Fall and Winter Foods of Northern Pintails in the Sacramento Valley, California

Michael R. Miller

Abstract: Food habits of northern pintails (Anas acuta) were investigated on 3 national wildlife refuges in the western portion of the Sacramento Valley, California, from August to March 1979-82. Pintails consumed ≥97% (aggregate % dry wt) plant food during diurnal foraging on national wildlife refuge rice, summer-irrigated, and summer-dry habitats from August through January. Invertebrate use increased to 28.9-65.6% of the diet in these habitats during February and March. Rice, swamp timothy (Heleochloa schoenoides), flatsedges (Cyperus spp.), common barnyardgrass (Echinochloa crusgalli), southern naiad (Najas guadalupensis), and smartweed (Polygonum spp.) seeds, miscellaneous vegetation, snails (Gastropoda), and midge (Diptera) and water beetle (Coleoptera) larvae were most important. These foods usually were taken proportional to or greater than availability. Rice was the most important food of pintails feeding nocturnally off the refuges in harvested rice fields from October through January (99.7%) and February and March (63%; barnyardgrass formed 31% of the diet). In August and October, some pintails consumed invertebrates or bulrush (Scirpus spp.) seedlings in marshes soon after feeding in refuge rice (Aug) or harvested commercial rice fields (Oct), thereby increasing dietary protein. In late winter, females and males obtained similar (P > 0.05) percentages of invertebrates from refuge habitats. Important dietary seeds and invertebrates contained high protein or metabolizable energy content. Management should maintain adequate seed production in fall and mid-winter and invertebrate biomass in late winter.

Table of Contents

• Introduction
• Study Area
• Methods
• Results
• Discussion
• Literature Cited

Tables

• Table 1 -- Esophageal contents and preference ranks of major foods of northern pintails collected while feeding diurnally in unharvested rice fields left flooded and rice fields flooded after harvest on 3 northern California wildlife refuges, 1979-82.
• Table 2 -- Esophageal contents and preference ranks of major foods of northern pintails collected while feeding diurnally in summer-dry seasonal marshes on 3 northern California wildlife refuges, 1979-82.
• Table 3 -- Esophageal contents and preference ranks of major foods of northern pintails collected while feeding diurnally in summer-irrigated seasonal marshes on 3 northern California wildlife refuges, 1979-82.
• Table 4 -- Esophageal contents and preference ranks of major foods of northern pintails collected while feeding in marshes of 3 northern California wildlife refuges shortly after returning from feeding in commercial rice fields, 1979-82.
• Table 5 -- Esophageal contents of northern pintails collected returning to roost on 3 national wildlife refuges before sunrise from nocturnal feeding in harvested commercial rice fields, northern California, 1979-82.
• Table 6 -- Aggregate percent of invertebrates in habitat samples taken at pintail collection sites on 3 northern California national wildlife refuges, 1979-82.
• Table 7 -- Proximate analysis and apparent metabolizable energy of selected foods making up ≥5% dry weight of the esophageal content of northern pintails from 3 northern California wildlife refuges, 1979-82.

Introduction

Northern pintails begin to arrive in the Sacramento Valley, California, from northern breeding and staging areas the 1st week of August. About 315,000 pintails are present by mid-September (LeDonne 1980) and > 1.5 million in January (LeDonne 1981); many remain through March. This large number of ducks present over this extended time period represents a significant demand for food resources.

About 96% of the original marsh present in the Central Valley (the Sacramento Valley forms the northern half of the Central Valley) has been converted to other uses including rice farming. Waste grain left after harvest is an important waterfowl food (Gilmer et al. 1982). However, the importance to waterfowl of seeds and invertebrates from remaining marshes must be understood to manage food resources in Sacramento Valley wetlands. The role of winter food supply in the dynamics of waterfowl populations is poorly understood, but winter habitat conditions may affect recruitment (Heitmeyer and Fredrickson 1981).

Recently, Beam and Gruenhagen (1980), Connelly and Chesemore (1980), and Euliss (1984), using esophageal samples, demonstrated significant use of invertebrates by pintails during winter in the San Joaquin Valley (the southern half of the Central Valley). However, food habit data available for Sacramento Valley waterfowl were collected in the 1940's and 1950's and were based almost entirely on analysis of gizzard contents (Calif. Dep. Fish and Game, Sacramento, and U.S. Fish and Wildl. Serv. [1954], unpubl. rep., Willows, Calif.). Such analyses were biased toward hard seeds (Dillon 1958, Swanson and Bartonek 1970). Specific foods consumed and their availability in the environment were not compared. Therefore, I used esophageal analysis to investigate foods consumed by northern pintails during late summer, fall, and winter in the Sacramento Valley. My objectives were to determine food preference and to describe seasonal variation in consumption of specific items relative to habitat types and behavioral and nutritional requirements.

I thank the following people for their critical review of the manuscript: J. M. Hicks, G. L. Krapu, P. A. Opler, S. L. Paulus, K. J. Reinecke, T. L. Shaffer, and G. A. Swanson. D. S. Gilmer gave administrative and technical support and provided detailed review and helpful comments. J. H. Day, J. M. Hicks, S. R. Lawry, C. A. Lefever, D. L. Orthmeyer, and D. C. Weinrich processed esophagus and habitat samples. J. Ruckman of the Agron. Dep., Univ. California, Davis, provided protein analyses of seeds. J. J. Chesi, S. M. Dobbins, and B. J. Hass of the Calif. Dep. Food and Agric. identified many seeds. I thank J. B. Helvie, J. S. Miller, and J. L. Wilson of Sacramento Natl. Wildl. Ref. for their support and cooperation during the study. D. W. Sparling processed quantitative data, and D. H. Johnson and T. L. Shaffer provided statistical consultation.

Study Area

I conducted the study on Sacramento, Delevan, and Colusa national wildlife refuges (NWR's) located in Glenn and Colusa counties in the central Sacramento Valley. These refuges are in the rice-growing region of California and total 8,275 ha. Refuge crops, marsh units, and ponded areas are supplied with about 26,000 ha-m of drain water from rice fields and fresh water from the Sacramento River (Sacramento NWR, unpubl. rep., Willows, Calif.).

Habitat features on each refuge consisted of managed marsh, seashore saltgrass (Distichlis spicata) uplands, Fremont cottonwood (Populus fremontii) and willow (Salix spp.) woodlots, and rice fields. Marsh units were either dry in the summer and flooded from about 1 September to 1 April (summer-dry), or periodically flooded through the summer as well (summer-irrigated). Permanent ponds made up <5% of refuge marsh habitat. Dominant vegetation of summer-irrigated marshes included common barnyardgrass, smartweeds, knotgrass paspalum (Paspalum distichum), and bearded sprangletop (Leptochloa fascicularis). Flatsedges, roughseed bulrush (Scirpus mucronatus), common arrowhead (Sagittaria latifolia), and southern naiad were common in rice fields. Plants characteristic of alkaline soil conditions such as swamp timothy, curlycup gumweed (Grindelia squarrosa), common Bermuda grass (Cynodon dactylon), and woolly marbles (Psilocarphus brevissimus) characterized summer-dry habitats. Tule bulrush (Scirpus acutus) and southern cattail (Typha domingensis) were dominant emergents in all habitats. Rice was grown on the refuges, and 75-150 ha were left unharvested on each refuge for duck use to prevent depredation of commercial crops from August through September. Several thousand ha of private duck clubs, consisting of rice fields (flooded after harvest) and managed marsh, were present throughout the region (Gilmer et al. 1982).

Methods

Pintails were collected by shooting each month from August through March 1979-82. Contents of 187 usable esophagi ≥ 5 food items present) were available for analysis. A sample of pintails was collected from October to March, before sunrise, from flocks returning to refuge roosts from nocturnal feeding locations outside the refuges (nocturnal feeders). However, most data were obtained from ducks feeding diurnally on the refuges. These pintails were collected at locations where they were feeding, not at random or preselected locations. Normally, actively feeding pintails were observed for ≥10 minutes (Swanson and Bartonek 1970) before they were collected. In August and September, dense vegetation made observation impossible in some habitats, so ducks were collected and assumed to have been feeding if food was present in the esophagus. Esophageal contents were removed immediately and preserved in 80% ethyl alcohol (Swanson and Bartonek 1970). Sample bags were chilled in ice until transferred to frozen storage. Habitat samples, obtained at pintail collection sites, to measure food availability included 5 benthic core (Swanson 1978) and water column (Pederson and Pederson 1983) samples and 5 line transect samples for emergent vegetation. The latter were 10 m in length with plant species recorded at 10-cm intervals.

Data were analyzed separately for nocturnal feeders and for pintails feeding on refuges in flooded-unharvested rice fields (Aug and Sep only) and rice fields flooded after harvest, in summer-dry and summer-irrigated seasonal marshes, and on 1 permanent pond. In addition, ducks collected while feeding in refuge marshes shortly after returning from nocturnal feeding in rice fields were analyzed separately if rice was present in esophagi along with foods obtained at the feeding site (Marsh Feeders With Rice = MFWR). Foods consumed in rice fields would have biased results if data from MFWR were simply added to the respective habitat-type category.

Foods from esophageal and habitat samples were identified by published descriptions (Mason 1957, Pennak 1978) or by California Department of Food and Agriculture personnel in Sacramento. Foods were dried to constant weight at 65 C. Dry weight was used instead of volumetric measurements (Swanson et al. 1974a) because of direct application to bioenergetic and nutritional considerations (Reinecke and Owen 1980) and differences in moisture content of seeds and invertebrates (Sugden 1973).

Crude protein, crude fat, crude fiber, ash, and nitrogen-free extract (NFE) were determined for selected pintail foods obtained from esophagi (Assoc. Off. Anal. Chem. 1980). Apparent metabolizable energy (AME) (Harris 1966:3) values were calculated from digestibility coefficients (Harris 1966:15) and metabolizable energy of digestible protein, fat, and NFE (Vohra 1972). Crude fiber was assumed to be undigestible (Almquist and Halloran 1971).

Data were summarized by percent occurrence (Swanson et al. 1974b) and aggregate percent dry weight (Prevett et al. 1979) for fall (Aug-Sep), mid-winter (Oct-Jan), and late winter (Feb-Mar). These seasonal groupings reflected major changes in percent of the day spent feeding (Miller 1985) and the close relation of food habits within the periods, as determined a posteriori. Comparisons of invertebrate consumption between sexes were made with Mann-Whitney U-tests (Siegel 1956:116-127). Proportion of invertebrates available at feeding sites and in esophageal samples among fall and mid-and late winter were compared using Kruskal-Wallis 1-way analysis of variance (Siegel 1956: 184-193), or Mann-Whitney U-tests (nocturnal feeders only).

Preferences were determined for foods that comprised ≥1% of the dry weight of esophageal or habitat samples. Food items that were present in esophagi of feeding birds, but not in habitat samples obtained at that site, and foods present but not eaten were not included in preference analyses. Preference was based on mean difference between ranks of components by usage and by availability (Johnson 1980). Formal statistical analyses (Johnson 1980) were not done because the number of foods exceeded the number of ducks in most cases. Therefore, mean rank differences ≥0.5 or ≤ –0.5 were used to roughly indicate limits of consumption of foods in excess and deficit of availability, respectively. Heavy vegetation (stems, leaves, roots, and stolons) was generally not considered potential pintail food and was included in calculating preference in only 1 instance where pintails were feeding exclusively on bulrush seedlings.

Results

Refuge Rice Ponds.—Plant foods accounted for nearly 100% of the diet of pintails feeding in flooded-unharvested rice fields during fall (Table 1); rice seed constituted nearly 94% of this total. During mid-winter, plant foods accounted for nearly 100% of foods eaten in flooded-harvested rice fields (Table 1). However, rice seed was much less important than flatsedge, barnyardgrass, southern naiad, and miscellaneous vegetation. In late winter, animal material, primarily midge larvae (Chironomidae) and snails, increased to 40% of the diet (Table 1). Southern naiad seeds and vegetation also were important.

Summer-dry Seasonal Marsh.—Plant foods comprised 98% of the diet of pintails feeding in these wetlands in fall (Table 2). Swamp timothy seeds made up almost 78% of this total and occurred in nearly 94% of esophagi. Common Bermuda grass seeds, together with miscellaneous vegetation, accounted for 14% of the diet. In mid-winter, plant foods, especially swamp timothy seeds, constituted nearly the entire diet. Other important seeds included Bermuda grass, alkali bulrush (Scirpus paludosus), woolly marbles, dock (Rumex spp.), smartweed, saltbush (Atriplex spp.), and flatsedge. Vegetation, especially stems and leaves of swamp timothy, comprised 21% of the diet. In late winter, plant foods, primarily seeds of swamp timothy and miscellaneous vegetation, made up 71% of the diet. Animal matter increased to 29% of the diet and included primarily larvae, pupae, and adults of beetles (especially Hydrophilidae) and midge larvae.

Summer-irrigated Seasonal Marsh.—Nearly 100% of food consumed during fall was plant material (Table 3); about 75% was barnyardgrass and 21% smartweeds. In mid-winter, seeds of goosefoot (Chenopodium spp.), barnyardgrass, spikerush (Eleocharis spp.), smartweeds, and dock, and miscellaneous vegetation formed 98% of the diet; barnyardgrass predominated. Some important foods, such as goosefoot and dock, were not recorded at feeding sites. During late winter, animal matter increased to 65% of the diet and consisted of dipteran larvae and pupae (especially those of midges), snails, and miscellaneous fragments. Overall, fewer plant (species) and animal (orders) taxa (includes those in tr amounts) were consumed by pintails in this than in other habitat types in fall (9 vs. 21-38), mid-winter (13 vs. 17-34), and late winter (8 vs. 14-28).

Marsh Feeders with Rice.—The diet of 1 male feeding in a permanent pond in August immediately after feeding in refuge rice contained about 32% animal matter (rice portion deleted), predominantly (97%) snails. This is a conservative estimate because most of the plant foods, in addition to rice, could have been obtained in the rice field. During October, animal material, including water boatmen (Hemiptera [Corixidae]), midge larvae, and water beetles, comprised 40% of the diet (summer-dry habitats) (Table 4). October samples contained about 30% bulrush seedlings (misc. vegetation in Table) taken from a shallowly flooded fallow field. Several important foods (barnyardgrass, southern naiad, beetle and midge larvae, and misc. invertebrate fragments) were not found in habitat samples. In late winter, smartweed seeds and miscellaneous vegetation comprised 88% and animal matter, primarily midge larvae, nearly 12% of the diet (summer-irrigated, N = 8 and summer-dry, N = 1). Invertebrates comprised 33.3% of the late winter diet for 7 of 9 ducks in this category; the other 2 ducks had very high percentages (71 and 83%) of smartweed seeds.

Harvested Commercial Rice Fields.—During mid-winter, rice seed was present in all esophagi and accounted for >97% (dry wt) of the diet of nocturnal feeders (Table 5). During late winter, plant material again accounted for >97% of the diet, but barnyardgrass as well as rice was important at this time. Invertebrates (snails and midges) accounted for 3% of the late winter diet.

Food Preferences

Seeds and Vegetation.—Food preferences were exhibited by pintails in all habitat types and seasons (Tables 1-4). Seeds of rice, swamp timothy, and barnyardgrass (foods most characteristic of rice ponds, summer-dry marshes, and summer-irrigated wetlands, respectively), were consumed proportional to availability in fall when these foods were most abundant. In mid- and late winter, these foods were preferred or consumed proportional to availability only when the food comprised a significant portion of the diet. Smartweeds were abundant in summer-irrigated habitats with barnyardgrass, but were underrepresented in the diet until mid-winter when most of the barnyardgrass was exhausted. Southern naiad in rice ponds was preferred or consumed proportional to abundance after fall. Flatsedge seeds were preferred or consumed as available through mid-winter in rice ponds and summer-dry habitats. Bulrush seedlings were consumed proportional to availability by MFWR. Dock, roughseed bulrush, and alkali bulrush seeds never accounted for more than a small proportion of the diet and were underrepresented.

Invertebrates.—Invertebrates consistently were preferred or taken as available in late winter when proportionately abundant in all habitat types (Table 6). Marsh Feeders With Rice preferred or consumed invertebrates as available in all seasons, but the proportion of invertebrates at these sites did not differ from other habitat categories (Table 6). Invertebrate preference ranks were inconclusive in fall and mid-winter because animal material was relatively unavailable and consumed in low amounts.

No significant differences (P > 0.05) between sexes in invertebrate consumption were present in fall or mid-winter in any habitat type. In late winter, animal material accounted for a greater proportion of the diet (P < 0.05) of female than male nocturnal feeders (6%, N = 5 vs. 0.6%, N = 8), but there were no significant differences (P > 0.05) between sexes in refuge habitats.

Nutritional Considerations and Foraging Success.—Preferred foods or those taken as available (e.g., southern naiad, rice, swamp timothy, and bulrush seedlings) contained high AME or protein content (Table 7). In contrast, roughseed bulrush and dock seeds, generally not preferred foods, contained relatively little energy and protein. Barnyardgrass, in fall, contained low AME because awns increased crude fiber; the virtual absence of awns in late winter resulted in a higher AME. Water boatmen and midge larvae contained more protein than other invertebrate groups, but only midges were eaten regularly.

During mid-winter, daily energy intake was acquired primarily during nocturnal feeding in harvested rice fields. This resulted in collection of fewer feeding birds/month ( = 2.8 vs. 8.5-15.0, P < 0.01, Kruskal-Wallis) and collection of the greatest proportion (36 vs. 11-18%, P < 0.01, x²) of actively foraging pintails with empty esophagi during November-January.

Discussion

Numerous seeds and invertebrates were consumed by pintails in fall and mid-winter, but relatively few of these were consumed in large quantity. This resulted because only a few acceptable plant foods were available in quantities large enough to sustain the energy requirements of the large populations. Foods such as smartweeds, flatsedges, southern naiad, and invertebrates were byproducts of management for rice and barnyardgrass, swamp timothy germinated and matured naturally, and bulrush seedlings grew fortuitously after flooding of fallow fields. Enhancement of these foods and reductions in little-used foods (e.g., dock and alkali bulrush) could be achieved by management.

Although marsh seeds were used intensively in fall, rice seed was a dominant factor in pintail foraging routines. For example, in fall, when swamp timothy and barnyardgrass were abundant and consumed in large amounts, pintails regularly made feeding flights from these respective habitats to feed in refuge rice crops (Miller 1985). This use of rice delayed depletion of native marsh seeds and provided a readily available, abundant source of energy. More land would be required to supply this energy with natural foods because native seeds have less AME/gram (except swamp timothy) and less quantity can be produced/unit area. Food supplies in existing marshes in mid- and late winter would not be sufficient to replace the loss of rice seed if rice culture on private land was replaced with nongrain crops. However, marsh foods are important to pintails because rice, an excellent source of energy, has low protein content. Swamp timothy was the most important food in summer-dry habitats, as it was in the San Joaquin Valley (Beam and Gruenhagen 1980, Connelly and Chesemore 1980, Euliss 1984). This food was used by large numbers of pintails as soon as ponds were 1st flooded in August, and pintails often dove to obtain it (Miller 1983). Timothy contains high energy and protein, and it is suitable for many refuges and duck clubs where irrigation water is expensive or difficult to obtain; however, production could be increased with water management.

Barnyardgrass and smartweeds were abundant and occurred together in summer-irrigated habitats in fall. Nutritional values of the seeds were similar, but barnyardgrass was preferred where both occurred together. Barnyardgrass ponds (summer-irrigated) were often 1 m deep, putting benthos out of reach; therefore, fewer taxa were consumed. Barnyardgrass was not always mature when pintails arrived in August, and rice and swamp timothy were important alternatives. Examination of esophagi filled with barnyardgrass in fall revealed that much of the volume consisted of awns. Thus, a full esophagus would have fewer seeds and a lower AME content than if awnless varieties (e.g., junglerice barnyardgrass [Echinochloa colonum]) were available and consumed.

Pintails consumed predominantly (>96%) plant foods during the hunting season in Louisiana (Glasgow and Bardwell 1965, Wills 1972) and through mid-winter in the Sacramento Valley (this study). However, MFWR demonstrated that consumption of high protein sources occurred at certain intervals. Also, animal matter was more important in fall and mid-winter in other areas. For example, clams (Pelecypoda) formed 11% (vol) of the fall diet of pintails on Humboldt Bay (Yocom and Keller 1961). In the San Joaquin Valley invertebrates (primarily midges) formed a small portion of the diet (1-7%) in fall (Beam and Gruenhagen 1980, Connelly and Chesemore 1980), as in the Sacramento Valley, but in mid-winter, animal material increased to about 18% of the diet in 1 study (Beam and Gruenhagen 1980) and 60-85% in another (Connelly and Chesemore 1980). Late winter pintail diets in the San Joaquin Valley, as in the Sacramento Valley, consistently contained a large proportion of animal foods (65-70%). Flooded-harvested rice fields were excellent sources of midge larvae in late winter on the refuges, but these ponds on private lands normally are drained starting the last day of the hunting season (usually mid-Jan). This causes a dramatic decline in the availability of midges, as larvae die within 10 days of drainage (Darby 1962).

Invertebrates are important to nesting pintail pairs, especially females (Krapu 1974a,b). The minimal differences between sexes in invertebrate consumption in late winter suggest that requirements for protein were similar. Paulus (1982) also found no dietary differences between sexes in wintering gadwalls (Anas strepera) in Louisiana. Marked sex-specific changes in invertebrate consumption by pintails must occur after migrants depart the Sacramento Valley.

Preference

Selection of habitat types in which to feed may have been more important to pintails than preference for specific seeds within habitats (Johnson 1980, Pederson and Pederson 1983). Potential for this "higher order" selection (Johnson 1980) is substantial in the Sacramento Valley because many habitat types are present as units of adjacent monocultures. Important seeds, such as rice, swamp timothy, and barnyardgrass were taken proportional to, rather than in excess of, availability at sample sites in early fall when foods were most abundant. This perhaps reflected preference for rice, summer-dry, and summer-irrigated habitat categories and the absence of other suitable foods in these habitats.

Conversely, preference rankings for midges and beetles (invertebrates consumed in greatest quantity) showed positive selection in late winter even though use of invertebrates paralleled abundance in habitat samples (Table 6). This occurred because the weight of these foods in esophageal samples from each habitat category increased by 10-15 × from mid- to late winter (P < 0.05), whereas weight in habitat samples did not change significantly (P > 0.05) (Mann-Whitney U-tests). Thus, preference ranks seemed to correctly reflect active within-habitat selection of midge and beetle larvae. However, food weight in habitat samples may not efficiently measure food abundance because sampling locations were not selected randomly; i.e., they were located where pintails fed (Pederson and Pederson 1983).

Nutritional Considerations

Rice and barnyardgrass seeds may have been consumed primarily to supply energy for weight gain and for foraging on invertebrates, seedlings, and other seeds rich in protein or specific amino acids, analogous to the bioenergetic implications of foraging for invertebrates on breeding grounds (Krapu 1974a). For example, several pintails collected while feeding in rice fields in September had fed heavily on flatsedge seeds (high protein) in addition to the abundant rice. In August and October, MFWR supplemented rice with protein-rich invertebrates or bulrush seedlings obtained on refuges, even though invertebrates formed a small proportion of the diet during most diurnal feeding. In Saskatchewan grain-eating mallards (Anas platyrhynchos), when feeding in wetlands, consumed more animal matter than wetland-feeding mallards that had not eaten grain recently (Sugden and Driver 1980). In late winter, pintails in the Sacramento Valley fed heavily on invertebrates during the day and consumed rice and barnyardgrass at night in harvested rice fields.

Use of foods by pintails was related to documented seasonal changes in body weight (Miller 1986) and behavior (Miller 1985). Weight gains in fall were associated with feeding a large portion of the day on the abundant and diverse foods. Losses in body weight in mid-winter were associated with marked reductions in diurnal feeding activity and less successful feeding; most daily energy intake was acquired during nocturnal feeding in harvested rice fields. In late winter, increased feeding activity and the increased consumption of invertebrates and, hence, dietary protein provided a full complement of amino acids (Krapu and Swanson 1975) and may have contributed to an observed increase in body weight (McLandress and Raveling 1981). Increased dietary protein may be required several weeks prior to rapid growth of reproductive organs (Thompson 1973, Swanson et al. 1974b, Taylor 1978), so late winter consumption of invertebrates may directly affect reproductive performance.

This resource is based on the following source (Northern Prairie Publication 0677):

Miller, Michael R. 1987. Fall and winter foods of northern pintails in the Sacramento Valley, California. Journal of Wildlife Management 51(2): 405-414.

This resource should be cited as:

Miller, Michael R. 1987. Fall and winter foods of northern pintails in the Sacramento Valley, California. Journal of Wildlife Management 51(2): 405-414. Jamestown, ND: Northern Prairie Wildlife Research Center Online. http://www.npwrc.usgs.gov/resource/birds/fwfoods/index.htm (Version 30APR2001).

fwfoods.zip ( 24K ) -- Fall and Winter Foods of Northern Pintails in the Sacramento Valley, California