Botanical Composition of Cattle and Vizcacha Diets in Cen- Tral Argentina

J. Range Manage. 52:370Ð377 Botanical composition of cattle and vizcacha diets in central Argentina

ELIANA E. BONTTI, ROBERTO M. BOO, LILIA I. LINDSTRÖM, AND OMAR R. ELIA

Authors are assistant and fellow, Departamento de Biología and Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET); professor and researcher, Departamento de Agronomía (UNS) and Comisión de Investigaciones Científicas (CIC); assistant, Departamento de Agronomía (UNS); research technician, CONICET. Centro de Recursos Naturales Renovables de la Zona Semiárida (CERZOS). Avenida San Andres 800 (8000) Bahía Blanca, Argentina.

Abstract Resumen

Cattle (Bos taurus) and vizcacha (Lagostomus maximus) Las dietas de vacunos (Bos taurus) y vizcachas (Lagostomus diets were examined monthly in the semiarid Caldenal in cen- maximus) fueron examinadas mensualmente en el Caldenal tral Argentina. Cow-calf operations are the most important semiárido del centro de Argentina. La ganadería es la princi- economic activities within the region. In spite of a widespread pal actividad en la región. A pesar que la vizcacha se encuen- distribution of the vizcacha in Argentina, comparative studies tra ampliamente distribuida en la Argentina, existen pocos of the diet of cattle and vizcacha are scarce. The objective of estudios comparativos de la dieta de vacunos y vizcachas. El this work was to analyze the botanical composition, seasonal objetivo de este trabajo fue analizar la composición botánica, trends, and possible dietary overlap between cattle and viz- tendencias estacionales y posible superposición dietaria entre cacha. Diets were determined by microscopic analysis of cattle vacunos y vizcachas. Las dietas se determinaron por análisis and vizcacha feces collected from November 1994 through microhistológico de heces de vacunos y vizcachas recogidas December 1995 in a shrubland community of the southern desde noviembre de 1994 hasta diciembre de 1995 en una Caldenal. Grasses were the bulk of the diet for both herbi- comunidad arbustiva del sur del Caldenal. Las gramíneas vores. Piptochaetium napostaense (Speg.) Hack. was the most fueron el principal componente de la dieta de ambos her- abundant grass in vizcacha (53%) and cattle (40%) diets. bívoros. Piptochaetium napostaense (Speg.) Hack fue la Prosopis Caldenia Burk. pods partially (34%) replaced this gramínea más abundante en las dietas de vizcachas (53 %) y grass in cattle diets during late summer and fall. Consumption vacunos (40 %). Los frutos de Prosopis caldenia Burk. reem- of P. napostaense was generally higher (13%) in vizcachas plazaron parcialmente (34%) a esta gramínea en la dieta de than in cattle, especially during the dry period of the study vacunos durante el verano tardío y otoño. Las vizcachas gen- (21%). During the drier months, cattle consumed more of the eralmente consumieron más Piptochaetium napostaense (13 less preferred grasses (48%). Forbs were poorly represented %) que los vacunos, especialmente durante el período seco del in the diets perhaps because of scarce rains and low availabili- estudio (21%). Durante los meses más secos los vacunos con- ty. Classification and ordination techniques revealed seasonal sumieron gramíneas menos preferidas (48 %). Las trends and overlapping diets. A greater overlap (75%) was dicotiledóneas herbáceas fueron escasas en la dieta de ambos found during the wet period due to simultaneous consumption herbívoros, quizás por la baja disponibilidad debida a las of P. napostaense by both herbivores. Trends in diet diversity escasas lluvias. Las técnicas de clasificación y ordenamiento were similar with indices generally higher for cattle than for revelaron las tendencias estacionales y la superposición de vizcachas, especially during the dry period. dietas. Una gran superposición de dietas (75%) se encontró durante el período húmedo del estudio debido a que ambos herbívoros consumieron principalmente Piptochaetium napos- Key Words: free ranging cattle, diet composition, semiarid taense. Las tendencias en la diversidad dietaria fueron simi- area, microhistology. lares con índices generalmente más altos para el ganado que para las vizcachas, especialmente durante el período seco. Wild and domestic herbivores directly influence plant communities with their selective grazing. Floristic changes are highly variable and depend on many factors such as preferences of herbivores, the relative abundance and availability of forages, season of the year, and temporal and spatial overlap We would like to thank H. Canessa, F. Zanetti, S. Aman and M. U. García when more than one species grazes the same area. Lorenzana for invaluable assistance. Two anonymous reviewers and the associate The Southern Espinal (Fernández et al. 1989) or District of editor made helpful suggestions which improved a previous version of the manuscript. Research was funded by Consejo Nacional de Investigaciones Científicas y the Caldén (Cabrera 1976), commonly referred to as the Técnicas (CONICET), Universidad Nacional del Sur (UNS), and Comision de Caldenal, is a temperate semiarid region in central Argentina Investigaciones Científicas (CIC). characterized by the presence of Prosopis caldenia Burk. The Manuscript accepted 3 Oct. 1998. landscape in the southern Caldenal is dominated by shrub-

370 Journal of Range Management 52(4), July 1999 lands with a very rich grass layer. Poor site is a Calciustoll with a petrocalcic were analyzed using Student’s t test range management is common through- horizon at an average depth of 0.60 m (Ho: the relative composition of dietary out the region, and soil erosion and plant (Fernández et al. 1989). components were equal in vizcacha and community deterioration are incipient The research area supports a plant cattle feces; n = 10 for vizcacha and n = but generalized (Fernández et al. 1989, community defined as "Unidad C" by 15 for cattle). Bóo and Peláez 1991). Bóo and Peláez (1991). This community Vegetation cover was visually esti- The plains vizcacha (Lagostomus is characterized by a woody species mated with the Braun-Blanquet (1979) maximus, Chinchillidae) is a native cover of 50Ð70 % with Prosopis calde- abundance-cover scale in March 1995 rodent of Paraguay, Bolivia, and nia Burk., P. flexuosa D. C. and (late summer), August 1995 (winter), Argentina that inhabits grasslands and Condalia microphylla Cav. as the most and December 1995 (late spring), and scrub (Branch and Sosa 1994). It is abundant. The herbaceous cover is about we compared diet composition and for- abundant in the Southern Caldenal and 40Ð50% with Piptochaetium napos- age abundance for those months. Cover has been considered a plague in taense (Speg.) Hack. and Stipa tenuis estimation permits a rapid and non- Argentina since 1907 because of its bur- Phil. as the major grasses. Stipa spe- destructive measure of food abundance, rowing and foraging habits. Adult ciosa Trin. et Rupr. and S. gynerioides and estimates were made in and out of females (2.9Ð3.3 kg), adult males Phil. are frequent, and S. clarazii Phil., vizcacha colonies to detect possible dif- (4.5Ð5.6 kg) and young individuals live Poa ligularis Nees and Pappophorum ferences in community composition. in groups of 9 to 14 individuals per mucronulatum Nees are scarce. P. Plant taxa were grouped in 4 classes: colony (Branch et al. 1993). Daily napostaense and S. tenuis are good for- woody species, forbs, preferred grasses intake by adults is 1.7Ð4.7% of body age species that replace Stipa clarazii and non-preferred grasses. This group- weight (Jackson 1985), and areas around Phil and Poa ligularis Nees under mod- ing was made because of the difficulty the vizcacha colonies are characterized erate continuous grazing (Distel and of identifying some taxa to species level by bare soil with scarce, low-growing Bóo 1995). Bare ground (30–40 %) is especially during the winter when vege- forbs and grasses. colonized during wet years by annual tative forms of grasses are similar. Studies of the diet of cattle and viz- forbs, mostly Medicago minima (L.) However, it was important to differenti- cacha in semiarid Argentina are very Grufberg and Erodium cicutarium ate preferred and non-preferred grasses scarce. Giulietti and Jackson (1986) L’Herit. Annual above ground net pri- because diets may be correlated with studied food habits of vizcacha and cat- mary production in this plant communi- forage availability. The woody species tle and found little similarity in their ty for S. tenuis was 862 kg ha-1, and 684 group included Prosopis caldenia, diets. They concluded that potential kg ha-1 for P. napostaense in a year with Prosopis flexuosa, Condalia microphyl- competition was low between both her- 532 mm of rainfall (Distel and la and Ephedra spp. The forb group bivores, but knowledge of the botanical Fernández 1986, Distel 1987). included Erodium cicutarium and composition of vizcacha and cattle diets Fifteen independent fecal samples Medicago minima. The preferred grass- where both herbivores graze a common from different mature cows were col- es included Piptochaetium napostaense, area is much needed. The objective of lected monthly from November 1994 to Stipa tenuis, Pappophorum mucronula- this work was to analyze the botanical December 1995. These samples were tum, Poa ligularis and Stipa clarazii. composition, seasonal variation, and analyzed following Bóo et al. (1991) to The non-preferred grasses included dietary overlap of cattle and vizcacha obtain 20% (P<0.05) precision in the Stipa speciosa, Stipa gynerioides, Stipa diets in a plant community typical of the estimation of the mean for species trichotoma and Stipa tenuissima. southern Caldenal. exceeding 6% of the total sample. On Student’s t test was used to detect possi- each sampling date, fresh fecal pellets ble differences in botanical composition from 10 vizcacha colonies were collect- of the pasture in and out of vizcacha Study Area and Methods ed and analyzed in the same way. This colonies (Ho: plant community compo- sampling intensity gave a 20% (P<0.05) sition was identical in and outside of Cano et al. (1980) have described the precision for species making up at least vizcacha colonies; n = 10 on each site). vegetation of La Pampa where most of 4% of the sample (Bontti et al. 1995). A matrix of the Bray and Curtis the Caldenal occurs. The study was con- Composition of cattle and vizcacha diets (1957) similarity index (BCjk) between ducted in a 600 ha pasture located in was determined by microscopic analysis all possible pairs of average monthly southeast La Pampa (Argentina) at of fecal samples (Sparks and Malechek diets was classified using complete link- 38¡45'S, 63¡45'W with mean elevation 1968). Five slides were prepared from age (Sneath and Sokal 1973, McInnis et of 80 m. Mean annual rainfall is 448 each sample and 40 microscope fields al. 1990, Bóo et al. 1993). A possible mm with about 60% occurring in spring per slide were systematically viewed distortion produced by the classification and fall. Mean annual potential evapo- using 100X magnification. Fragments procedure was evaluated by the calcula- transpiration is 800 mm (Peláez 1986), were identified by comparing fecal tion of the cophenetic correlation coeffi- and the mean annual temperature is material with a reference collection that cient (rxy) following McInnis et al. 15.3¡C with a minimum mean of 7¡C included most of the forages present in (1990), and by ordination of the data (July) and a maximum mean of 23.6¡C the area (Lindström 1994). Epidermal matrix (DECORANA, Hill 1979). (January). Recorded maximum and min- fragments were identified to species Rainfall and temperature data were imum temperatures are 42.5¡C and level except for Ephedra spp. and a few obtained locally with an automatic Ð12.8¡C respectively. Soil at the study grasses and forbs. Differences between Zeeman station. Monthly diversity of cattle and vizcacha diet components

JOURNAL OF RANGE MANAGEMENT 52(4), July 1999 371 cattle and vizcacha diets (N2) were cal- tenuis (Fig. 2B), culated as the reciprocal of Simpson's a high yielding index (Hill 1973). Regression analyses cool season using monthly rainfall and mean tem- grass (Distel peratures on monthly diet diversity for and Fernández both herbivores were performed. 1986), was present in low proportions in the Results diets with a maximum of Cattle diets. 16% in July During the study, rains were scarce 1995. Forbs (388 mm) and erratic, with a relatively were scarce in wet period during the first 6 months fol- cattle diets, and lowed by a pronounced drought and rain they reached a near the end of the study in November maximum of 1995 (Fig.1). Grasses dominated the 7% in Septem- Fig. 1. Monthly rainfall and monthly averages of daily temperatures diets of both herbivores throughout the ber. Medicago from November 1994 through December 1995 in a 600 ha. pasture study. However, diets varied among minima was the in southeast La Pampa, Argentina. months and between herbivores within only forb in cat- the same month. tle diets in appreciable amounts (Fig. of S. tenuis showed the lowest seasonal Cattle relied heavily on Piptochaetium 2F). Prosopis caldenia pods consistently variation of all dietary components, and napostaense, one of the dominant peren- appeared in cattle diets (Fig. 2G). They it was second to P. napostaense as the nial cool-season grasses on the study were abundant from February through most important component of the viz- site, throughout the study (Fig. 2A). May (>30%) with a maximum of 40.5% cacha’s diet. Vizcacha consumption of Monthly diet composition of this grass in February. Other woody plants such as S. tenuis peaked at 26% in November exceeded 65% during the first 3 months Condalia microphylla (Fig. 2H) and 1995 but still ranked below P. napos- of the study. From February to July diet Prosopis flexuosa were very scarce in taense. Other grasses such as Stipa composition still exceeded 35%, but P. the cattle diets. tenuissima contributed less than 1% to napostaense was partially replaced by vizcacha diets. Prosopis caldenia pods (Fig. 2G). For Vizcacha diets. Forbs made a maximum contribution the rest of the study, P. napostaense Grasses made up the majority of viz- (16%) to vizcacha diets in August 1995. made up less than 15% of the cattle cacha diets throughout the study. P. M. minima was the most abundant forb diets, and it was basically replaced by 3 napostaense contributed more than 64% in their diet. Because M. minima was other grasses: Stipa speciosa (Fig. 2C), of the vizcacha diets from November not detected growing in the field during S. gynerioides (Fig. 2E) and S. tri- 1994 through May 1995. From June to our study, we assumed the fruit frag- chotoma (Fig. 2D). These 3 grasses have August vizcacha consumption of P. ments detected in their feces were pro- been given low palatability rankings for napostaense declined and vizcacha con- duced in a previous growing season. cattle by Cano (1988), Bóo and Peláez sumed more S. speciosa, C. microphylla Woody plants were generally scarce in (1991), and Distel and Bóo (1995). Stipa and M. minima. Vizcacha consumption vizcacha diets. The most prominent was C. microphylla (23% in August 1995), and it was detected from June to Table 1. Mean (n = 10) botanical composition (X ± SE) of the pasture by selected plant groups and mean (n = 10 for vizcacha; n = 15 for cattle) botanical composition (X ± SE) of cattle and viz- December 1995. P. caldenia was not cacha diet in March, August and December 1995 in Southeast La Pampa, Argentina. detected in vizcacha diets, and P. flexuosa contributed only trace amounts. Pasture and diet composition Month Preferred Non preferred Comparisons of cattle and viz- grasses grasses Forbs Woody species Bare soil cacha diets. ------(%) ------There were marked similarities and March Pasture composition 32.6 ± 3.0 19.9 ± 3.5 2.9 ± 0.7 27.6 ± 3.8 15.7 ± 2.6 differences among the principal dietary Cattle diet 64.1 ± 3.3 0.7 ± 0.2 0.7 ± 0.2 33.9 ± 3.2 — components of both herbivores (Fig. 2). Vizcacha diet 91.6 ± 1.7 0.7 ± 0.4 5.0 ± 1.5 0.3 ± 0.2 — In December 1994 cattle consumed August — more (p<0.05) P. napostaense than viz- Pasture composition 27.4 ± 2.6 25.3 ± 2.9 1.1 ± 0.3 34.7 ± 5.5 16.5 ± 2.7 cachas. Equal proportions of P. napos- Cattle diet 20.4 ± 2.1 54.9 ± 1.7 6.5 ± 0.9 8.3 ± 1.1 — taense occurred in the diets of both Vizcacha diet 30.9 ± 2.7 23.4 ± 2.1 9.9 ± 2.6 24.1 ± 2.3 — species in November 1994, and January December — and August 1995. For the rest of the Pasture composition 24.0 ± 2.5 22.9 ± 2.3 1.1 ± 0.2 29.2 ± 3.3 17.7 ± 1.6 study, vizcachas consistently consumed Cattle diet 19.6 ± 3.5 56.8 ± 2.9 4.0 ± 0.7 1.9 ± 0.4 — more P. napostaense than cattle. Vizcacha diet 62.3 ± 2.7 13.4 ± 1.2 6.3 ± 1.1 2.2 ± 0.5 — Vizcachas consumed more (p<0.05) S.

372 Journal of Range Management 52(4), July 1999 that hypothesis and the data were pooled. Table 1 summarizes those data and includes cattle and vizcachas diet composition data for the same periods.

Cluster analysis. Seventeen of 19 taxa were present in both herbivore diets, indicating in qualitative terms, a high degree of dietary overlap. However, there were significant differences (p<0.05) in the proportions of components that appeared in the diet of each herbivore. Seasonal differences also occurred. These differences are evi- dent from the groupings generated by the cluster analyses (Fig. 3). By cutting the tree in the range 620.05). subcluster, contains the greatest overlap between the diets of both herbivores, tenuis than cattle during most of the Fruits of M. minima, the only appre- and it encompasses the mid spring to study. Exceptions were November 1994, ciable forb in the diets, were generally mid summer (November and December when cattle consumed more S. tenuis more abundant in vizcacha than in cattle 1994 and January 1995) diets of cattle than vizcachas, and January 1995 when diets. Among the woody plants, P. and vizcachas. This period contained the dietary proportions were equal. In most caldenia pods were found only in cattle greatest consumption of P. napostaense, months when Stipa speciosa was found diets. Leaves of C. microphylla were moderate intake of S. tenuis, and low in cattle and vizcacha diets, its contribu- found from June to December 1995 in intakes of S. speciosa and M. minima tion was higher for cattle than viz- vizcacha diets and in cattle diets only in (Fig. 2F). The second subcluster con- cachas. During March, April, and May, July 1995. tains the mid summer to mid fall Stipa speciosa was a trace component of (February, March, April, and May 1995) both herbivore diets. Other grasses such diets of vizcachas which consisted Plant community composition. almost entirely of P. napostaense and S. as S. Gynerioides and S. trichotoma Although we suspected there were dif- tenuis. Cattle diets were segregated from were detected in moderate (14%) to low ferences in plant community composi- this subcluster into cluster F probably (2%) amounts in the diets of cattle from tion within and adjacent to vizcachas because of the high intake of Prosopis June to December 1995. colonies, our analyses did not support

JOURNAL OF RANGE MANAGEMENT 52(4), July 1999 373 Ordination output of the study, diversity remained low and Ordination output is fairly constant, especially in the case of shown in Figure 4. The vizcachas. There was a sharp increase first 2 axes (DC1 and from May to August, more pronounced DC2) had eigenvalues in cattle than in vizcacha diets. A of 0.48 and 0.05 decrease occurred between August and respectively. The dif- October, another increase occurred in ferent clusters generat- November, and a final decline in ed by the classification December. Low but significant (P<0.05) are also depicted in negative correlations between monthly Figure 4. The samples rainfall and diversity of vizcacha diets (r span a range of 250 = Ð0.53) and cattle diets (r = Ð0.53) units on the first axis were found. There was also a negative indicating a high correlation (P<0.05) between monthly degree of homogeneity mean daily temperature and diversity of in the data. Two sam- vizcacha (r = Ð0.65) and cattle diets (r = ples separated by 400 Ð0.55). Although rains during the study would have no species were lower than average and erratic, in common. This analy- there was a general association between sis suggested consider- months with relatively high rainfall and able overlap occurred high temperatures, with low diversity for the entire study indices. period. Cluster B, C, and F were homogeneous and Discussion well separated on the first axis, whereas The main components of cattle and Clusters A, D and E vizcacha diets were grasses during most were less homogeneous of the study. Cattle preference for grass- and did not show clear es has been reported in many regions separation. The diets of (Johnson 1979, Migongo-Bake and cattle had a wider dis- Hansen 1987, Bóo et al. 1993), and tribution on the axes Giulietti and Jackson (1986) have also than vizcacha diets. reported vizcachas preferences for Fig. 3. Dendrogram of cluster analyses of monthly diets of Cluster F, characterized grasses. Although cattle predominantly vizcacha (VIZ) and cattle (CAT). A, B, C, D, E and F by high proportion of P. consumed grasses, P. Caldenia pods indicate defined clusters. caldenia pods in cattle were an important food source from late diets, was located near summer through the fall. Bóo et al. caldenia pods. the origin of the first axis, whereas clus- (1993) reported similar use of P. Cluster D contains the diets of viz- ter B with a high proportion of non-pre- Caldenia pods during a year with well cacha from June, July, October and ferred grasses in cattle diets was at the above average rainfall, but they detected December 1995. The main difference other extreme. All clusters indexing viz- a lower level of use (23%) than our between cluster D and cluster A is a cacha diets occupied intermediate posi- study (40%). Menvielle and Hernández higher proportion of P. napostaense and tions on the first axis. Diets of cattle, (1985) suggest that P. caldenia may a lower proportion of S. speciosa in D. with the exception of those included in have special significance as a food Cluster E, which is the smallest group, cluster C, were found on the lower half resource for cattle during dry years. By includes only the June and July diets of of the second axis. The diets of vizcacha consuming pods cattle may expedite cattle. Based on the main diet compo- were concentrated on the upper half of seed dispersal and shrub encroachment nents (Fig. 2) it is similar to cluster B the second axis. The ordination con- in the Caldenal (Peláez et al. 1992, Bóo but had a greater proportion of P. napos- firmed our interpretation of the cluster et al. 1993). Pods were not found in viz- taense and S. tenuis, and a lower propor- analyses and also revealed higher level cacha diets, so these rodents are proba- tion of non-preferred grasses. of homogeneity in the diets of vizcacha bly not a dispersal agent of P. caldenia. The cophenetic correlation coefficient, than in cattle. P. napostaense and S. tenuis were pre- an index of possible distortion between sent in cattle diets throughout the study, the resemblance matrix and the classifi- Diversity indices. with a very high consumption made of cation procedure, was found to be rx,y = Diversity indices (Fig. 5) exhibited P. napostaense. Cattle made a shift to 0.86 which is considered acceptable. less preferred grasses in November and ≥ some similar patterns between the 2 her- Values of rx,y 0.80 indicate that distor- bivores. Cattle indices showed more December 1995. A less pronounced but tion is not great (McInnis et al. 1990). variability and reached higher values similar trend was observed for P. napos- than vizcacha. During the first 7 months taense in vizcachas. Vizcachas con-

374 Journal of Range Management 52(4), July 1999 ables like S. gynerioides and S. tenuissima (Distel and Bóo 1995). Forbs were found in low proportions in the diets of both herbivores. Annuals like M. minima and Erodium cicutarium may provide good seasonal forage during spring (Fresnillo-Fedorenko 1990) in the southern Caldenal. Being annuals, they produce numerous seeds (Mayor 1996), and growth is quite variable between dry and wet years. During a wet year, M. minima contributed heavily to cattle diets (Bóo et al. 1993). In our study forb production was low. M. minima pods contributed more heavily to vizcacha than cattle diets, possibly because the pods are difficult for cattle to pick up. Although high diet overlap (90%) occurred between both herbivores in November 1994 through January 1995, exploitative competition cannot be Fig. 4. Spatial distribution of monthly diets defined by the first 2 axes of ordination assessed (Begon et al. 1986). During (DC1 and DC2). Monthly diet of different clusters produced in the classification are: Cluster A: O, Cluster B: g , Cluster C:◊ , Cluster D: u , Cluster E: r and Cluster this period rains occurred and stimulated F: . the growth of preferred grasses. During the dry period of the study, cattle and vizcacha diets diverged (Figs. 3 and 4) sumed more S. tenuis in 1995 than in these species are consumed only when in qualitative and quantitative terms 1994, and they exhibited a less substan- better quality forage becomes scarce (Fig. 2). We assume then that the great- tial shift to the less palatable S. speciosa. (Cano 1975). It is important, in terms of est competition for preferred grasses This was reflected in the classification management and planning, to take into (i.e. P. Napostaense and S. Tenuis) coin- and ordination analyses, where vizcacha account the combined effects of heavy cided with the lower levels (38Ð48%) of and cattle diets of November and grazing by cattle and vizcachas. Under dietary overlap. At that time vizcachas, December 1994 clustered together, but severe grazing conditions some species and especially cattle, were forced to rely their diets were well separated in these (P. napostaense and S. tenuis) are more on the less preferred grasses (i.e. S. same months in 1995. This was probably tolerant of defoliation than others (Stipa Speciosa, S. Gynerioides, and S. due to variations in forage availability, Clarazii and Poa ligularis). Less toler- Tricotoma) that have low nutritive value since plant growth was lower in 1995. ant forages may be replaced by undesir- (Cano 1988). This would probably The drought period that began in April, and continuous heavy grazing, probably caused a decline in available forage and brought about a greater reliance on less preferred forages. Branch et al. (1994) reported that vizcachas relied heavily on P. napostaense throughout the year. We had similar findings for both vizcacha and cattle. The more stable vizcacha diets, compared to cattle, may be related to differences in mouth size and morphology. Cattle may be constrained by bite size and their demands as bulk grazers (Schwartz and Ellis 1981). Size, morphology and grazing behavior may also explain the rapid shift to preferred grasses in vizcachas that was detected in December (Fig. 2) after abundant rains in November 1995 stimulated forage growth. Cattle grazed less preferred grasses Fig. 5. Monthly diversity indices of cattle and vizcacha's diets sampled in a 600 ha pas- during the dry period of the study, and ture in southeast La Pampa from November 1994 through December 1995.

JOURNAL OF RANGE MANAGEMENT 52(4), July 1999 375 affect cattle performance more than viz- high intake of the available, good quali- Bóo, R., L. Lindström, and E. Lutz. 1991. cachas, because the cattle relied more ty species during wet and warm months. Número de preparados por muestra y heavily on the lower quality herbage During the dry months, when the best número de muestras en la estimación de la Holechek et al. (1982) reported an grasses were scarce, both herbivores composición botánica de heces de vacunos. (In Spanish). Rev. Arg. Prod. Anim. increase in the diversity of cattle diets as grazed on less preferred and lower qual- 11:13Ð18. forage availability increased in late ity grasses. Managing stocking rate and Bóo, R.M., L.I., Lindström, O.R Elía, and spring. Because forage biomass depends rodents density should prevent an M.D. Mayor. 1993. Botanical composi- on rainfall (Sala et al. 1988), one would extreme situation of forage depletion. A tion and seasonal trends of cattle diets in expect a positive correlation between low density of vizcachas would also central Argentina. J. Range Manage. diversity of the diets and rainfall. Bóo et increase availability of desirable grasses 46(6): 479Ð482. al. (1993) found no correlation between for cattle during the wet season. Branch, L.C., D. Villarreal, and G.S. diversity and rainfall. Kufner and Prosopis caldenia pods were the only Fowler. 1993. Recruitment, dispersal, and group fusion in a declining population of Pellizza de Sbriller (1987) found a high- contribution by woody species to cattle the plains vizcacha (Lagostomus maximus, er trophic diversity in cattle and mara diets, and they were not detected in viz- Chinchillidae). J. Mamm. 74(1): 9Ð20. (Dolichotis patagonum) diets in the arid cacha feces. While pods could potential- Branch, L.C, and R.A. Sosa. 1994. plains of eastern Mendoza during sum- ly be a good source of food for cattle Foraging behavior of the plains vizcacha, mer, which was considered the lean sea- during the fall, the possibility of seed Lagostomus maximus (Rodentia: son. In our study the warmest months dispersal and P. caldenia encroachment Chinchillidae), in semi-arid scrub of cen- coincided with the highest rainfall. represents a serious risk. tral Argentina. Vida Silvestre Neotropical. Consequently, there was more forage Dietary overlap was highest during 3(2):96Ð99. Branch, L.C., D. Villarreal, A. Pelliza available than in the coldest and driest wet months when Piptochaetium napos- Sbriller, and R.A. Sosa. 1994. Diet selec- months. The negative correlation found taense was consumed in high proportion of plains vizcacha (Lagostomus max- among mean monthly daily tempera- tions by both herbivores. Dietary over- imus, Family Chinchillidae) in relation to tures, monthly rainfall and botanical lap was lower during the fall than during resource abundance in semi-arid scrub. diversity of the diets indicated that summer because cattle relied on Can. J. Zool. 72: 2210Ð2216. diversity increased as available forage Prosopis caldenia pods. The lowest Braun-Blanquet, J. 1979. Fitosociología. decreased. When favorable conditions overlap was observed during the driest (In Spanish). Ed. H. Blume. 820 pp. for plant growth occurred, vizcachas and months when cattle consumed more of Bray, J.R., and Curtis, J.T. 1957. An ordi- cattle consumed mainly preferred grass- the less preferred grasses than vizcacha. nation of the upland forest communities of Southern Wisconsin. Ecol. Monog. es. Under unfavorable conditions their In this study, diversity indices were 27:325Ð349. diets diverged because cattle shifted to higher in cattle than in vizcacha diets. Cabrera, A.L. 1976. Regiones fitogeográfi- other resources. During the drier months, diversity of cas argentinas. In: Enciclopedia argentina In a study on vizcacha diets Branch et both herbivore diets was higher, because de agricultura y jardinería. (In spanish) al. (1994) found a decrease in diet they consumed more of the less pre- ACME, Buenos Aires. breadth concomitant with scarce ferred species. The combined effect of Cano, E. 1975. Pastizales de la región cen- resources in the field. A decrease in heavy grazing by cattle and vizcachas tral de la provincia de La Pampa. (In span- selectivity occurred when annual forbs should be prevented to avoid displacing ish) IDIA 331Ð333:1. Cano, E. 1988. Pastizales naturales de La were abundant. In our study, higher high quality forages with undesirable Pampa, tomo 1. Convenio AACREA - diversity indices occurred during the grasses or woody vegetation. These Provincia de La Pampa. (In spanish). drier months. This may have been due to shifts in community composition would Cano, E., B. Fernández, and M.A. Montes. a less diverse plant community on our be difficult to reverse. 1980. Región meridional (vegetación) (In study site than on the natural reserve Spanish). p. 359Ð443. In: Prov. De La where Branch’s research was conducted. Pampa, and Univ. Nac. de La Pampa (ed.). Cattle and vizcachas have overgrazed Literature Cited Inventario Integrado de Los Recursos our study site for years, and highly pre- Naturales de la Provincia de La Pampa. Buenos Aires, Argentina. ferred grasses like S. clarazii and P. Begon, M., J.L. Harper, and C.R. Distel, R.A. 1987. Crecimiento aéreo y radi- ligularis are scarce. Monthly diversity Townsend. 1986. Ecology. Individuals, cal, germinación y supervivencia en was generally lower for vizcacha than Populations and Communities. Blackwell, Piptochaetium napostaense (Speg.) Hack for cattle diets, especially during most Oxford. y Stipa tenuis Phil (In Spanish). M. S. of the driest months. Vizcachas also Bontti, E., R. Bóo, and L. Lindström. Thesis, Univ. Nac. del Sur, Bahía Blanca, consistently consumed more of the pre- 1995. Número de preparados por muestra Argentina. y número de muestras en la estimación de Distel, R.A., and R.M. Bóo. 1995. ferred grasses and M. minima fruits than la composición botánica de heces de viz- cattle. We speculate that vizcacha Vegetation states and transitions in tem- cacha. (In spanish). Memorias de la XVII perate semiarid rangelands of Argentina. require a higher quality diet than cattle. Reunión Argentina de Ecología. Mar del Proceedings of the Vth International Plata. Rangeland Congress, Salt Lake City, Utah, Bóo, R.M. and D.V. Peláez. 1991. USA p. 117. Conclusions Ordenamiento y clasificación de la veg- Distel, R.A. and Fernández, O.A. 1986. etación en un área del sur del Distrito del Productivity of Stipa tenuis Phil. and This study indicates that both herbi- Caldén. (In spanish). Bol. Soc. Argent. Piptochaetium napostaense (Speg.) Hack vores relied heavily on grasses, with Bot., 27:135Ð141. in semi-arid Argentina. J. Arid Environments. 11:93-96.

376 Journal of Range Management 52(4), July 1999 Fernández, O., R. Bóo, and L. Sánchez. Jackson, J.E. 1985. Ingestión voluntaria y Menvielle, E.E. and O.A. Hernández. 1989. South American shrublands. In: digestibilidad en la vizcacha (Lagostomus 1985. El valor nutritivo de las vainas de McKell, C. (Ed.). The biology and utiliza- maximus). (In spanish). Rev. Arg. Prod. Caldén (In Spanish). Rev. Arg. Prod. tion of shrubs. Academic Press, USA. Anim. 5 (1-2): 113Ð119. Anim. 5: 435-439. Fresnillo-Fedorenko, D.E. 1990. Johnson, M.K. 1979. Foods of primary con- Migongo-Bake, W. and R. Hansen. 1987. Estrategias ecológicas de Medicago mini- sumers on cold desert shrub-steppe of Seasonal diets of camels, cattle, sheep, and ma (L.) Grufb. var minima y Erodium south-central Idaho. J. Range Manage. goats in a common range in eastern Africa. cicutarium (L.) L’Herit, dos anuales de 32(5):365Ð368. J. Range Manage. 40:76Ð79. valor forrajero en el Caldenal (In Spanish). Kufner, M.B. and A. Pelliza de Sbriller. Peláez, D.V. 1986. Análisis de algunos factores M. S. Thesis, Univ. Nac. del Sur, Bahía 1987. Composición botánica de la dieta ambientales y morfológicos y su relación con Blanca, Argentina. del mara (Dolichotis patagonum) y del la aplicación de herbicidas en cinco especies Gauch, H.G. 1982. Multivariate analysis in ganado bovino en el monte mendocino. (In arbustivas del distrito fitogeográfico del Caldén (In Spanish). M.S. Thesis, Univ. Nac. community ecology. Cambridge Univ. spanish) Rev. Arg. Prod. Anim. del Sur, BahIa Blanca, Argentina. Press 7(3):255Ð264. Giulietti, J.D. and J.E. Jackson. 1986. Peláez, D.V., R.M. Bóo, and O.R. Elia. Lindström, L.I. 1994. Estudio microhis- 1992. Emergence and seedling survival of Composición anual de la dieta de vizcacha tológico de las especies de gramíneas que caldén in the semiarid region of Argentina. (Lagostomus maximus) en pastizales natu- integran la dieta de vacunos en el caldenal: J. Range Manage. 45:564Ð568. rales en la provincia de San Luis, Elaboración de una clave y su aplicación Sala, O.E., W.J. Parton, L.A. Joyce, and Argentina. (In spanish). Rev. Arg. Prod. al análisis de muestras fecales (In W.K. Lauenroth. 1988. Primary produc- Anim. 6 (3-4):229Ð237. Spanish). M. S. Thesis, Univ. Nac. del tion of the central grassland region of the Hill, M.O. 1973. Diversity and evenness: An Sur, Bahía Blanca, Argentina. United States. Ecol. 69:40Ð45. unifiyng notation and its consequences. Mayor, M.D. 1996. Banco de semillas de Schwartz, C.C. and J.E. Ellis. 1981. Feeding Ecol. 54:427Ð432. un pastizal-arbustal natural del sudeste de ecology and niche separation some native Hill, M.O. 1979. DECORANA. A fortran La Pampa, su variación estacional y and domestic ungulates on the shortgrass program for detrended correspondence relación con la vegetación existente (In prairie. J. of Appl. Ecol., 18, 343Ð353. analysis and reciprocal averaging. Cornell Spanish). M.S. Thesis, Univ. Nac. del Sur, Sneath, P. and R. Sokal. 1973. Numerical Univ., Ithaca, N.Y. Bahía Blanca, Argentina. taxonomy. W. H. Freeman and Co., USA. Holechek, J., M. Vavra, J. Skovlin, and McInnis, M.L., L.L. Larson, and M. Sparks, D.L. and J.C Malechek. 1968. W.C. Krueger. 1982. Cattle diets in the Vavra. 1990. Classifying herbivore diets Estimating percentage dry weight in diets Blue Mountains of Oregon, I. Grasslands. using hierarchical cluster analysis. J. using a microscopic technique. J. Range. J. Range Manage. 35:109Ð112. Range Manage. 43:271Ð274. Manage. 21 (4):264-265.

JOURNAL OF RANGE MANAGEMENT 52(4), July 1999 377