Soil Ecology Restoration Group

last update February 1, 2000

Restoration of Quercus agrifolia and Nassella pulchra in an oak savannah in the Cuyamaca Mountains

Tree shelters improve growth and seed set for Nassella pulchra

There is a general, though not unanimous, agreement that grasslands were once more common in many areas of California including San Diego (Burcham, 1957; Crampton, 1973; Bartolome and Gemmil, 1981). The accounts of the early Spanish travelers, particularly Fr. Juan Crespi and Fr. Francisco Palou, highlight the abundant good pasture near San Diego (Bolton, 1971; Englehardt, 1920; Palou, 1988). Quoting Crespi as he headed north from San Diego in July 1769, "We ascended a large grassy hill.... and found ourselves on some broad mesas... all covered with grass... except here and there some very small oaks and chaparral (Bolton, 1971). In addition the Kumeyaay remember a now extinct domesticated large seeded grass that was an important food source (Shipek, 1989). As Fr. Palou noted in his memoirs, "The heathen live on grass seeds which they harvest in their season and make into sheaves as is usually done with wheat..." (Palou, 1988).

The dominant grass in many of these grasslands may have been purple needlegrass, Nassella pulchra (formerly Stipa pulchra), but disturbance was so complete before botanists arrived that it may simply be the bunchgrass that was best adapted to survive disturbance (Bartolome and Gemmill, 1981). Purple needlegrass is favored by frequent fire (Biswell, 1956; Langstroth, 1991), which may have made it a camp follower for the Native Californians who used fire extensively and intensively (Blackburn and Anderson, 1993). Many other grasses, flowers and forbs were also found in these complex grasslands (Burcham, 1957).

The conversation from perennial native grasses to weedy annuals was clearly driven by several types of disturbance and fire suppression (Burcham, 1957; Adams, 1964). In many cases overgrazing for more than 200 years is implicated, because even short period low intensity grazing can have a major impact on fragile arid ecosystems. Fire suppression and/or increased fire frequency have also been blamed for dramatic changes in these native plant communities (Minnich and Dezzani, 1998; Burcham, 1957). Air pollution and nitrogen deposition have been suggested as causes of declines in native ecosystems in southern California (Bainbridge, 1997; Padgett et al., 1999). Agricultural operations have also played a role in the decline of the native grasslands, particularly the widespread but often now forgotten dry farming near the turn of the century. Purple needlegrass was apparently particularly abundant in bottomlands, the first areas converted to farmland (Adams, 1964).

Only scattered remnants of semi-healthy grasslands can be found in Southern California, most notably on the Santa Rosa Plateau Preserve and Camp Pendleton.

For restoration purposes it is desirable to return native grasses to these degraded ecosystems (Amme, 1991; Bugg et al., 1997; Adams et al., 1999). They provide important food and cover for animals, birds (seeds), and butterflies (when in bloom) and reduce soil erosion. They may remain green and growing late into the summer, rather than dying and drying out in May, reducing fire hazard in the mid summer.

These grasses can be returned to disturbed areas by direct seeding or container planting. (Anderson, 1993; Bugg et al., 1997). Direct seeding methods have now been developed, but it requires considerable seed, herbicide and persistent management. Establishing plants from containers is faster and more certain than direct seeding in arid areas (Bainbridge et al., 1995). It also requires less seed, and this can critical because local seed sources are often restricted. Use of commercial cultivars of N. Pulchra may not be desirable, because of genetic and quantitative trait differences have been identified (Knapp and Rice, 1998; Adams et al., 1999). Local ecotypes are more likely to be adapted to site climate and soils (Lippit et al., 1995).

The best approaches for planting native grass from containers have yet to be determined. In some situations simply ripping and or scraping to reduce compaction and competition may surface (Delman et al., in review). But other innovations may also be required. In 1997 we initiated a test of a new strategy for improving growth from plants grown in plant beds and outplanted in a degraded site at United States International University in San Diego on the eastern boundary of the mesas.

Materials and Methods

N. pulchra seed was purchased from S&S Seeds. It was sent to the CDF Seed Lab in Davis, California for cleaning and upgrading and production of seedlings in 5 x 5 x 15 cm plant bands. In the spring of 1997 matched seedlings were planted in a random arrangement, half with and half without Treepee tree shelters in a degraded eucalyptus plot. These were watered occasionally but equally during the first season, watered twice in 1998, and ignored thereafter. In June 1999 the plants were relocated and survival and height were measured for 20 tree sheltered and 20 unsheltered plants. Seed set of a representative tree sheltered plant was also counted.


Survival was excellent for both cohorts; but the differences in height were striking. The tree sheltered plants were tall (mean height 118.8 cm) and covered with seeds while unsheltered plants were healthy but short (mean height 11.3 cm) from intensive browsing by cottontail rabbits. This is highly significant with a p-value of 0.0001 using Fisher's Protected LSD (SuperAnova). All of the tree sheltered plants set seed, while no seed was set on the unsheltered plants. More than 600 seeds were counted on a representative treesheltered plant.

Figure 6.1-1. Photo of a representative seed head from treesheltered Nassella pulchra


Although most previous research has examined the effects of fire and competition on growth of N. Pulchra it appears that herbivory by cottontail rabbits is limiting growth and seed set at this location. Protection from herbivory releases the grass from this pressure. Perhaps one of the benefits of fire is reduced herbivory. It is possible the intensive use of rabbits and hares as food by the Native people may have provided similar benefits near camps. The mean height of the treesheltered plants was almost double that found in native stands by Ahmed (1981). The height of the three year old unsheltered, heavily browsed, plants was comparable to the first year height of natural germinants in another study (Fossum, 1990).

The benefits may also reflect the improved microenvironment in the tree shelter (Bainbridge, 1994). Ahmed (1981) discovered the roots of N. pulchra were more vigorous than roots of Bromus mollis, growing 78% faster in coarse soil and 53% faster in fine textured soil. Although this should give N. pulchra an edge in the competition for water, root growth in N. pulchra usually begins weeks or months after the exotics, perhaps until early Spring the treeshelter may increase the growth and competitiveness of N. pulchra when soil moisture and nutrients are readily available.

Planted at at density of only 2,000 plants per hectare with treeshelters a grassland could produce more than a million seeds per hectare to fill adjacent areas. In this initial trial the benefits of treeshelters were striking and shelters for at least a percentage of out planted grass seedlings may be recommended.


Adams, M.S. 1964. Ecology of Stipa pulchra with a special reference to certain soil characteristics. MS Thesis, Range Management, UC Davis, Davis, CA 76 p.

Adams, T.E., C.E. Vaughn and P.B. Sands. 1999. Geographic races may exist among perennial grasses. California Agriculture 53(2): 33-38.

Ahmed, E.O. 1981. Fire ecology of Stipa pulchra in California. PhD Thesis, Ecology. UC Davis, Davis, CA 64 p.

Amme, D. 1991. Working with native perennial grasses. Grasslands 1(1):1-3.

Anderson, J. 1993. Strategies for establishing native grasses. Grasslands 3(1):1-3.

Bainbridge, D.A. 1994. Tree shelters improve establishment on dry sites. Treeplanters notes. Winter 45(1): 13-16.

Bainbridge, DA 1997. The nitrogen pollution problem. Newsletter of the Society for Ecological Restoration, California Section. 7(3):3-4.

Bainbridge, DA, M. Fidelibus and R. MacAller. 1995. Techniques for plant establishment in arid ecosystems. Restoration and Management Notes 13(2):198-202.

Bartolome, J.W. and B. Gemmill. 1981. The ecological status of Stipa pulchra in California. Madrono 28(3): 172-184.

Biswell, H.H. 1956. Ecology of California grasslands. Journal of Range Management. 9:19-24.

Blackburn and Anderson, 1993. Before the Wilderness. Ballena Press. Menlo Park, CA 476 p.

Bolton, H.E. 1971 [1921] Fray Juan Crespi. AMS Press, NY 402 p.

Bugg, R.L., C.S. Brown, and J.H. Anderson. 1997. Restoring native perrenial grasses to rural roadsides in the Sacramento Valley of California: establishment and evaluation. Restoration Ecology 5(3):214-228.

Burcham, L.T. 1957. California Rangeland: An Historico-Ecological Study of the Range Resource in California. California Division of Forestry, Sacramento. 260 p.

Crampton, B. 1974. Grasses in California. UC Press, Berkeley. 178 p.

Delman, S., DA Bainbridge, S. Rood. 1999. Revegetation of Nassell pulchra in a Cuyamaca Mountain oak savannah. In review.

Englehardt, Fr. Z. 1920. San Diego Mission. J.H. Barry, San Francisco. 358 p.

Fossum, H.C. 1990. Effects of prescribed burning and grazing on Stipa pulchra seedling emergence and survival. MS Thesis, Ecology, UC Davis, Davis, CA 69 p.

Knapp, E.E and K.J. Rice. 1998. Comparison of isozymes and quantitative traits for evaluating patterns of genetic variation in purple needlegrass (Nassella pulchra). Conservation Biology 12:1031-1041.

Langstroth, R.P. 1991. Fire and grazing ecology of Stipa pulchra grassland. A field study at Jepson Prarie. MS. Thesis, Range and Wildland Science, UC Davis, Davis, CA 76 p.

Lippit, L., M. Fidelibus and DA Bainbridge. 1994. Native Seed collection, processing and storage. Restoration Ecology 2(2): 120-131.

Minnich, R.A. And R.J. Dezzani. 1998. Historical decline of Coastal Sage Scrub in the Riverside-Perris Plain. Western Birds 29:366-391.

Padgett, P.E., E.B. Allen, A. Bytnerowicz, and R.A. Minnich. 1999. Changes in soil inorganic nitrogen as related to atmospheric nitrogenous pollutants in Southern California. Atmospheric Environment., 33:769-781.

Palou, Fr. F. 1998 [1774]. A Spaniard explores the southern California Landscape, 1774. Pp. 67-73. In C. Merchant, ed. Green Versus Gold. Island Press, Washington DC.

Shipek, F.C. 1989. An example of intensive plant husbandry: the Kumeyaay of southern California. Pp. 159-170. In D.R. Harris and G.C. Hillman, eds. Foraging and Farming. Unwin-Hyman, London.

Revegetation of Nassella pulchra in a Cuyamaca Mountain Oak Savanna

In recent years severe grazing by domestic animals, agriculture, fire suppression and urban development have eliminated and changed the oak savanna landscape of southern California. Nassella pulchra (formerly Stipa pulchra), purple needlegrass, may once have been a dominant plant of California grasslands before the invasion of Eurasian annuals (Burcham, 1957). In less disturbed areas, N. pulchra may make up to 90% of the stand in perennial grasslands (Adams, 1964). In a study at Hastings Reserve N. pulchra at 11% of cover, made up 37% of the standing crop (White, 1967). But today introduced annual grasses and weeds dominate disturbed grasslands in both density and cover, even in fields which have been protected from domestic animal grazing for more than 30 years.

While many native species have survived the onslaught of these exotic invaders, the native species have dropped from a major component of these systems to rare or isolated patches. N. Pulchra is favored by livestock and was virtually grazed out on many rangelands; but it may be the most common native grass today because it tolerates disturbance better than the other native grasses (Bartolome and Gemmill, 1981). It requires mineral soil (lack of litter) to establish, and germinates later in the season than the introduced exotics, Avena and Bromus (Fossum, 1990); Langstroth, 1991). Consequently, N. pulchra seedlings must contend with reduced nutrients, moisture and light and rarely get established. Increased competition in disturbed habitats, fire suppression (which favors exotic annual grasses), and a loss of habitat have led to a continuing decline in N. pulchra over the years (Bartolome and Gemmill, 1981; Ahmed, 1983; Langstroth, 1991). This project explored techniques for restoring oak savanna grassland communities by outplanting N. pulchra in a highly disturbed area.

The Study Site

The study site is a Caltrans ecological preserve at the western base of Cuyamace Peak in San Diego County that had been used as grazing land for perhaps 150 years. The preserve is located on a 20-30 degree north facing slope of a rounded ridge, with a sandy loam soil and approximately 75% grassland cover and 25% canopy cover of Quercus agrifolia and Quercus englemanii trees. Julian, elevation 1,300 m (4240 feet), the nearest town, about 10 miles to the north-east, receives an average of 66 cm (26 inches) of precipitation per year (Western Regional Climate Center, 1998). The elevation at the ecological preserve is 1,037 m (3,400 feet). Based on the difference in elevation, exposure and a map developed by Minnich and Chou (1997) the precipitation at the preserve is probably only slightly less than at Julian.

A four foot metal posed barbed wire fence has been in place for about four years and domestic animal grazing (cattle) has been eliminated, while deer grazing may have been slightly reduced. The fence, however, does not keep out the gophers, as evidenced by gopher tailings, which cover 5-10% of the area.

Materials and Methods

N. Pulchra seeds were purchased and then shipped to Davis, California where they were germinated and grown in 5 x 5 x 15 cm inch plant bands at the California Department of Forestry L.A. Moran Reforestation Center. Once the seedlings reached a height of about 15 cm with 5-10 tillers, the N. pulchra plant bands were sent to the Soil Ecology and Restoration Group at San Diego State University. On 13-14 December 1997, one thousand seedlings were planted at the ecological preserve.

A Ford 1215 T.L.S. tractor was used to prepare the site for planting. A total of thirty two plots were scraped clear in the open grassland to a depth of 5-10 cm. This removed nearly all existing vegetation. Sixteen 1.2 x 1.2 m foot plots and sixteen 1.8 x 1.8 m plots were established. The different plot sizes were cleared to see if the larger plots had an edge effect, making it harder for the exotic annual grasses to reach the middle of the plot. In eight of the smaller plots, N. pulchra was planted in a grid pattern at a density of one plant per 0.009 m2. In the other eight plots N. Pulchra was planted in a grid pattern at a density of one plant every 0.18 m2. The same density patterns were repeated for the larger scraped plots.

Similar plot sizes and densities were grouped into fours and subjected to nutrient and/or organic matter treatments. One plot served as a control; the second plot received a commercial fertilizer application (Scott's Winterizer Lawn Fertilizer, a slow release fertilizer with a nutrient content of 22-4-14); a third plot received mulch (Vigora Compost, which has a nitrogen content of 0.5%); and a fourth plot received both fertilizer and mulch. The fertilizer was applied at a rate of 0.17 kg per 10m2 and the mulch was applied 0.6 cm deep. There were two replicates for each independent variable. Overall, 792 N. pulchra seedlings were planted in the test plots.

The tractor scarifiers were then used to rip the soil to a depth of 16-18 cm down the slope in the open areas. We then planted 208. N. pulchra seedlings into the ripped slots without the addition of fertilizer or mulch. Since disturbance was minimal to surrounding vegetation, this was in effect and open field planting that could be compared to the control plots of the cleared areas. No supplemental water was applied to any of the transplanted grasses.


Data was collected on 6 May 1998, five months after planting, and on 17 July 1999, nineteen months after planting. A three way ANOVA was used to test the dependent variables of percent cover N. pulchra, percent cover other species (including both the native and non- native grassland plant species), percent cover gopher tailings, height of leaf blades (cm), percent N pulchra plants producing seeds, and percent N. pulchra survival.

After five months N. pulchra survival average 82%. Plot size had no significant effect on any of the dependent variable (p>0.05). Density was significant in relation to percent cover N. pulchra (p=0.003), with 35 percent cover in the high density plots and 18 percent in the low density plots. The density plots did not have an affect on the over dependent variables (p>0.05). Soil amendment treatments did not have a significant effect on percent cover of N. pulchra, percent gopher cover or percent N. pulchra survivorship (p<0.05). The treatments did, however have significant effect on percent other species cover (p=0.003), this was highest in the fertilizer + mulch plots at 36 percent and lowest in the control plots at 14 percent, percent N. pulchra producing seed (p=0.002), highest in the mulch plots at 63 percent and lowest in the control plots at 14 percent), and N. pulchra height (p=0.002), highest in the mulch plots at 34.1 and lowest in the control plots at 22.1 cm. There were no differences of dependent variable between the amended plots (p>0.05), but all the amended plots had higher values than the control plots (p<0.05). Although not affected by the independent variables, gopher activity averaged 6%. The bunchgrass seedlings planted in the groves created by the ripping bars on the tractor were not significantly different (p>0.05) than the control plots in height, percent of plants producing seeds or percent survival.

After nineteen months overall survival of N. Pulchra had dropped to 41%. The data collected at nineteen months differed from the data collected on May 6, 1998. Plot size was deemed an inappropriate independent variable due to the relatively small difference in size of the plots. Percent cover of other species data was not collected again because many of the plants had dried up, making judgments of percent cover difficult to measure. Also, percent of N. Pulchra plants producing seeds was not tallied because after nineteen months, nearly all of the N. Pulchra plants still alive had produced seeds. For the nineteen month observations, the density treatment still had a significantly higher (p=0.006) percent cover of N. Pulchra in the high density plots (24 percent) than in the low density plots ( 7 percent). The low density plots had N. Pulchra plants averaging 24.2 cm tall, significantly taller (p=0.0002) than in the high density plots, 21.1 cm. Density treatments did not have an effect (p>0.05) on the other dependent variables, percent cover gopher tailings and percent N. Pulchra survival. Soil amendment treatments still did not have a significant effect on percent cover of N. Pulchra (p>0.05). The treatments did, however, have a significant effect on N. pulchra height (p=0.03) with the mulch plots tallest, 23.1 cm, and shortest in the control, 20.0 cm; percent N. Pulchra survivorship (p=0.0001), with survival in the control plots highest at 61 percent and the fertilizer+mulch plots lowest at 23 percent; and percent gopher cover (p=0.02) with the fertilizer + mulch plots highest at 29 percent and the control plots lowest at 3 percent. Overall gopher activity averaged 17 percent, an 11 percent increase since the five month monitoring. The bunchgrass seedlings planted in the grooves create by the ripping bars on the tractor were significantly different (p<0.05) than the control plots in height (20.0 cm for control plots and 17.6 cm for plants in the grooves) and percent survival (61 percent for control plots and 29 percent for plants in the grooves.)


N. pulchra is thought by many to have been one of the dominant grassland species in California before overgrazing, fire suppression, and the invasion of exotic annuals and forbs. By clearing small plots of land and planting perennial bunchgrass seedlings, it was thought that a decrease in competition for nutrients and moisture might enable N. pulchra to become established, and once established, N. pulchra has a tendency to persist. Different plots sizes were compared to see if the larger plots would have a beneficial edge effect; but the dispersal range of the other species, exotic forts and annual grasses, as well as the native annual forbs, was greater than both plot sizes; and plot size had no effect on percent cover N. pulchra percent cover other species, height of N. pulchra, percent survival of N. pulchra, percent of N. pulchra producing seeds and percent cover of gopher tailings. Although plot size did not matter in this experiment, it would probably be better to cultivate small plots or horizontal strips than large plots because it would be less disruptive to the native annual forb seed bank, as well as minimizing the risk of erosion. Dispersal of N. pulchra seeds generally does not exceed a distance of one meter from the maternal plant (Stromberg and Griffin, 1996). A larger number of smaller plots over a specified area would thus have the potential to fill in a restoration site more quickly.

A previous study (Eliason, 1995) showed that the native plants in low density plots grow larger than in the high density exotic grass plots, but in this experiment similar results were not evident until the nineteenth month of monitoring, when the plants in the low density plots were taller than the plants in the high density plots. Differences in planting density also had a significant effect on percent cover of N. pulchra in this study at both the five and nineteen month monitorings, with the high density plots having a greater percent cover. It was though that an increase in percent cover of N. pulchra would decrease the percent cover of other species, thereby increasing the survival for N. pulchra; but this was not the case at the five month monitoring. If the percent cover goal for N. pulchra in an undisturbed grassland area is a minimum of 10%, then at what density do you originally plant? In this experiment, N. pulchra cover averaged 27 percent at five months and 16 percent at nineteen months. Assuming all N. pulchra that are still alive have now been in the ground long enough to be considered mature, then the survival rate is likely to decline slowly. With percent cover of N. pulchra averaging 7 percent in the low density plots and 25 percent in the high density plots, then in may be unnecessary to plant more densely than a spacing of 0.2 m x 0.2 m.

The addition of nutrients and/or organic matter seemed to have the biggest effect on the dependent factors. While it did not have an impact after nineteen months on percent cover of N. pulchra, it did significantly effect percent cover of other species (at five months), height of N. pulchra, percent cover of gopher tailings, percent survival of N. pulchra and percent of N. pulchra producing seed. At the five month monitoring, percent cover of other species increased with the addition of nutrients and/or organic matter. It was thought than and increase in size of other species might help them outcompete N. pulchra and affect its survival. At five months this was not the case, however, at nineteen months a significantly lower average of 35 percent of N. pulchra survived in the amended plots compared to 61 in the early stages, but they had trouble in the long run. Still, mature N. pulchra is more resistant to competition than seedlings. It is estimated that only 1 percent of N. pulchra seeds make it to a mature plant (Stromberg and Griffin, 1996), while in this experiment 41% of the semi-mature transplants made it to a mature plant. This project suggests that for many restoration projects, it may be advantageous to plant bunchgrass.

Perhaps the most important indicator of plant establishment is the production of seeds. The addition of nutrients and/or organic matter significantly increased seed production at the five month monitoring. Sixty percent of the soil amendment treated plants were producing seeds, compared to only 15% of the control plants. More seeds will increase chances N. pulchra further colonizing the site. At both the five and nineteen month monitorings, plant height was significantly taller in the soil amended plots, another indicator that N. pulchra matures more quickly with higher nutrient levels.

The relation between planting success, treatments, and pocket gophers was also of interest. The modern day gopher diet in disturbed California grasslands consist of exotics such as Bromus, Avena, Vulpia, Erodium, and Hypochaeris (Stromberg and Griffin, 1996). Pocket gophers (Thomomys bottae) are active as a group 24 hours a day, 365 days a year (Grinnell, 1923). Their tunnels are 10-20 cm deep as they search for plant roots to feed on. Gophers spend 99% of their time underground, but will at times surface to eat grass and annual plant leaves. Excess dirt is removed from the tunnels and left as piles of dirt (tailings) on the surface of the ground, on the average of 6 cm high by 25 cm wide. At the nineteen month monitoring, percent cover of gopher tailings was not affected by plot size or density, but activity was significantly higher in the soil amended plots (23 percent tailing cover) than in the control plots (3 percent tailing cover). Mean gopher mound coverage for all plots was 17 percent.

It has been reported that gophers have tendency to dig in soils low in nitrogen (Stromberg and Griffin, 1996). In this experiment, however, this was not the case. Results suggest the gophers targeted the soil amended plots. The gophers did not occur more frequently in the high density plots. This suggests the gophers were targeting those plots where N. pulchra and other plants grew larger, greener and more tender and perhaps had increased nitrogen leaf content. At the five month monitoring there was a trend in percent cover gopher tailings to percent cover other species (p=0.058). After nineteen months, gophers could have been eating more exotic plants, with damage occurring to N. pulchra because they were nearby. There are still questions surrounding the growth and competition of native versus nonnative plants in gopher areas. In an early study by Stromberg and Griffin (1996), exotics outcompeted N. pulchra in germinating and growing on the gopher tailings, but Edwards (1995) found there was an increased density of native annuals in gopher areas.

Based on this experiment, the recommendation from planting N. pulchra in disturbed areas is to scrape plots at least 1.2 x 1.2m2 and spacing at about 0.2 x 0.2 m. The use of slow release fertilizer (recommended because it is cheaper than commercial mulch and has a comparable effect) will depend on the priorities of the restoration ecologist. If quick seed production is important, then the use of fertilizer is recommended, if long term plant survival is more important, then no fertilizer is recommended.

The N. pulchra planted in the grooves created by the tractor ripper bar was significantly shorter than the control plots in height and percent survival, but this process was easier and quicker, and probably less damaging than soil scraping. It might be possible to establish N. pulchra by ripping the soil at two foot intervals and planting seedlings in longitudinal strips. However, the low survivorship (29 percent) of N. pulchra in the grooves would require a greater number of plants to achieve an appropriate percent cover. This method could be augmented by using a mechanical transplanter to rip and plant rows of seedlings, something we hope to test in the coming year.


Adams, MS 1964. Ecology of Stipa pulchra with special reference to certain soil characteristics. MS Thesis, Range Management, UC Davis, Davis, CA 76 p.

Bartolome, J.W. an B. Gemmill. 1991. The ecological status of Stipa pulchra in California. Madrono 28(3):172-184.

Burcham, L.T. 1957. California Rangeland: An Historico-Ecological Study of the Range Resource of California. California Division of Forestry, Sacramento, CA 260 p.

Edwards, F.S. 1995. Indicators of low intensity ecosystem perturbation: shifts in herbaceous plants and arbuscular mycorrhizal fungi in two semi arrid ecosystems. MS thesis, San Diego State University, San Diego, CA. 56 p.

Eliason, S.A. 1995. Competition between Artemisia californica and Mediterranean annual grasses. MS thesis, San Diego State University, San Diego, CA, 68 p.

Fossum, HC 1990. Effects of prescribed burning and grazing on Stipa pulchra seedling emergence and survival. MS Thesis, Ecology, UC Davis, Davis, CA 69 p.

Grinnell, J 1923. The burrowing rodents of California as agents in soil formation. Journal of Mammology 4(3):137-149

Langstroth, RP 1991. Fire and grazing ecology of Stipa pulchra grassland. A field study at Jepson Prairie. MS Thesis, Range and Wildland Science, UC Davis, Davis, Ca 76 p.

Minnich, R.A. And Y.H. Chou. 1997. Wildland fire patch dynamics in the chaparral of Southern California and Northern Baja California. International Journal of Wildland Fire 7(3):221-248.

Stromberg, M.R. And J.R. Griffin. 1996. Long-term patterns in Coastal California grasslands in relation to cultivation,gophers, and grazing. Ecological Applications 6(4): 1189-1211.

Western Regional Climate Center. 1998.

White, K.L. 1967. Native Bunchgrass (Stipa pulchra) on Hastings Reservation, California. Ecology 48(6):949-955.