Arctostaphylos canescens ssp. sonamensis not in Humboldt County?

The application of the name Arctostaphylos canescens Eatwood ssp. sonamensis (Eastwood) Wells to plants of the High North Coast Range region of Humboldt County, California is doubtful. Records for this CNPS List 1B.2 taxon are attributed to Mendocino and Humboldt Counties by both CNPS and CNDDB. An alternative taxonomic view is here offered that considers the northern, glandular plants to be hybrids between Arctostaphylos canescens ssp. canescens X A. viscida ssp. pulchella, and thus the subject plants are not of conservation concern.

Eastwood (1933) described Arctostaphylos sonamensis from a type from Rincon Ridge, Sonoma County. Knight (1985) later recognized the taxon as distinct from A. canescens. Wells (1988) submerged A. sonamensis as subspecies of A. canescens, commenting;

"a consistently different glandular race of A. canescens with a wide but segregated (allopatric) distribution relative to the nominate subspecies...Although subsp. sonamensis occurs on volcanic and other rocks, it appears to be restricted to serpentinite at the northern limits of its known range, as on the summit of Horse Mountain, Humboldt Co., (unpublished collection). Perhaps the glandulosity of pedicles and fruit and serpentine tolerance derive from some genes of A. viscida ssp. pulchella having introgressed into A. canescens ssp. canescens at some time and place."

Wells (2000) continued to attribute Arctostaphylos canescens ssp. sonamensis to Horse Mountain, Humboldt County.

Taxa of Arctostaphylos are well known to form fully fertile hybrids where related, diploid clades come into close contact (Dobzhanskey 1953, Schmid et al. 1968, Gottleib 1968, Schierenback et al. 1992), particularly where manzanitas are abundant in early successional plant communities where ecological segregation present in 'climax' settings has broken down by disturbance (Anderson 1948, Kruckeberg 1977). Several named hybrids involving Arctostaphylos canescens and A. viscida are known (Wells 1988).

Pubescence Features of the Plants in Question

Eastwood (1933) circumscribed Arctostaphylos canescens ssp. sonamensis on the basis of having two trichome types within the inflorescence: the twigs, pedicles and rachices bearing long, multicellular white trichomes as well as glandular trichomes, and the fruits bearing only glandular trichomes. Similarly, Wells (2000) characterizes the difference between Arctostaphylos canescens ssp. canescens and Arctostaphylos
canescens ssp. sonamensis on the following basis:

Arctostaphylos canescens ssp. canescens:

pedicles, rachises and twigs without glands, white hairy; ovary and fruit egandular

Arctostaphylos canescens ssp. sonamensis:

pedicles, rachises and twigs minutely glandular, glands often obscured on twigs by downy white hairs; ovary and fruit with minute glands

Based on both Eastwood (1933) and Wells (2000), plants that lack glandular trichomes on the pedicles or inflorescence rachises do not conform to the circumscription of Arctostaphylos
canescens ssp. sonamensis.

Plants on in the vicinity of Board Camp Mountain

Plants of Arctostaphylos canescens were found at two sites in the vicinity of Board Camp Mountain in July 2009 (vouchered Taylor #20626 bound for JEPS). Both sites [T4N R4E & T5N R3E] support only isolated, waif volunteers along roads. At both sites, the plants are invariant: glandular pubescence is confined entirely to the fruits, being absent from the pedicles and rachices of the inflorescence (see photo). In this regard, these plants do not fit the circumscription of Arctostaphylos canescens ssp. sonamensis because they lack glandularity on the pedicles and rachices. Gottleib (1968) showed that the anatomical distribution of glandularity of hybrids between A. viscida ssp. pulchella and Arctostaphylos canescens on serpentine at Waldo, Oregon was variable, and because of the lack of glandularity other than on the fruit, the subject plants are most similar to Arctostaphylos x cinerea T.J. Howell rather than Arctostaphylos x bracteata T.J. Howell or A. x. oblongifolia T.J. Howell (Wells 1988).

Ultimately, there is no single model of manzanita taxonomy that is correct to the exclusion of alternative hypotheses [that is, absent a genomic-level understanding]: this instance in my view is that the weight of the evidence suggests that Arctostaphylos canescens ssp. sonamensis is a narrow endemic of Sonoma County, and that similar plants found in Humboldt County, including those attributed to Horse Mountain by by Wells (1988, 2000), and subsequently elsewhere in the region are plants of hybrid origin between A. viscida ssp. pulchella and Arctostaphylos canescens, and these plants should not be treated as rare and are not typical Arctostaphylos
canescens ssp. sonamensis.Literature Cited:

Literature Cited:

Anderson, E. 1948. Hybridization of the habitat, Evolution 2:1-9.

Dobzhansky, T. 1953. Natural hybrids of two species of Arctostaphylos in the Yosemite region of California. Heridity 7:73-79.

Eastwood, A. 1933. New species of Arctostaphylos. Leaflets Western Botany
1:61-62

Gottlieb, L.D. 1968. Hybridization between Arctostaphylos
viscida and A. canescens in Oregon. Brittonia 20:83-93.

Kruckeberg, A.R. 1977. Mantanita (Arctostaphylos)
hybrids in the Pacific Northwest: effects of human and natural disturbances. Systematic Botany 2(4:233-250.

Schmid, R., T. Mallory and J.M. Tucker. 1968. Biosystematic evidence for hybridization between Arctostaphylos nissenana and A. viscida. Brittonia 20:34-43.

Schierenbeck, K. , R.W. Patterson and G.L. Stebbins. 1992. Morphological and cytological evidence for polyphyletic allopolyploidy in Arctostaphylos mewukka (Ericaceae). Plant Systematics & Evolution 179:187-205.

Wells, P.V. 1988. New combinations in Arctostaphylos (Ericaceae): annotated list of changes in status. Madrono 35(4):330-341

Wells, P.V. 2000. The Manzanitas of California also Mexico and the World. Privately printed