Pentachaeta lyonii A. Gray is a state- and federally-listed endangered species, endemic to heavily invaded southern California grasslands. Recent population extirpations resulting in a decrease in range size have prompted investigation into the effects of invasive annual plants on this species. The goals of this study were (1) to examine the impacts of competition from non-native species from three different functional groups (annual grasses, early-season forbs, and late-season forbs) on P. lyonii success in the field and in pots, (2) to determine which non-native species/functional groups have the greatest competitive effect on P. lyonii, and (3) to evaluate the environmental conditions that contribute to the displacement of P. lyonii by non-native plants. In the field, at two sites over two years, control plots were paired with plots in which non-native competitors were clipped at the soil surface. In pots, individual P. lyonii plants were grown in competition with all three groups of non-native competitors at both high and low density. In both the field and in the pots, all three non-native plant groups had negative effects on P. lyonii reproductive potential, with Centaurea melitensis L. having the greatest effect. The effects on P. lyonii height were variable among non-native competitors and between years. Comparisons made of environmental features of sites where P. lyonii has been extirpated to those where it persists suggested that the presence of annual grass is associated with P. lyonii extirpation. Management of P. lyonii presents a challenge considering the tendency of this species to coexist with non-native annual plants due to their common disturbance-dependence, and the ubiquity of European annuals in P. lyonii habitat.

Cryptic self-incompatibility (CSI) is a type of non-random mating observed in self-compatible plants in which outcross pollen sires proportionally more seeds than self pollen when both pollen types are available on the stigma. Levels of CSI are known to vary among individuals and populations. We conducted competitive pollinations consisting of mixtures of self and outcross pollen to investigate reports of CSI in certain populations of Clarkia unguiculata Lindl. We also investigated how the order of self and outcross pollen deposition on the stigma influences the degree of nonrandom mating. Finally, we looked at whether the source population of outcross pollen affected the ability of outcross pollen to outcompete self pollen. We utilized recessive (white-petaled) maternal (and self pollen donor) plants from a locality near Morro Bay, California, and dominant (pink-petaled) outcross pollen donor plants from 17 localities widespread through the species range in California. Progenies from pollinations made with equal mixtures of self and cross pollen included significantly more outcross-pollinated than self-pollinated offspring in 10 of the 17 cross-pollen donor populations. However six populations showed no significant difference between self- and outcross-pollinated offspring, and one population yielded a majority of progeny sired by self pollen. Progenies from sequential self pollen followed by outcross pollinations included significantly more self offspring in 12 of the donor populations, no significant difference between outcross and self offspring in four donor populations and significantly more outcross offspring in one donor population. Progenies from sequential outcross pollen followed by self pollinations included significantly more outcross offspring in 15 donor populations and no significant difference between outcross and self offspring in two donor populations. Our results confirm the occurrence of non-random mating in C. unguiculata, and demonstrate that the degree of non-random mating can depend on the order of self vs. outcross pollen deposition and the source population of outcross pollen. This non-random mating can influence the proportion of self and outcross progeny in sequential pollinations.

Olea europaea (European olive) is invasive in Australia and widely planted in California. Vertebrates, particularly birds, mediate Olea seed dispersal. Fruits are large, but their sizes range widely. We measured fruit widths from 12 study stands in California and constructed esophageal probes in diameters spanning the resulting size range. We then obtained whole-bird frugivorous bird carcasses and used the probes to determine the fruit sizes that each bird species would be anatomically capable of swallowing. This allowed us to develop lists of potential dispersers for each study stand. Even stands with the largest measured fruits had several potential disperser species, but the list of species expanded greatly as fruit sizes decreased. Feral Olea stands with remarkably small fruits have been observed in California and Australia. An increase in the incidence of such stands might augment the regional spread rate for the species.

We present a detailed comparison of Papaver californicum and Stylomecon heterophylla, which earlier were found to be sister species and most closely related to Meconopsis cambrica Papaver s.str. from western Eurasia. The two species of winter annuals differ mainly in the shape of their distal cauline leaves, coloration of petals and staminal filaments, and most notably morphology of the gynoecium and capsule, with Papaver californicum having a sessile stigmatic disc and Stylomecon heterophylla having a distinct style. They were earlier found to differ in ploidy, with chromosome numbers of 2n  =  28 (Papaver californicum) and 2n  =  56 (Stylomecon heterophylla). Mapped distributions of the two species indicate that the range of S. heterophylla encompasses and exceeds that of P. californicum; both are known only from the California Floristic Province except for one collection of S. heterophylla from central Baja California. Whereas Papaver californicum is most commonly found in burn localities in the first wet season after fire, Stylomecon heterophylla is active under a broader range of environmenal conditions and often occurs in habitats that appear to be somewhat more mesic. Both species are self-compatible and autonomously self-pollinating. Experimental hybridization resulted in well-developed but entirely sterile hybrids; no hybrids are known from nature. Based on these and earlier findings, we conclude that Stylomecon heterophylla is best treated in Papaver, as P. heterophyllum.

Fir dwarf mistletoe (Arceuthobium abietinum, Viscaceae) is a common parasite of California white fir (Abies lowiana) and red fir (Abies magnifica) in California. Based on its host specificity, fir dwarf mistletoe consists of two special forms: A. abietinum formae specialis concoloris on California white fir and A. abietinum f. sp. magnificae on red fir. I sampled 17 populations of each special form in the Sierra Nevada and extreme southern Cascade Mountains (Mt. Lassen area) and completed additional morphological measurements of male and female plants, flowers, and fruits. As reported by previous studies, my results demonstrated that these special forms are morphologically similar. No significant differences were detected between the plant, flower, or fruit dimensions measured. The plant color of white fir and red fir dwarf mistletoe was also similar for both male and female plants, but some plants of red fir dwarf mistletoe are more brown-green than white fir dwarf mistletoe, particularly in the northern end of its geographic range. Based on the results of this study no change in the taxonomic status of the special forms of fir dwarf mistletoe was recommended.

The Mason Valley cholla, Cylindropuntia ×fosbergii (C. B. Wolf) Rebman, M.A. Baker & Pinkava, is the putative hybrid of C. bigelovii (Engelm.) F. M. Knuth and some other species of Cylindropuntia. We used AFLPs to screen chollas of the Anza-Borrego Desert in southern California to test this hypothesis of hybrid origin and identify the parental species involved. Other species scrutinized as potential parents include C. echinocarpa (Engelm. & J. M. Bigelow) F. M. Knuth, C. ganderi (C. B. Wolf) Rebman & Pinkava, C. californica var. parkeri (J. M. Coulter) Pinkava, and C. wolfii (L. D. Benson) M.A. Baker. Patterns of band sharing clearly testify to the close relationship between C. ×fosbergii and C. bigelovii. None of the other species screened came close to that level of similarity. Moreover, the numbers of total loci and unique loci in C. ×fosbergii do not meet the expectations of a hybrid taxon. We propose the alternative hypothesis that C. ×fosbergii is the sister species of C. bigelovii.

Gabbro is a mafic rock with many species that do not occur on soils of granitic or ultramafic rocks. Although some gabbro soils harbor many unique or endemic species, others do not, as botanists have noted from the Appalachian Piedmont to the mountains of California and Oregon. Gabbro soils were sampled in the Peninsular Ranges and in the foothills of the Sierra Nevada to identify special features of the gabbro soils with unique plant species distributions. No soil morphological or chemical differences were found between gabbro soils with and without unique plant species that might explain the differences in plant distributions. Although many unique species may occur only on gabbro soils, apparently their distributions cannot be explained primarily by soil differences among gabbro soils.

The unusual edaphic habitats of late Tertiary lacustrine deposits in Sonoran Desert basins of central Arizona have previously been shown to harbor endemic taxa and disjunct taxa from other floristic regions, which inhabited the Sonoran Desert during previous climatic regimes. The infertile limestone soils contrast sharply with the surrounding volcanic soils, excluding the dominant Sonoran Desert vegetation and thereby providing an ecological opening for the disjuncts. The disjunctions provide clues for interpreting the biogeographical history of Arizona. Here, thirteen additional examples are documented. Taxonomic changes to two of the earlier examples, Eriogonum apachense and Hymenoxys acaulis var. arizonica, are discussed. Erigonum apachense from the San Carlos Basin is revealed to be a disjunct population of the northern E. heermannii var. argense, not a separate endemic species; and, a recently named taxon, Tetraneuris verdiensis, from the Verde Valley, is shown to be a rayless form of T. acaulis var. arizonica, not a separate endemic species. The biogeography of Ericameria nauseosa var. juncea and Quercus havardii are considered in more detail. The type locality of Ericameria nauseosa var. juncea is one of the disjunct localities, not from the main range of the variety. An unusual thicket-forming oak in the Verde Valley is determined to be a disjunct population of Quercus havardii from Staked Plains of New Mexico and Texas and the Four Corners of the Colorado Plateau.