenized, equally partitioned, and placed on 5 microscopic slides. These contents were considered as a whole or 100% irrespective of the extent of stomach fill (Hansson 1970). Twenty randomlyselected microscopic fields with food items were examined per slide under 100X magnification (total fields examined = 14,900). Identified food items in each field were recorded. Reference slides were also prepared of plant and animal matter from the collection site after mincing in a blender to mimic mastication.
Identification of food items and classification categories were based on anatomical or morphological structures of plant and animal matter. However, some matter due to extensive mastication remained unidentified. Ingested bait was ignored. Food categories applied to stomach contents were: (1) fruits and seeds-remains of testa, endosperm, exocarp, or individual minute seeds; (2) green foliage-leaf or stem tissues; (3) larval insects-nonsclerotized soft-bodies; (4) adult insects-sclerotized parts such as antennae, elytra, mouthparts, wings, or appendages; and (5) animal matter-primarily muscle tissue and hair. Fruits and seeds were combined as a food category because of a strong tendency of co-occurrence (Pitts & Barbour 1979). Calculations of percent frequency (%f), percent volume (%v) and relative importance (1) for each food item followed Obrtel & Holisova (1974; 1981). Composition percentages were obtained by dividing the total number of fragments of a given food item by the total number of fragments of all foods encountered. Analyses showed no difference in percent frequency of food items in diets of males and females; therefore, data for males and females were pooled.
Seasonal variation in food habits of small mammals has been primarily demonstrated by two methods. A season is defined by one method as a set number of months, usually four periods of three months, and food items enumerated within a season (Hamilton 1941; Whitaker 1966; Houtcooper 1978; Luo & Fox 1994). The second method lists food items consumed each month (Jameson 1952; Myers & Vaughan 1964; Bradley & Mauer 1971; Luce et al. 1980; Armgard & Batzli 1996). Both methods fail to reveal the dynamics of rapid dietary change. It is possible that seasonal change in animal diets may not be reflected by the four customary seasons; rather, based on dietary changes, one might find fewer or a greater number of seasons.
An Index of Change based on dietary dynamics was developed using a two-day moving average of the abundance of food items for groups of a set number of mice. The difference between group means for two consecutive such groups of mice was divided by the mean collection date of the mice in both groups. The Index of Change is represented by the formula:
[[[v.summation over (j = 1)]([[m + k].summation over (i = m)][x/k]) - [v.summation over (j = 1)]([[m + 2k].summation over (i = m + k + l)][x/k])]/[[[[m + k].summation over (i = m)][di/k]) - ([[m - 2k].summation over (i = m + k + 1)][di/k])] + 1
k = number of mice included in the first or second half (group) of the moving mean interval
m = accession number of the first mouse in the interval, with m incrementing by unity after each solution of the expression
x = percent composition of food item j in mouse i
v = total number of different food items
d = Julian day of capture for mouse i
Seasonal recognition was determined by the extent of an abrupt change in the diet. For the purposes of this study, any change [greater than or equal to]10 was defined as a substantial dietary change and recognized as the beginning of a season. Also, the stability of the period following an abrupt change was considered. If the Index of Change did not vary by a factor > 5 during a period of time, this sequence of days was considered as the term of a season. The next change in the index [greater than or equal to]10 marked the start of the next season.
Preference for food items was determined by a modified Ivlev's Index of Electivity (Ivlev (1961; Jacobs 1974; Krebs 1999). The electivity index indicated whether a plant was consumed in amounts similar to availability in the environment or active selection of the food. A value of zero indicated randomness in resource selection, a positive index indicated the food was selected in amounts greater than would be expected by chance encounter, and a negative index signified consumption of a food in quantities lower than predicted by randomness. The z test for comparing sample proportions (Sokal & Rholf 1969) was used to determine differences in male and female diets.
Seasonal diversity in the diet was calculated by Brillouin's Index of Diversity because samples were treated as collections rather than random samples from a larger biological community (Margalef 1958; Pielou 1966; Krebs 1999). Seasonal overlap in diet was determined by Morisita's Index of Similarity (Morisita 1959; Krebs 1999). The overlap coefficient varies from 0 when diets are completely distinct (no food categories in common) to 1 when diets are identical. Values [greater than or equal to]0.60 indicate significant overlap (Zaret & Rani 1971), but this should not be construed as statistically significance. Seasonal differences in the proportion of plant and animal matter in the diet were tested by a Goodness-of-Fit test.
Herbaceous vegetation occurring on the study site was assessed in January, April, August and November by randomly dropping a 0.5 m x 1 m quadrat 100 times in the study area. From these data, percent cover of each plant species was estimated (Myers & Vaughan 1964). Seed abundance was sampled by placing a 25 cm x 10 cm quadrat in the lower right corner of the larger quadrat at 10 sampling locations, removing the top 1-2 cm of soil, and screening soil samples through a series of descending size sieves (2.0 mm-180 [micro]m). Seeds were removed, counted, and identified to species.
The arthropod fauna was sampled monthly by sweeping through the ground vegetation along each trap-line with an insect net (38-cm diameter) (Beiner 1955). Arthropods inhabiting the foliage of trees were not sampled.
Twenty-three food items (18 plant and 5 animal), broadly categorized into 3 trophic groups (fruit and seed, green foliage, and animal matter), comprised the seasonal diet of P. pectoralis in central Texas (Table 1). No mouse exclusively consumed plant or animal matter. There was no difference in percent frequency of food items in diets of males and females (z = 0.0936, P > 0.34). The percent frequency, percent volume, and importance value of food items in the diet varied seasonally (Table 2). Prickly pear foliage, larval and adult insects, and arachnids were the only food items consumed in all seasons. Ten food items had a percent frequency of [less than or equal to] 5%. These foods were a miscellaneous assortment of animal tissues, green foliage, seeds, fruits, and possibly flower inflorescence. Twenty-two percent of stomachs had plant fragments so masticated that identification only as dicot material was possible. No monocot, lichen, or fungal materials were identified in any stomach. The greatest number of different food items found in any one stomach was 6 ([bar.X] = 3.26, SE = 0.73). Based on percent volume and percent frequency, 2 trophic categories (fruit and seed and animal matter) composed the bulk of seasonal diets. Seeds and fruits were the most frequently encountered category in the seasonal diet, occurring in 87% of stomachs. Ashe juniper berries were the most frequently consumed fruit. Arthropods were the primary animal matter in the seasonal diet with 84% of stomachs containing mostly insect fragments. Insects consumed were from the following taxa: Order Hymenoptera, Families Formicidae and Apidae; Order Coleoptera, Family Scarabaeidae; Order Orthoptera, Family Gryllidae; Order Collembola; Order Lepidoptera, Family Pyralidae; and Order Diptera.
Table 1. Characteristics of the seasonal diet of Peromyscus pectoralis in central Texas. Criterion Winter Early Late Summer Autumn Spring Spring Sample size (n) 39 21 23 31 35 Number of 8 10 8 15 12 items (S) Plant (%) 5 (62.5) 7 (70.0) 5 (62.5) 11 (73.3) 8 (66.7) Animal (%) 3 (37.5) 3 (30.0) 3 (37.5) 4 (26.4) 4 (33.3) Equally common 6.19 7.44 7.25 9.32 7.82 species Trophic 2.52 2.78 2.71 3.736 2.87 diversity (H) Table 2. Percent frequencies (%f), percent volumes (%v) and importance values (I) of items identified in the seasonal diet of Peromyscus pectoralis in central Texas. Winter Early Spring Late Spring N = 39 N = 21 N = 23 Item %f %v I %f %v I %f %v I Fruit and Seed Juniperus ashei 80 69.2 74.6 27 21.2 24.1 Rhus virens 41 14.1 27.6 8 0.8 4.0 Celtis laevigata 18 10.6 14.3 32 3.8 17.6 Opuntia lindheimeri Verbesina virginica 15 3.8 9.4 Diospyros texana 2 1.3 1.7 Sorghum halepense Yucca rupicola Bothriochola ischaemum Cassia lindheimeriana Unidentified Green Foliage Melilotus albus 43 24.3 33.7 18 11.2 14.6 Opuntia lindheimeri 16 2.6 9.3 8 1.5 4.8 18 5.8 11.9 Croton monanthogynus 27 12.6 19.8 Heydotis nigricans Verbesina virginica 18 1.9 10.0 Labitata Unidentified dicot 35 2.3 18.7 67 19.9 42.9 Animal Matter Larval insects 19 2.6 10.8 35 13.6 24.3 82 36.6 59.2 Adult insects 40 3.8 21.9 54 11.6 32.8 67 18.2 42.6 Arachnid 3 2.6 2.8 9 1.5 5.3 18 2.6 30.3 Gastropod Sylvilagus floridanus Summer Autumn N = 31 N = 35 Item %f %v I %f %v I Fruit and Seed Juniperus ashei 73 53.1 63.1 Rhus virens 17 4.7 10.9 Celtis laevigata 5 2.3 3.7 Opuntia lindheimeri 9 11.7 10.4 5 4.8 4.9 Verbesina virginica 5 4.7 4.9 Diospyros texana 25 9.4 17.2 Sorghum halepense 26 3.7 14.9 21 7.7 14.4 Yucca rupicola 8 5.9 7.0 Bothriochola ischaemum 37 6.1 21.6 Cassia lindheimeriana 5 1.2 3.1 Unidentified 18 5.1 11.6 Green Foliage Melilotus albus 9 3.5 6.3 Opuntia lindheimeri 21 6.3 13.7 5 1.5 3.3 Croton monanthogynus Heydotis nigricans 9 3.3 6.2 Verbesina virginica Labitata 5 3.5 4.3 Unidentified dicot 32 7.0 19.5 Animal Matter Larval insects 5 1.2 3.1 27 2.3 14.7 Adult insects 93 36.3 64.7 37 7.6 22.3 Arachnid 17 3.5 10.3 19 2.1 10.6 Gastropod 1 0.3 0.7 Sylvilagus floridanus 2 1.2 1.6
Five seasons (winter, December-January; early spring, February-March; late spring, April-May; summer, June-August; and autumn, September- November) were delineated by an Index of Change (Fig. 1). Morisita's Index of Similarity indicated no two seasons had unity (Table 3), yet some seasons were similar because of dietary overlap >0.60. The temporal continuity of resource use by white-ankled mice and dynamic changes in the diet demonstrated a general trend in which the immediate juxtaposed seasons to a given season were most similar (i.e., winter diet was most similar to the previous season, autumn, and the following season, late spring). The exception to this trend was autumn. A similarity value (0.42) between summer and autumn diets indicated substantial dissimilarities between these juxtaposed seasons; however, autumn and winter diets had the highest similarity. Furthermore, the autumn diet was more similar to the early spring diet. The communality of diet for these three seasons was based on the extent of fruit and seed use (primarily Ashe juniper berries) by white-ankled mice.
[FIGURE 1 OMITTED]
Table 3. Seasonal overlap in the diet of Peromyscus pectoralis (above diagonal) in centra Texas as measured by Morisita's Index of Similarity and number of food item common to all pair-wise seasonal comparisons (below diagonal). Season Winter Early Spring Late Spring Summer Autumn Winter - 0.61 0.367 0.394 0.88 Early Spring 6 - 0.688 0.622 0.603 Late Spring 4 7 - 0.657 0.392 Summer 4 7 7 - 0.418 Autumn 7 6 4 6 -
The substantial and almost exclusive use of the fruit and seed category was the distinctive feature of the winter diet (Table 2, Fig. 2). As a result, the winter diet was the most homogenous and least diverse of any season. The high use of fruit and seed in comparison to minimal consumption of animal matter (arthropods) and green foliage resulted in low seasonal diversity. Only Ashe juniper and green sumac berries were important staples, continuing the trend first seen in autumn of a preponderance of fruit and seed in the diet. The importance value (74.6) for Ashe juniper berries in winter was the highest importance value of any food item for any season. The highest seasonal percent volume (88.4%) of a trophic category (fruit and seed) occurred during winter. Plant matter in the winter diet was greater than animal matter ([X.sup.2] = 67.2; df = 1; P < 0.01).
[FIGURE 2 OMITTED]
The early spring was transitional with a decrease in the use and importance of the fruit and seed category (32.6% of overall seasonal percent volume) and an increase in the consumption and importance of green foliage. Three trophic categories (fruit and seed, green foliage, and animal matter) comprised the diet. Residual Ashe juniper berries and hackberries in the environment (Table 4) and new, succulent foliage of white sweet clover, oneseed croton, and unidentified dicot herbage were important plant items in the diet. Larval and adult insect consumption increased (26.7% of overall seasonal percent volume), but arthropod use ranked lower in importance than fruit and seed and green foliage (40.7% of overall seasonal percent volume). Early spring trophic diversity was higher than winter. Consumption of residual winter berries and new green foliage resulted in higher plant matter use than animal matter in the early spring diet ([X.sup.2]= 21.67; df =1; P < 0.01).
Table 4. Seasonal availability of the food resource of Peromyscus pectoralis. Fruits and seeds are the total number of items counted in ten 0.25 [m.sup.2] quadrats. Green foliage is the estimated percent occurrence in one hundred 0.5 [m.sup.2] quadrats. Item Winter Spring Summer Autumn Fruit and Seed Juniperus ashei 222 38 61 Rhus virens 21 5 Celtis laevigata 4 Quercus fusiformis 5 23 Sorghum halepense 40 60 Cassia Lindheimeriana 20 Diospyros texana 3 Green Foliage Sorghum halepense 8 14 23 20 Bothriochola ischaemum 8 17 28 31 Stipa leucotricha 22 15 2 15 Melilotus albus 28 16 8 Opuntia lindheimeri 5 3 2 4 Croton monanthogynus 1 4 9 Shrankia roemeriana 1 1 Torilis nodosa 19 8 2 Lepidium virginicum 1 2 Houstonia nigricans 4 1 Lupinus texensis 1 Rumex acetosella 4 Abutilon incanum 3 3
A shift in resource use in late spring to consumption of green foliage and animal matter supplanted fruit and seed use in the diet. The array of herbaceous plants in the diet expanded; however, extreme mastication of foliage made identification difficult and most material was classified as unidentified dicot herbage. Based on percent frequency, percent volume, and importance value, the most important food consumed during late spring was larval insects. Because of high consumption of larval insects, late spring was the only season where percent volume of animal matter (57.4%) exceed percent volume of plant matter in the diet. The most common larval insect in stomachs was the lepidopteran moth family, Pyralidae. Although there was a higher consumption of animal matter for this season, animal and plant matter use was similar ([X.sup.2] = 3.24; df= 1;P> 0.1).
The summer diet was the most heterogeneous of any season. White-ankled mice continued to eat green herbaceous foliage (23.6% of overall seasonal percent volume), but percent volume for most herbs was small. Stem and fruit use of prickly pear was the highest for any season. However, with maturation of grass and woody plant seeds, the importance of seeds and fruits in the diet increased (34.2% of overall seasonal percent volume), but percent volume for most species was low. Adult insect consumption was greatest in summer. Arthropods as a trophic group persisted as the most important food category (41.0% of overall seasonal percent volume). Adult crickets (Family Gryllidae) and beetles (Family Scarabaeidae) were the most common insects consumed. The highest consumption of arachnids occurred during summer. Trophic diversity (3.74) and richness (15 different food items) were the highest for any season. Materials attributable to 3 trophic categories were consumed during this season, and with increased seed use, there was a preponderance of plant matter in the summer diet. However, there was no difference between plant and animal matter use ([X.sup.2] = 2.56; df = 1; P > 0.2).
Autumn was a transitional season because of shifts and changes in the diet. Fruit and seed consumption (86.2% of the overall seasonal percent volume) increased and supplanted the importance of green foliage and arthropods in the diet (Fig. 2). This season was distinguished by an abrupt increase in the consumption of Ashe juniper and green sumac berries and prickly pear fruit. Arthropod use (12.0% of overall seasonal percent volume) decreased substantially in comparison to summer. Trophic diversity decreased because fewer food items were consumed. The consumption of berries and seeds and a diminished use of arthropods resulted in a diet dominated by plant matter ([X.sup.2] = 56.8; df = 1; P < 0.01).
Peromyscus pectoralis preferred certain foods in different seasons. Texas persimmon seeds had the highest electivity value (0.66) for any food item. No Texas persimmon trees occurred along trap-lines, and few trees were in the immediate study area. There were, however, an abundance of scattered raccoon (Procyon lotor) scats with Texas persimmon seeds in the study area. These scats concentrated Texas persimmon seeds in high density patches that provided an immediate access to seeds that otherwise were unavailable.
The most common fruit eaten was Ashe juniper berries. Even though the ground under and around juniper trees was often covered by berries, the electivity value (0.12) indicated only a slight preference for this food item. Since the quantity of Ashe juniper berries consumed by white-ankled mice was commensurate with availability (Table 4), P. pectoralis opportunistically fed on these berries. Hackberry and green sumac berries were not as common as Ashe juniper berries and availability was lower; however, electivity indices for these fruits (0.53 and 0.42, respectively) indicated selection by white-ankled mice. White sweetclover was the most abundant herbaceous plant in the spring plant availability sample (28% ground coverage). The electivity index for white sweetclover foliage (0.13) was similar to Ashe juniper indicating consumption was commensurate with availability. The negative electivity indices for seeds and/or foliage of Johnsongrass (-0.93), King Ranch bluestem grass (-0.96), prickly pear (-0.59), oneseed croton (-0.91), prairie bluet (-0.95), and Lindheimer senna (-0.99) indicated avoidance. No stomachs contained live oak acorns, agarito berries, and seeds or foliage of little bluestem grass, sensitivebriar, knotted hedgeparsley, pepperweed, Texas bluebonnet, common sorrel, or Indianmallow.
Species of Peromyscus are opportunistic feeders with variable use of food resources by season and availability. Major trophic categories in the diet are seeds, fruits, green plants, and animal matter (Montgomery 1989). Studies of the food habits of P. maniculatus, P. leucopus, P. californicus, P. eremicus, P. truei, and P. boylii (= attwateri) indicated either a moderate to common use of seeds, rare to common use of animal matter, rare to moderate use of green vegetation in all species except P. attwateri, and rare to moderate use of fruits (Brown 1964; Whitaker 1966; Flake 1973; Kritzman 1974; Vaughan 1974; Meserve 1976; Wolff et al. 1985).
Peromyscus food habits vary from season to season. Wolff et al. (1985) found univariate differences between seasons for seven of eight categories of food items in diets of P. maniculatus and P. leucopus. Both species ate more fleshy fruit in summer, more moths and butterflies in autumn, and more nuts in autumn and winter than other seasons. Hamilton (1941) reported that 180 P. leucopus noveboracensis collected between November and April in central New York consumed more arthropods than nuts/seeds or green plant matter. The diet between May and October was primarily arthropods with lesser amounts of fruits, nuts/seeds, and fungi. Whitaker (1963) found the primary food in the summer diet of 142 P. leucopus from New York was nuts/seeds with lesser amounts of arthropods and green plant matter. In addition, Whitaker (1966) stated 113 P. maniculatus from Indiana consumed principally nuts/seeds with lesser amounts of arthropods and green plant matter. Martell and Macauley (1981) found arthropods were the most common item and nuts/seeds, fruits, and fungi were less common with green plant matter and achlorophyllous plant matter present in miniscule amounts in stomach contents of 712 P. maniculatus taken between May and September in northern Ontario. Brown (1964) analyzed stomach contents of 20 P. attwateri collected in March, June, September, and December in southern Missouri and found seed use highest in June and lowest in March, insect use highest in December and lowest in June, fruit or berry use highest in March and lowest in June, consumption of green plant matter highest in June and September and lowest in March and December, and bulb fragments use highest in September and lowest in June.
There was a definite seasonal variation in the diet of P. pectoralis in contrast to the suggestion by Montgomery (1989) that southern populations of Peromyscus indicate little or no annual dietary cycle. Dietary trends of this species resemble the diet of Peromyscus attwateri and other species of Peromyscus; yet, there are differences. The seasonal diet of P. pectoralis appears opportunistic, especially in autumn and winter use of Ashe juniper berries, spring use of green foliage, and spring and summer consumption of insects. In contrast, plant matter dominated the winter diet (percent volume 91%), but the percent volume in the summer diet was only 42.6%. No other studies of Peromyscus have indicated this high a consumption of plant matter during winter (Montgomery (1989). Brown (1964) found only low or trace amounts of green plant matter in the diet of P. attwateri with the greatest amount of plant matter consumed being seeds.
As in other species of Peromyscus, consumption of animal matter by the white-ankled mouse had seasonal importance, especially insect use. Although insect abundance was monitored using sweeping of vegetation, this method did not provide adequate samples of the availability of the insects consumed by P. pectoralis during this study. Most adult insects eaten by P. pectoralis were ground crickets or beetles. The most consumed larval insects were catepillars of a pyralid moth that inhabits Ashe juniper trees. Insect consumption in late spring and summer were the highest reported for Peromyscus for these seasons (Flake 1973, Kritzman 1974, M'Closkey and Fieldwick 1975, Batzli 1977). Overall a similar consumption of animal matter by P. pectoralis and P. attwateri occurred in spring, summer, and autumn (Brown 1964). Insect use by P. pectoralis and P. attwateri in early spring and autumn was comparable, but winter consumption was dissimilar. The disparity in insect consumption during winter by P. pectoralis in central Texas compared to other species of Peromyscus was probaly phenological. The mild winters of central Texas allows for an extended period of activity by adult insects compared to those inhabiting northern latitudes and high altitude environs where winter temperatures are below freezing for an extended time. In Hays County, Texas, it is not unusual to have < 5 days in winter with temperatures at or below freezing.
A major difference in the diet of P. pectoralis compared to other species of Peromyscus was the importance of fruits in the seasonal diet. Peromyscus pectoralis began to eat friuts and seeds in summer as they matured with highest consumption of these foods in autumn extending into winter. Ashe juniper and green sumac berries were important components of the autumn through winter diets when abundance and availabilty was at a maximum in the habitat of P. pectoralis. Residual hackberries in the environment were important in the early spring through summer. Prickly pear fruits became important in summer through autumn with maturation of fruits. All of these berries and fruits became abundant and important in the diet of P. pectoralis based on their maturation and abundance in the environment (Table 4). Although the electivity index for Ashe juniper berries indicated a slight preference by white-ankled mice, the index value did not indicate the importance of this food in the diet. The high percent frequency of occurrence in the environment and percent volume in stomachs confirm the importance of this food in the seasonal diet and opportunism of P. pectoralis. Ashe juniper berries were most abundant between 15 December and 25 February. This period of high abundance of Ashe juniper berries overlapped the time when juvenile mice were most common in the population. Green sumac berries also had high abundance during this period. The high use of Ashe juniper and green sumac berries may have been the result of young white-ankled mice learning to eat this food in their natal environment and conditioning of adults to associate with habitats where berries are abundant (Drickamer 1970; 1976).
The abundance of green foliage was highest in spring through autumn for most herbaceous species except Stipa leucotricha that had highest availability in winter. The phenology for most herbaceous vegetation is invigorated growth during spring (Schmidly 2004). Peromyscus pectoralis had the highest consumption of the green foliage of herbaceous vegetation in early and late springwhen availability was highest. When available, green vegetation was a very common food of P. maniculatus (Kritzman 1974), moderately common in the diets of P. californicus and P. truei (Merritt 1974), but rarely consumed by P. leucopus (Whitaker 1966) and P. eremicus (Meserve 1976).
In a comparison of the size of food items consumed, larger berries and seeds had a positive electivity index; whereas, minute seeds had negative values (i.e., compare hackberry and green sumac berries vs. Johnsongrass and King Ranch bluestem grass seeds). Kantak (1983) found northern populations of both wild caught Peromyscus maniculatus bairdii and Peromyscus leucopus noveboracensis preferred larger size grass seeds of Andropogon. Moriarty (1977) suggested that size and selections of food items were critical expenses related to search time incurred by an animal while foraging. Both Ashe juniper and sumac berries were readily available seasonally, and foraging and handling times necessary to encounter and eat these items were probably minimal (Moriarity 1977). Otherwise, the diversity of food items consumed by P. pectoralis and the implied trophic niche was similar to other species of Peromyscus.
In one of the first food habits of a southern species of Peromyscus, this study delineated by an Index of Change a five season (winter, early spring, late spring, summer, and autumn) pattern of trophic ecology for Peromyscus pectoralis in central Texas. The species had considerable variation in the seasonal diet, and is primarily a frugivorous/granivorous herbivore with omnivorous tendencies reflecting opportunistic feeding habits.
Thanks are extended to R. W. Manning, M. F. Small, J. G. Brant, T. R. Simpson, and an anonymous reviewer for critical reviews of the manuscript. A special thanks is extended to E. Longcope for permission to collect on his property. Funding for this study came from a Texas State University Faculty Enhancement grant to J. T. Baccus.
Alvarez, T. 1963. The recent mammals of Tamaulipas, Mexico. Univ. Kansas Publ. Mus. Nat. Hist., 14:363-473.
Armgard, E. H. & G. O. Batzli. 1996. Effects of availability of food and interspecific competition on diets of prairie voles (Microtus ochrogaster). J. Mamm., 77:315-324.
Baccus, J. T. & J. K. Horton. 1984. Habitat utilization by Peromyscus pectoralis in central Texas. Pp. 7-26, in Festschrift for Walter W. Dalquest in honor of his sixty-sixth birthday, (N. V. Horner, ed.). Depart. Biol., Midwestern State Univ., Wichita Falls. Texas. 163 pp.
Baker, R. H. 1971. Nutritional strategies of myomorph rodents of North American grasslands. J. Mamm., 52:800-805.
Batzli, G. O. 1977. Population dynamics of the white-footed mouse in floodplain and upland forest. Amer. Midl. Nat., 97:18-32.
Beiner, B. P. 1955. Collecting, preparing and preserving insects. Sci. Serv. Entomolog. Div., Canada Depart. Agri., Publ. 932. Ottawa, Canada. 133 pp.
Bradley, W. G. & R. A. Mauer. 1971. Reproduction and food habits of Merriams kangaroo rat, Dipodomys merriami. J. Mamm., 52:497-507.
Brown. L. N. 1964. Ecology of three species of Peromyscus from southern Missouri. J. Mamm., 45:189-202.
Carron. P. L., D. C. D. Happold & T. M. Bubela. 1990. Diet of two sympatric Australian subalpine rodents, Mastacomys fuscus and Rattus fuscipes. Australian Wildl. Res., 17:479-489.
Cockburn, A. 1980. The diet of the New Holland mouse (Psettdomys novaehollondiae) and the house mouse (Mus musculus) in a Victorian coastal heathland. Australian Mamm., 3:31-34.
Cogshall, A. S. 1928. Food habits of deer mice of the genus Peromyscus in captivity. J. Mamm., 9:217-221.
Copley, P. B. & A. C. Robinson. 1983. Studies on the yellow-footed rock wallaby. Petrogale xanthopus Gray (Marsupialia: Macropodidae). II. Diet. Australian Wildl. Res., 10:63-76.
Copson, G. R. 1986. The diet of the introduced rodents Mus musculus L. and Rattus rattus L. on subantarctic Macquarie Island. Australian Wildl. Res., 13:441-445.
Davis, W. B. 1974. The mammals of Texas. Bull. No. 41. Texas Parks and Wildl. Depart., Austin, Texas. 257 pp.
Dawson. T. J. & B. A. Ellis. 1979. Comparison of the diets of yellow-footed rock-allabies and sympatric herbivores in western New Wales. Australian Wildl. Res., 6:245-254.
Drickamer, L. C. 1970. Seed preference in wild caught Peromyscus maniculatus bairdii and Peromyscus leucopus noveboracensis. J. Mamm., 51:191-194.
Drickamer, L. C. 1976. Hypotheses linking food habits and habitat selection in Peromyscus. J. Mamm., 57:763-766.
Ellis, B. A., E. M. Russell, T. J. Dawson & C. J. F. Harrop. 1977. Seasonal changes in diet preferences of free-ranging red kangaroos, euros, and sheep in western New South Wales. Australian Wildl. Res., 4:127-144.
Etheredge, D. R., M. D. Engstrom & R. C. Stone. Jr. 1989. Habitat discrimination between sympatric populations of Peromyscus attwateri and Peromyscus pectoralis in west-central Texas. J. Mamm., 70:300-307.
Flake, L. D. 1973. Food habits of four species of rodents on a short-grass prairie in Colorado. J. Mamm., 54:636-647.
Free, J., R. M. Hansen & P. L. Sims. 1970. Estimating dry weights of food plants in feces of herbivores. J. Range Mgmt., 23:300-302.
Grant, P. R. 1978. Competition between species of small mammals. Pp. 38-51, in Populations of small mammals under natural conditions (D. A. Snyder, ed.). Univ. Pittsburgh Press. Pittsburgh. Pennsylvania. 237 pp.
Hamilton, W. J. 1941. The food of small forest mammals in eastern United States. J. Mamm., 22:250-263.
Hansson, L. 1970. Methods of morphological diet microanalysis in rodents. Oikos, 21:255-266.
Hatch. S. L., K. N. Gandhi & L. E. Brown. 1990. Checklist of the vascular plants of Texas. Bull. MP-1655. Texas A&M Agri. Ext. Serv., College Station, Texas. 158 pp.
Houtcooper, W. C. 1978. Food habits of rodents in a cultivated ecosystem. J. Mamm., 59:427-430.
Ivlev, V. S. 1961. Experimental ecology of the feeding of fishes [translated from Russian by Douglas Scott]. Yale Univ. Press, New Haven, Connecticut. 302 pp.
Jacobs, J. 1974. Quantitative measurement of food selection. Oecologia, 14:413-417.
Jameson. E. W. 1952. Food of deer mice. Peromyscus maniculatus and P. boylii. in the northern Sierra Nevada, California. J. Mamm., 33:50-60.
Kantak, G. E. 1983. Behavioral, seed preference and habitat selection experiments with two sympatric Peromyscus species. Amer. Midl. Nat., 109:246-252.
Kilpatrick, W. C. & W. Caire. 1973. First record of the encinal mouse, Peromyscus pectoralis, for Oklahoma and additional records for north-central Texas. Southwestern Nat., 18:351.
Knuth, B. A. & G. W. Barrett. 1984. A comparative study of resource partitioning between Ochrotomys nuttalli and Peromyscus leucopus. J. Mamm., 65:576-583.
Krebs, C. J. 1999. Ecological methodology. Second Edition. Addison Welsey Longman, Menlo Park, California, xii+620 pp.
Kritzman, E. B. 1974. Ecological relationships of Peromyscus maniculatus and Perognathus parvus in eastern Washington. J. Mamm., 55:172-188.
Luce, D. G., R. M. Case, & J. Stubbendieck. 1980. Food habits of the plains pocket gopher on western Nebraska rangeland. J. Range Mgmt., 33:129-131.
Luo, J. & B. J. Fox. 1994. Diet of the eastern chestnut mouse (Pseudomys gracilicudatus). II. Temporal and spatial patterns. Australian Wildl. Res., 21:419-431.
Margalef, D. R. 1958. Information theory in ecology. Gen. Systematics, 3:36-71.
Martell, A. M. & A. L. Macauley. 1981. Food habits of deer mice (Peromyscus maniculatus) in northern Ontario. Canadian Field Nat., 95:319-324.
McInnis, M. L., M Vavra & W. C. Krueger. 1983. A comparison of four methods used to determine the diets of large herbivores. J. Range Mgmt., 36:302-306.
M'Closkey, R. T. & B. Fieldwick. 1975. Ecological separation of sympatric rodents (Peromyscus and Microtus). J. Mamm., 56:119-129.
Meserve, P. L. 1976. Food relationships of a rodent fauna in a California coastal sage scrub community. J. Mamm., 57:300-319.
Montgomery, W. I. 1989. Peromyscus and Apodemus: Patterns of similarity in ecological equivalents. Pp. 293-366, in Advances in the study of Peromyscus (Rodentia) (G. L. Kirkland, Jr. and J. N. Layne, eds.). Texas Tech Press, Lubbock, Texas. 366 pp.
Moriarity, D. J. 1977. Effect of search time on food preference in Peromyscus leucopus (Cricetidae). Southwestern Nat., 21:469-474.
Morisita, M. 1959. Measuring of interspecific association and similarity between communities. Memoirs of the Faculty of Science Kyushu Univ. Series E, 3:65-80.
Mullican, T. R. & J. T. Baccus. 1990. Horizontal and vertical movements of the white-ankled mouse (Peromyscus pectoralis) in central Texas. J. Mamm., 71:378-381.
Myers, G. T. & T. A. Vaughan. 1964. Food habits of the plain pocket gopher in eastern Colorado. J. Mamm., 45:588-598.
Orbetel, R. & V. Holisova. 1974. Trophic niches of Apodemus flavicollis and Clethrionomys glareolus in a low-land forest. Acta. Sci. Nat. Brno, 8:1-37.
Orbetel, R. & V. Holisova. 1981. Mean individual dietary diversity and its variation in selected small rodents. Folia Zool., 30:11-22.
Pielou, E. C. 1966. The measurement of diversity in different types of biological collections. J. Theoretical Biol., 13:131-144.
Pitts, W. D. & M. G. Barbour. 1979. The microdistribution and feeding preferences of Peromyscus maniculatus in the strand at Point Reyes National Seashore, California. Amer. Midl. Nat., 101:37-48.
Schmidly, D. J. 1972. Geographic variation in the white-ankled mouse, Peromyscus pectoralis. Southwestern Nat., 17:113-138.
Schmidly, D. J. 1974. Peromyscus pectoralis. Mamm. Species, 49:1-3.
Schmidly, D. J. 2004. The mammals of Texas. Univ. Texas Press, Austin, Texas. 501 pp.
Sokal, R. R. & J. F. Rholf. 1969. Biometry. W. H. Freeman & Co., San Francisco, California. 776 pp.
Sparks, D. R. & J. C. Maleckek. 1968. Estimating percentage dry weight in diets using a microscope technique. J. Range Mgmt., 21:264-265.
Thornthwaite, C. W. 1948. An approach toward a rational classification of climate. Geogr. Rev., 38:55-94.
Vaughan, T. A. 1974. Resource allocation in some sympatric, subalpine rodents. J. Mamm., 55:764-795.
Waser, N. M. 1978a. Competition for hummingbird pollination and sequential flowering in two Colorado wildflowers. Ecology, 59:934-944.
Waser, N. M. 1978b. Interspecific pollen transfer and competition between co-occurring plant species. Oecologica, 36:223-236.
Westoby, M., G. R. Rost & J. A. Weis. 1976. Problems with estimating herbivore diets by microscopically identifying plant fragments from stomachs. J. Mamm., 57:167-172.
Whitaker, J. O. 1963. Food of 120 Peromyscus leucopus from Ithaca, N. Y. J. Mamm., 44:418-419.
Whitaker, J. O. 1966. Foods of Mus musculus, Peromyscus maniculatus, and Peromyscus leucopus in Vigo County, Indiana. J. Mamm., 47:473-486.
Williams, O. 1959. Food habits of the deer mouse. J. Mamm., 40:415-419.
Wolff, J. O., R. D. Dueser, & K. S. Berry. 1985. Food habits of sympatric Peromyscus leucopus and Peromyscus maniculatus. J. Mamm., 66:795-798.
Zaret, T. M. & A. S. Rani. 1971. Competition in tropical stream fishes: support for the competitive exclusion principle. Ecology, 52:930-938.
JTB at: firstname.lastname@example.org
John T. Baccus, John M. Hardwick, David G. Huffman and Mark A. Kainer
Wildlife Ecology Program, Department of Biology Texas State University, San Marcos, Texas 78666
|Printer friendly Cite/link Email Feedback|
|Author:||Baccus, John T.; Hardwick, John M.; Huffman, David G.; Kainer, Mark A.|
|Publication:||The Texas Journal of Science|
|Date:||May 1, 2009|
|Previous Article:||Vegetation of South Padre Island, Texas: freshwater and brackish wetlands.|
|Next Article:||Changes in vegetation patterns and their effect on Texas kangaroo rats (Dipodomys elator).|