Endangered Species Recovery Program
Taxonomy.-- Kern mallow was first described as Eremalche kernensis, based on a specimen from the "Temblor Valley, 7 miles. northwest of McKittrick", in Kern County (Wolf 1938, p.67). Both Kearney (1951) and Munz (1958) transferred this species to the genus Malvastrum then reconsidered (Kearney 1956, Munz 1968) and returned to the original name. Other combinations have been suggested (Leonelli 1986) but were not validly published. The most recently-published treatments (Bates 1992, 1993) assign Kern mallow the name Eremalche parryi ssp. kernensis. However, the taxonomy of Kern mallow remains controversial in terms of its rank and its relationship to Parrys mallow (Eremalche parryi ssp. parryi). Most local botanists continue to use the scientific name Eremalche kernensis (Medlin in litt. 1995a) for this member of the mallow family (Malvaceae).
Description.-- The height and habit of Kern mallow (Figure 9) vary depending on seasonal precipitation. The form can vary from single-stemmed to multiple-stemmed, with the central stem erect and the lateral stems trailing along the ground. Stem lengths at flowering may range from less than 2.5 centimeters (1 inch) to nearly 50 centimeters (20 inches). The flowers have five petals, and the wheel-shaped fruits are divided into single-seeded segments (Bates 1993).
Identification.-- The taxonomic debate centers around the gender, color, and size of flowers indicative of Kern mallow versus Parrys mallow. Some populations in the Kern/Parrys mallow complex exhibit a condition known as gynodioecy, meaning that a population contains a mixture of plants that have only pistillate (female) flowers and plants that have only bisexual flowers (with both male and female parts). Bates (1992, 1993) considered any gynodioecious population in the Kern/Parrys mallow complex to be Kern mallow and those populations with only bisexual flowers to be Parrys mallow. On the other hand, Taylor and Davilla (1986) maintained that both Kern mallow and Parrys mallow were gynodioecious. Neither Wolf (1938) nor authors of early regional floras (Abrams 1951, Munz and Keck 1959) mentioned flower gender. Bisexual Kern mallow flowers produce fewer seeds per fruit (7 to 13) than do pistillate flowers (8 to 19). Parrys mallow and desert mallow (Eremalche exilis) fruits contain 10 to 22 and 9 to 13 segments, respectively (Abrams 1951, Munz and Keck 1959, Bates 1992, 1993, Mazer et al. 1993).
The strictest definition of Kern mallow applies only to populations in which white-flowered individuals predominate. Even in these areas, a few individuals may have pale lavender flowers (Wolf 1938, Bates 1992, Mazer et al. 1993), but lavender-flowered plants represented less than 10 percent of one population in 1994 (E. Cypher unpubl. data). Definite Parrys mallow populations consist of only pinkish-purple flowers, whereas those of questionable taxonomic affinity contain either exclusively pinkish-purple flowers or a very small proportion of white-flowered plants. Regardless of color, pistillate flowers have shorter petals than bisexual flowers in the same population (Bates 1992, 1993). Parrys mallow has larger flower parts than Kern mallow. Another closely-related species that infrequently occurs with the other two taxa is desert mallow, which has trailing stems and bisexual flowrs that are smaller than those of Kern mallow (Twisselmann 1956, Twisselmann 1967, Hoover 1970, Bates 1993). The populations of Kern mallow that are predominantly white-flowered are the object of conservation concern, and thus the strict interpretation is used in the following sections unless otherwise noted.
Figure 9. Illustration of Kern mallow (from Abrams, Vol. 3, 1951, with permission).
Historical Distribution.-- Kern mallow has always had a highly-restricted distribution. In the original description, Wolf (1938) mentioned specimens from the Temblor Valley, Belridge Oil Field, and two sites west of Buttonwillow; all these occurrences were in western Kern County north of McKittrick.
Current Distribution.-- A 1986 status survey reported three additional occurrences in Lokern, which is the local name for the area between Buttonwillow and McKittrick (Taylor and Davilla 1986). More intensive surveys during the past few years (Anderson et al. 1991, Olson and Magney 1992, CDFG 1995, Stebbins et al. 1992, S. Carter pers. comm.) revealed that Kern mallow occurs intermittently within an area of approximately 100 square kilometers (40 square miles) in Lokern, which is best described as a single metapopulation (Figure 10). The California Native Plant Society (Skinner and Pavlik 1994) and CDFG (1995) also accept reports of plants from three sites between Maricopa and McKittrick (in extreme western Kern County) as representing Kern mallow. Because specimens are not available to determine the color of the flowers and these sites are outside of the accepted range, they are treated here as representing Parrys mallow.
Pink-flowered plants fitting Bates (1992, 1993) broader concept of Kern mallow are widespread. Recent reports indicated that these plants occurred in several areas of Kern County, including Buena Vista Valley, Elk Hills, Lost Hills, McKittrick Hills, Stockdale, and the Temblor Range. Recent and historical reports elsewhere included Corcoran in Kings County; the Carrizo Plain, Elkhorn Plain, Panorama Hills, and Temblor Range in San Luis Obispo County; the Cuyama Valley in Santa Barbara County; and Pixley in Tulare County (Hoover 1970, Leonelli 1986, Olson and Magney 1992, Skinner and Pavlik 1994, CDFG 1995, Taylor and Davilla 1986, E. Cypher unpubl. observ., S. Wilson pers. comm.). Parrys mallow ranges from Alameda to Ventura Counties (Bates 1992).
As with many arid-land annuals, the form, density, phenology (timing of different stages in the life cycle), and reproduction of Kern mallow vary greatly depending on precipitation.
Reproduction and Demography.-- In Lokern, Kern mallow seeds typically germinate in January and February, and the plants begin flowering in March. Fruit production begins within a few days after flowers appear; flower and fruit production may continue into May if sufficient moisture is available. The seeds fall from the fruits as soon as they are mature. Seeds are capable of germinating in the following growing season, but at least some remain ungerminated. The duration of seed viability in the soil is not known. Seed dispersal agents are unknown but probably include animals and wind (Taylor nd Davilla 1986, Mazer et al. 1993, E. Cypher unpubl. observ.).
Preliminary studies showed that insects facilitated pollination of Kern mallow. However, small numbers of seeds were produced when pollinators were excluded, even in pistillate plants which did not produce pollen. Possible explanations for this phenomenon were apomixis (i.e., seed set without fertilization), contamination of the test plants by researchers, or wind pollination. However, a higher frequency of seed set would have been expected if pollen was carried by the wind (Mazer et al. 1993). The native solitary bee species Diadasia laticauda is one potential pollinator of Kern mallow. This bee species occurs in Kern County and is known to visit mallows of the genus Eremalche. Furthermore, many bees of the genus Diadasia restrict their pollen collection to members of the mallow family (Thorp in litt. 1998).
Population size of Kern mallow varies with rainfall. Several botanists familiar with this species were unable to find Kern mallow at known locations in years of below-average rainfall (Wolf 1938, Twisselmann 1956, Bates 1992). In Lokern, Kern mallow density was nearly 10 times as high in 1995, a year of much higher than average rainfall, as in 1994, which had below-average rainfall during the growing season. Similarly, the number of flowers per plant ranged from 1 to 8 in 1994 and from 1 to over 700 in 1995 (E. Cypher unpubl. data.).
Habitat and Community Associations.-- Kern mallow typically occurs in the Valley Saltbush Scrub natural community, where it grows under and around spiny and common saltbushes and in patches with other herbaceous plants, rather than in the intervening alkali scalds. Associated herbs include red brome, red-stemmed filaree (Erodium cicutarium), woolly goldfields (Lasthenia minor), and white Sierran layia (Layia pentachaeta ssp. albida). Kern mallow typically grows in areas where shrub cover is less than 25 percent (Taylor and Davilla 1986). The amount of herbaceous cover varies with rainfall and microhabitat; in occupied areas of Lokern, herbaceous cover averaged 80 percent in 1993 and 48 percent in 1994 (Cypher 1994a, 1994b, E. Cypher unpubl. data). Kern mallow occasionally has reinvaded disturbed sites when existing populations remained in adjacent areas to provide sources of seed (Mitchell 1989, E. Cypher unpubl. observ.).
Kern mallow occurs on alkaline sandy loam or clay soils at elevations of 95 to 275 meters (315 to 900 feet) (Wolf 1938, CDFG 1995). Leonellis (1986) comparison of Kern mallow habitat in Lokern with Parrys mallow habitat in the Temblor Range revealed that Kern mallow grew on soils that were more alkaline, less saline, and less sandy than those where Parrys mallow grew.
Reasons for Decline.--The loss and degradation of habitat in the Lokern area have been responsible for the decline of Kern mallow. Construction of the California aqueduct impacted Kern mallow both directly, by destroying plants in its path, and indirectly, by providing water that allowed cultivation of cotton and alfalfa in the area of endemism. The western portion of Lokern was developed for petroleum production, which eliminated Kern mallow at the type locality. Two disposal facilities for liquid waste were constructed in occupied habitat. Causes of habitat degradation, not only in Lokern, but also in the populations south to Maricopa, included installation of pipelines and transmission lines and off-road vehicle use (CDFG 1995, Taylor and Davilla 1986).
Threats to Survival.-- Approximately 85 percent of the Kern mallow habitat in Lokern is privately owned and thus is vulnerable to development for many potential uses (CDFG 1995, Taylor and Davilla 1986, Presley 1994). Although the current level of petroleum prodction does not seem to pose a threat to the portion of the metapopulation that remains, increased production levels could cause further fragmentation and loss of localized colonies of Kern mallow. Ongoing activities such as oil exploration and maintenance of pipelines and utility corridors continue to disturb occupied habitat. The maximum levels of development and habitat disturbance that would be compatible with the continued existence of Kern mallow are unknown. A more remote threat is the possibility of spills from tank trucks traveling through the area on highways and roads.
Paradoxically, both uncontrolled grazing and cessation of grazing have the potential to threaten the Kern mallow metapopulation. Sheep have grazed the Lokern area for decades (Presley 1994) and continue to graze on private lands during the growing season (E. Cypher pers. observ.). Grazing reduces the number of stems and branches on Kern mallow plants, which in turn reduces reproductive output (Mazer et al. 1993). In addition, trampling is likely to lead to localized destruction of Kern mallow in bedding areas where sheep are concentrated (Taylor and Davilla 1986). However, light to moderate grazing may serve to reduce competition in areas that are dominated by aggressive exotics (Cypher 1994b). Demographic studies indicated that the survival rate of Kern mallow seedlings was reduced in dense stands of exotic plants compared to sparsely-vegetated sites (Cypher 1994b). Furthermore, flower production was significantly increased in preliminary experiments where competitors were reduced through clipping (E. Cypher unpubl. data). The overall effects of sheep grazing on Kern mallow populations are unknown and require further investigation to determine appropriate management for the area.
Application of malathion in Lokern or other pesticides on adjacent agricultural fields could pose a threat to the long-term survival of Kern mallow by reducing pollinator populations. Malathion is sprayed periodically on natural lands in the San Joaquin Valley to control the beet leafhopper, which transmits diseases to crops (Clark 1991). Although current permit conditions for the California Department of Food and Agriculture prohibit malathion spraying within 1.6 kilometers (1 mile) of Kern mallow occurrences, research has not been conducted to determine whether or not this buffer size is adequate. If pollinator numbers were reduced, the Kern mallow metapopulation likely would experience reduced seed-set (Mazer et al. 1993). Also, if apomixis was the primary source of seeds, genetic variability could decline and the metapopulation could be more vulnerable to disease or other catastrophic events, such as has been observed in common species (Burdon and Marshall 1981).
Kern mallow was federally listed as endangered in 1990 (USFWS 1990; Table 1). Even before then, Lokern was a focus for protection because a variety of endangered and threatened species occupy the area. The California Energy Commission, California Department of Water Resources, and USBLM have sponsored biological surveys in Lokern (Anderson et al. 1991, Stebbins et al. 1992, S. Carter pers. comm.). Approximately 15 percent of the occupied Kern mallow habitat, primarily on the margins of the metapopulation, is owned by USBLM and The Nature Conservancy. An interagency cooperative acquisition and management plan for the entire 17,800-hectare (44,000-acre) Lokern Conceptual Area is in draft form; participants include USBLM, CDFG, California Energy Commission, The Nature Conservancy, Center for Natural Lands Management, and USFWS. Chevron USA may contribute to preservation of the area by establishing a mitigation bank on its lands, which constitute approximately 40 percent of the conceptual area and a substantial portion of the Kern mallow habitat (Presley 1994). The draft Kern County Valley Floor Habitat Conservation Plan specifies that no more than 10 percent of the natural land in the Lokern Conceptual Area may be disturbed under its section 10(a)(1)(B) permit (T. James pers. comm.), but protection efforts ould not necessarily target occupied Kern mallow habitat.
Efforts that specifically targeted the conservation of Kern mallow included (1) research on the demography and reproductive biology of Kern mallow funded by CDFG (Mazer et al. 1993), (2) salvage of plant specimens and seed from the Laidlaw Waste Disposal Facility by Endangered Species Recovery Program and Laidlaw in cooperation with USFWS, (3) ongoing population monitoring and research on the response of Kern mallow to cattle grazing jointly sponsored by the Biological Resources Division of the U.S. Geological Survey, USBLM, USFWS, the Endangered Species Recovery Program, CDFG, and other agencies, corporations, and organizations, and (4) exclusion of grazing from known Kern mallow habitat under the control of USBLM and Center for Natural Lands Management.
Considering that habitat loss is the primary reason that Kern mallow is listed as an endangered species, the top-priority task for recovery is to protect habitat in Lokern. The goal is to protect 90 percent of the remaining occupied habitat. This goal is based on the recognition that some development in Lokern must be allowed for economic reasons and on the assumption that loss of an additional 10 percent of the habitat will not jeopardize the continued survival of the taxon, provided that the protected habitat is managed appropriately. Unoccupied habitat within the metapopulation also is important for population expansion and movement of pollinators and seed dispersers. Thus, additional elements of the strategy are to protect land in blocks of at least 65 hectares (160 acres) and to avoid fragmenting the metapopulation into more than two blocks of contiguous, protected natural land. Buffer zones of 150 meters (500 feet) or more should be protected beyond the population margins to reduce external influences and to allow for population expansion.
The long-term prospects for survival of Kern mallow would be enhanced if more than one metapopulation was protected. The preferred approach is to determine the identity of the questionable populations in other areas and protect any others that are identified through biosystematic analysis as Kern mallow, rather than to attempt artificial introductions. However, the decision as to whether to protect existing populations outside of Lokern or to plant seeds from Lokern at other sites depends on the outcome of systematic research. A biosystematic study (research that uses evidence from several disciplines to determine taxonomic affiliations) should be completed within 5 years of recovery plan approval. Moreover, if the pink-flowered and white-flowered mallow populations are determined to represent a single taxon, the listing status for Kern mallow would need to be reevaluated.
Additional high-priority tasks are to continue demographic and ecological research. Demographic studies are a prerequisite to matrix projection modeling, which is necessary to identify vulnerable stages in the life cycle. Only if these limiting stages are managed properly can populations be assumed to be self-sustaining (Schemske et al. 1994). Research is required to determine the relative magnitude of threats posed by exotic plants and sheep and to formulate appropriate management strategies for all protected lands. Even after demographic studies are discontinued, population trends should be monitored annually and management strategies should be reassessed if the Lokern metapopulation begins to decline. Several important aspects of pollination ecology must be investigated in greater detail, including the identity of insect pollinators, their vulnerability to pesticides that are used locally, and other mechanisms of pollen transfer. Until more specific recommendations are available from research, pollinator availability should be considered a limiting factor and pesticide spraying should be avoided in Lokern during the Kern mallow flowering period.