Biometric and Allozymic Characterization of Three Coastal and Inland Killifish Populations (Pisces: Fundulidae) from the Peninsula of Baja California,Mexico

Type localities and/or central populations of coastal and inland killifishes along the peninsula of Baja California were sampled to survey the biometric and allozymic relationships among them. The standard stepwise discriminatory function analysis of biometric characters classified organisms into three well distinctive groups (p < 0.001). The number of scales in the lateral series successfully split the species, while the caudal length/standard length ratio did so between subspecies. Two diagnostic loci (PROT–2* and AAT–2*) segregated private alleles in each taxon. Fundulus lima showed lower genetic variability levels than F. parvipinnis spp., in both polymorphism (18.8% in F. lima vs. 25.0% in F. parvipinnisspp.) and expected heterozygosity (0.060 in F. lima vs 0.099 in F. p. brevis and 0.089 in F. p. parvipinnis). The reduced genetic variability in the threatened species F. lima may be related to its evolutionary history. Nei's genetic distance between F. lima and F. p. parvipinnis was 0.247, while 0.075 between F. parvipinnis spp.     

Population morphometric variation of the endemic freshwater killifish, <Emphasis Type="Italic">Fundulus lima</Emphasis> (Teleostei: Fundulidae), and its coastal relative <Emphasis Type="Italic">F. parvipinnis</Emphasis> from the Baja California Peninsula,Mexico

The population morphometric variation of the endangered freshwater killifish (Fundulus lima) was evaluated and compared with that of its euryhaline coastal relatives (F. parvipinnis parvipinnis and F. p. brevis) on the basis of 384 specimens from the Baja California peninsula, Mexico. Forty five standardized body distances were compared by means of discriminant function analysis (DFA). Sixteen body distances were significant to distinguish two groups of populations for F. lima: a first group represented by the Bebelamas and San Javier basins, and second group composed by the basins of San Ignacio, La Purísima, San Luis, San Pedro and Las Pocitas. When all freshwater and coastal populations were compared, the southernmost population of F. lima (Las Pocitas) showed a higher morphometric similarity with the southern coastal subspecies (F. p. brevis), while another southern population (San Pedro) had an intermediate position between the freshwater and coastal forms. This study suggests the presence of five evolutionary units (three freshwater and two coastal) for the genus Fundulus in the Baja California peninsula.     

Ichthyofaunal utilization of newly-created versus natural salt marsh creeks in Mission Bay,CA

Loss of wetland habitat has proceeded at an alarmingrate in southern California, and increasingly marshrestoration and creation are being used to mitigatethese losses. As part of an effort to evaluatefunctional equivalence of created systems, theichthyofaunal assemblages in a created and adjacentnatural marsh in Mission Bay, San Diego, Californiawere compared. Fishes trapped in both marshes includedFundulus parvipinnis, Gillichthysmirabilis, Acanthogobius flavimanus, Ctenogobius sagittula, Atherinops affinis, andMugil cephalus. Fundulus parvipinniswasnumerically dominant in both systems, representing onaverage 69% of all fishes trapped in the createdmarsh and 65% of all fishes trapped in the naturalmarsh. Gillichthys mirabiliswas the second-mostabundant species, representing on average 31% of allfishes trapped in the created marsh and 28% of allfishes trapped in the natural marsh. Species richnessand dominance measures were similar between the twosystems, while abundances were higher in the naturalrelative to the created marsh. The size-structure ofF. parvipinnisand G. mirabilisdifferedbetween the created and natural marsh creeks, with thecreated marsh populations being skewed towards largersize classes. These size differences are believed toarise from differences in creek morphology between thecreated and natural systems, and potentially affectboth predators and prey of these species in the marsh.Mark-release-recapture revealed considerable marshfidelity, with as many as 35% of the F.parvipinnistagged in a marsh being recovered one daylater in the same marsh. Stable isotope analyses ofF. parvipinnisrevealed similar ¹⁵Nand ³⁴S values between marshes; howeverthere was a consistent enrichment in ¹³C (>3per mil) in tissues of F. parvipinnisfrom thecreated marsh, supporting the high marsh fidelitysuggested by tagging results. This first publisheddocumentation of the Mission Bay marsh resident fishessuggests that the created marsh ichthyofaunalassemblage was distinct in density and size structurefrom the adjacent natural marsh, and provides lessonsfor future restoration efforts.     

Oxidative detoxification of sulfide by mitochondria of the California killifish Fundulus parvipinnis and the speckled sanddab Citharichthys sitgmaeus

Summary Earlier whole-animal experiments have shown that the California killifish (Fundulus parvipinnis) from tidal marshes is highly tolerant to sulfide while the speckled sanddab (Citharichtys stigmaeus) from the open coast is intolerant to sulfide. In the present paper, we demonstrate that the liver mitochondria of the California killifish detoxify sulfide by oxidizing it to thiosulfate and produce ATP in the process. Sulfide oxidation is obligately and stoichiometrically linked to mitochondrial electron transport to oxygen. Concentrations up to 20 M sulfide stimulate mitochondrial respiration while 50 M sulfide causes half-inhibition. Sulfide oxidation by mitochondria is adversely affected at pH<7.4. ATP production is maximal at 10 M sulfide. The finding of sulfide oxidation coupled to ATP production by killifish mitochondria is unprecedented among vertebrates. In comparison, mitochondria of the specked sanddab oxidize sulfide at a much lower rate. This is the first demonstration of biochemical adaptation to sulfide among coastal marine fishes.Abbreviations ADP adenosine diphosphate - APHA American Public Health Association - ATP adenosine triphosphate - BSA bovine serum albumin - EGTA ethyleneglycol-bis-(-aminoethyl-ether) N,N,N,N-tetraacetic acid - G-6-PDH glucose-6-phosphate dehydrogenase - HEPES N-2-hydroxyethylpiperazine-N-2-ethane-sulfonic acid - HPLC high-performance liquid chromatography; mBBr monobromobimane - NADP nicotinamide adenine dinucleotide phosphate, oxidized form - NADPH reduced form - RCR respiratory control ratio     

Single Versus Multiple Sources of Introduced Populations Identified With Molecular Markers: A Case Study of a Freshwater Fish

The presence of a widespread exotic raises the question as to whether the successful invasion can be attributed to spread after a single introduction or is the product of multiple introductions. The plains killifish, Fundulus zebrinus, is native to the western Great Plains of the United States but is also found throughout the Colorado River basin. We surveyed five introduced populations for genetic variation in allozymes and mitochondrial DNA (mtDNA) in order to compare the genetic structure of these populations with those from across the native range. Our survey revealed two genetically distinct stocks of introduced F. zebrinus, one in the San Juan and one in the Yampa and Colorado drainages. Each stock corresponds to at least one unique introduction event. In addition, data available from a genetic survey of the native range allowed us to identify the sources of these introduced populations. The combination of allozyme and mtDNA data for the San Juan population indicated that the most probable source for this population was the Pecos drainage of New Mexico and Texas. The Yampa and Colorado populations seem to be derived from the western Arkansas drainage, although the allozyme data indicated that another nearby drainage(s) might have also served as a source for the Colorado population.     

The invasive bullfrog <Emphasis Type="Italic">Lithobates catesbeianus</Emphasis> in oases of Baja California Sur,Mexico: potential effects in a fragile ecosystem

The American bullfrog (Lithobates catesbeianus), considered as one of the world’s worst invasive species because of its negative effects on native fauna, has been introduced into over 40 countries in four continents. The invasion status in Mexico is poorly known. The first known record for this species in the Peninsula of Baja California, northwestern Mexico is an individual collected 35 years ago. A great portion of the peninsula is an arid desert where the sources of freshwater are the spring-fed oasis and mountain arroyos that support rare and endemic species closely associated with wetland conditions. Humans, who have introduced alien species, inhabit many of those oases. The aim of this work was to document the presence of the exotic bullfrog in 40 oases and mountain arroyos in Baja California Sur state during 2006–2009. Bullfrogs were found in 25% of the oases sampled; 15% have confirmed records of breeding populations and five oases are new location records. Our results show a considerable increase (50%) in the number of oasis invaded by bullfrogs in only 8 years. The animal populations of rare or endemic species associated with oases would be threatened by the arrival of bullfrogs. Actions for their control and eradication are urgently needed.     

Microsatellite primers for the Atlantic coastal killifish,Fundulus heteroclitus,with applicability to related Fundulus species

The mummichog (Fundulus heteroclitus), a common Atlantic coastal killifish, is a model vertebrate species for the study of molecular genetic variation in natural populations and of environmental toxicology. We report the development of a set of 20 microsatellite loci in this species. Average expected heterozygosity across all loci was 0.84 (range: 0.60–0.97), revealing a high level of variability at most loci. A survey of seven additional Fundulus species yielded one or two robust amplification products in over half (63%) of the species–primer combinations tested. Therefore, many of these loci will also prove useful in studies of other members of the genus Fundulus.     

The relationship between handling time and cortisol release rates changes as a function of brain parasite densities in California killifish Fundulus parvipinnis

This study validated a technique for non‐invasive hormone measurements in California killifish Fundulus parvipinnis, and looked for associations between cortisol (a stress hormone) and 11‐ketotestosterone (KT, an androgen) release rates and the density or intensity of the trematode parasites Euhaplorchis californiensis (EUHA) and Renicola buchanani (RENB) in wild‐caught, naturally infected F. parvipinnis. In experiment 1, F. parvipinnis were exposed to an acute stressor by lowering water levels to dorsal‐fin height and repeatedly handling the fish over the course of an hour. Neither parasite was found to influence cortisol release rates in response to this acute stressor. In experiment 2, different F. parvipinnis were exposed on four consecutive days to the procedure for collecting water‐borne hormone levels and release rates of 11‐KT and cortisol were quantified. This design examined whether F. parvipinnis perceived the water‐borne collection procedure to be a stressor, while also exploring how parasites influenced hormone release rates under conditions less stressful than those in experiment 1. No association was found between RENB and hormone release rates, or between EUHA and 11‐KT release rates. The interaction between EUHA density and handling time, however, was an important predictor of cortisol release rates. The relationship between handling time and cortisol release rates was negative for F. parvipinnis harbouring low or intermediate density infections, and became positive for fish harbouring high densities of EUHA.     

Brain‐encysting trematodes and altered monoamine activity in naturally infected killifish Fundulus parvipinnis

This paper presents novel evidence to address mechanisms by which trematode parasites effect behavioural changes in naturally infected fish hosts. California killifish Fundulus parvipinnis infected with the brain‐encysting trematode Euhaplorchis californiensis display conspicuous swimming behaviours that render them 30 times more likely to be eaten by birds, the parasite's final host. Prevalence of E. californiensis reaches nearly 100% in most F. parvipinnis populations, with parasite biomass constituting almost 2% of F. parvipinnis biomass in some locations. Despite having thousands of cysts on their brains, infected fish grow and mature at rates comparable to those of uninfected populations. The lack of general pathology combined with the specificity of the altered behaviours suggests that the behavioural changes are due to parasite manipulation. The monoamine neurotransmitters serotonin and dopamine, which control locomotion and social behaviour in fishes and other vertebrates, were examined to explore the underlying mechanisms of this behaviour modification. Whereas previous studies were similarly conducted with experimentally infected fish, in this study, brain dopaminergic and serotonergic activity were analysed in naturally infected fish to assess how E. californiensis may alter F. parvipinnis monoamines in a naturally occurring system. A parasite density‐associated decrease in serotonergic activity occurred in the hippocampus of naturally infected fish, as well as a decrease in dopaminergic activity in the raphe nuclei, suggesting that E. californiensis inhibits serotonin and dopamine signaling in naturally infected F. parvipinnis. The neurochemical profile of infected fish is consistent with the hypothesis that E. californiensis affects brain monoaminergic systems in order to induce impulse‐driven, active, and aggressive behaviour in its hosts.     

GENE FLOW AND MITOCHONDRIAL DNA VARIATION IN THE KILLIFISH,FUNDULUS HETEROCLITUS

Two subspecies of killifish, Fundulus heteroclitus, inhabit the Atlantic coast of the United States. The contact zones between them are typified by morphological, physiological, nuclear gene, and mitochondrial DNA clines. Considerable debate exists about the importance of restricted gene flow and natural selection in maintaining these clines. To evaluate the relative importance of these two evolutionary forces we employed analysis of mtDNA as an independent measure of gene flow. Solutions of equilibrium diffusion equations suggested that killifish dispersal must be less than 2 km to maintain previously observed allozyme frequency clines in the absence of strong selection. To determine whether populations separated by distances greater than 2 km accumulate significant genetic differences we examined a total of N = 480 individuals from five killifish populations spanning a continuous shoreline distance of 8.4 km. Distribution of the 25 detected haplotypes indicated that most of the variation was within rather than between sampling locations. No evidence of geographic structure was detected, nor were there any significant genetic differences between killifish populations. With these data, gene flow was evaluated by estimating effective migration rate (N_em) between the populations. N_em was estimated from G_st(N_em = 49.4), from F_st (N_em = 24.1), by the private alleles approach (N_em = 18.5) and via four phylogenetic analyses (N_em ranged from 11.4 to 16.9). Regardless of the analysis, N_em was greater than one; the threshold level at which gene flow is strong enough to prevent differentiation due to genetic drift. This suggests that while the characteristics that distinguish Fundulus subspecies may have arisen in isolation, the current clinal distributions exhibited along the Atlantic coast are most likely maintained by selection.     

Fish community responses to ecosystem stressors in coastal estuarine wetlands: a functional basis for wetlands management and restoration

Functional responses of estuarine fish species to environmental perturbations such as wetland impoundment, changes in water quality, and sediment accretion are investigated. The study focuses on the feeding, growth and habitat use by California killifish (Fundulus parvipinnis), topsmelt (Antherinops affinis), and juvenile California halibut (Paralichthys californicus) in impacted coastal wetlands to provide an ecological basis for guidance on the management and restoration of these ecosystems. The ecology of California killifish, Fundulus parvipinnis, is closely tied with the marsh surface, which they access at high tide to feed and grow. Field estimates of food consumption show that killifish can increase their food intake by two-fold to five-fold by adding marsh surface foods to their diet. Bioenergetics modeling predicts that killifish can grow over an order of magnitude faster if they add intertidal marsh surfaces to their subtidal feeding areas. Tidal inlet closures and increased marsh surface elevations due to sediment accretion can restrict killifish access to the marsh surface, affecting its growth and fitness. An open tidal inlet and tidal creek networks that allow killifish to access the marsh at high tide must be incorporated into the restoration design. Topsmelt and California halibut are also adversely affected by tidal inlet closures. Food consumption rates of topsmelt are 50% lower when the tidal inlet is closed, compared to when the estuary is tidally-flushed. Tidal inlet closures inadvertently induce variations in water temperature and salinity and negatively affect growth of juvenile California halibut. Tidal creek networks which consist of channels and creeks of various orders are also important to halibut. Large halibut (>200 mm TL) inhabit deeper, high order channels for thermal refuge, while small halibut (<120 mm TL) are abundant in lower order channels where they can feed on small-sized prey which are typically less abundant in high order channels. Maintaining an open tidal inlet, implementing sediment management programs and designing coastal wetlands with tidal creek networks adjacent to intertidal salt marsh habitat (for fish access) are key elements that need to be considered during the planning and implementation of coastal wetland restoration projects.     

Use of Museum Specimens to Refine Historical Pronghorn Subspecies Boundaries

Endangered Sonoran (Antilocapra americana sonoriensis) and Peninsular (A. a. peninsularis) pronghorn persist largely because of captive breeding and reintroduction efforts. Recovery team managers want to re-establish pronghorn in their native range, but there is currently uncertainty regarding the subspecies status of extinct pronghorn populations that historically inhabited southern California, USA, and northern Baja California, Mexico. To address this uncertainty, we genotyped museum specimens and conducted phylogenetic and population genetic analyses of historical data in the context of 3 contemporary pronghorn populations. The historical northern Baja California pronghorn share the most ancestry with contemporary Peninsular pronghorn, whereas pronghorn in southern California share more ancestry with contemporary American (A. a. americana) pronghorn. For reintroductions into northern Baja California, the Peninsular subspecies is more appropriate based on museum genetic data. For reintroductions into Southern California, ecological and genetic factors are both important, as the subspecies most genetically related to historical populations (American) may not be well-adapted to the hot, low-elevation deserts of the reintroduction area. © 2019 The Wildlife Society.     

Effects of spironolactone and RU486 on gene expression and cell proliferation after freshwater transfer in the euryhaline killifish

We have explored the possible mechanisms by which mineralocorticoid (MR) and glucocorticoid (GR) receptors regulate the response to freshwater transfer in the gills of the euryhaline killifish Fundulus heteroclitus. Killifish were implanted with RU486 (GR antagonist) or spironolactone (MR antagonist) at doses of 0.1–1.0 mg g⁻¹, and subsequently transferred from 10‰ brackish water to freshwater. Compared to brackish water sham fish, mRNA expression of CFTR and NKCC1 decreased in the gills of sham fish transferred to freshwater, whereas Na⁺,K⁺–ATPase α_1a mRNA expression and α protein abundance, as well as cell proliferation (detected using BrdU) increased. Spironolactone inhibited the normal increase in cell proliferation and Na⁺,K⁺-ATPase expression after freshwater transfer. RU486 increased plasma cortisol levels and may have slightly inhibited Na⁺,K⁺–ATPase activity, but did not change α _1a expression. RU486 had no effect on cell proliferation in the non-lamellar region of the gills, but increased proliferation in the lamellar region. Neither antagonist inhibited the suppression of CFTR or NKCC1 expression after freshwater transfer. Glucocorticoid receptor expression was reduced in all sham and antagonist treatments compared to untreated controls, but no other consistent differences were observed. The effects of spironolactone suggest that MR is important for regulating ion transport in killifish gills after freshwater transfer.     

Microsatellite loci for the plains topminnow (Fundulus sciadicus,Fundulidae)

The plains topminnow (Fundulus sciadicus) is a killifish endemic to the Great Plains region. Recent field surveys indicate that populations may be declining in northerly portions of its range and perhaps throughout. Here, we describe the isolation and characterization of 11 microsatellite loci developed for population genetic studies in the species. We also conducted cross‐species amplifications to test primer efficacy in five additional taxa, including four killifishes.     

Development of microsatellite markers for the white mangrove (Laguncularia racemosa C.F. Gaertn., Combretaceae)

We developed nine new microsatellite markers for the white mangrove (Laguncularia racemosa C.F. Gaertn.), a mangrove species widely distributed in the Western Hemisphere. The loci were developed from a sample collected in Baja California Sur, Mexico, and were screened for polymorphism in 108 individuals from the Pacific Coast of Mexico and Costa Rica. The level of polymorphism was relatively low, from two to four alleles. These are the first set of polymorphic markers described for L. racemosa and should be of value in studying regional genetic differentiation among natural populations of this species.     

Genetic evidence for limited dispersal in the coastal California killifish, Fundulus parvipinnis

The California killifish, Fundulus parvipinnis, is a marine species that lives in salt marshes, estuaries and wetlands along the California and Baja California coasts. In order to estimate levels of dispersal between different coastal habitats over its range, we have studied six populations using morphological and genetic markers. Lateral line scale and vertebrae counts showed significant differences between individuals collected north of Punta Eugenia and south of Punta Eugenia. Morphological differences across Punta Eugenia were accompanied by large genetic differences at the mitochondrial control region (5.8%). Gene flow was in general very reduced over the range of the species (pairwise average F(st)=0.70, Nm=0.30), with a strong break at Punta Eugenia (F(st)=0.95, Nm=0.03). Such limited interchanges between coastal habitats have important theoretical and conservation implications.     

Appearance of cuboidal cells in relation to salinity in gills of Fundulus heteroclitus, a species exhibiting branchial Na+ but not Cl− uptake in freshwater

Fundulus heteroclitus (killifish) is a model organism for ionoregulatory studies, particularly because of its opercular epithelium, although the gills are the major sites of ion exchange. Whereas Na⁺ and Cl⁻ are excreted through the gills in seawater (SW), the killifish is unusual in taking up only Na⁺ and not Cl⁻ at the gills in freshwater (FW). We describe morphological changes in the branchial epithelium following transfer from an acclimation medium of 10% SW to 100% SW or FW. In 10% SW, mitochondria-rich cells resemble typical seawater chloride cells (SWCCs) with accessory cells. After transfer to 100% SW, no change occurs in pavement cell (PVC) morphology or mitotic rate (measured by bromo-deoxyuridine technique), although the density of SWCC apertures increases several fold because of the uncovering of buried SWCCs by PVCs, in accord with increased rates of Na⁺ and Cl⁻ efflux. After transfer to FW, PVC morphology remains unchanged, but SWCCs and accessory cells are quickly covered by PVCs, with many undergoing apoptosis or necrosis. The mitotic rate doubles by 10–14 h but typical freshwater chloride cells (FWCCs) do not appear. Instead, a wedge-shaped cell type that is moderately rich in apically oriented mitochondria, with a large ovoid nucleus, thin cytoplasmic layer, paucity of vesicular-tubular network, and variably villous surface rapidly (by 3 h) and progressively appears in the filament epithelium, by both uncovering and mitosis. This cell type is similar to that recently identified as the site of Na⁺ uptake in the FW trout gill. We propose the new term “cuboidal cell” for this cell, based on its morphology, to avoid confusion with traditional terminology (of PVC). We hypothesize that the cuboidal cells are the sites of active Na⁺ uptake in FW F. heteroclitus and suggest that the lack of Cl⁻ uptake is attributable to the absence of typical FWCCs previously described in teleosts.This work was supported by NSERC Discovery grants (to C.M.W.) and by an NSERC International Fellowship (to P.L.). C.M.W. is supported by the Canada Research Chair Program.     

Three reciprocally monophyletic mtDNA lineages elucidate the taxonomic status of Grant’s gazelles

The intraspecific phylogeography of Grant’s gazelles Nanger granti was assessed with mitochondrial DNA control region sequences. Samples of 177 individuals from 17 Kenyan and Tanzanian populations were analysed. Three highly divergent, reciprocally monophyletic lineages were found, with among group net nucleotide distances of 8–12%. The three lineages—notata, granti and petersii—grouped populations according to their geographic origin, encompassing populations in the north, southwest, and east, respectively. The mtDNA lineages reflected distinct evolutionary trajectories, and the data are discussed in reference to the four currently recognised subspecies. We suggest Grant’s gazelles be raised to the superspecies Nanger (granti) comprising three taxonomic units corresponding to the three mtDNA lineages. There was no evidence of gene flow between the notata and granti lineages, despite their geographic proximity, suggesting reproductive isolation. These constitute evolutionary significant units within the adaptive evolutionary framework. Due to its restricted geographic distribution and genetic and morphological distinctiveness, we suggest the petersii lineage be raised to the species Nanger (granti) petersii within the Grant’s gazelles superspecies.     

Molecular identification of vaginal lactobacilli isolated from Bulgarian women

Lactobacilli play an important role in maintaining the vaginal health of women. The development of suitable bacterial replacement therapies for the treatment of vaginosis requires knowledge of the vaginal lactobacilli species representation. The aim of this study was to identify at the species level vaginal Lactobacillus isolates obtained from Bulgarian women in childbearing age by using different molecular methods. Twenty-two strains of lactobacilli isolated from vaginal samples were identified and grouped according to their genetic relatedness. A combined approach, which included amplified ribosomal DNA restriction analysis (ARDRA), ribotyping and polymerase chain reaction (PCR) with species-specific oligonucleotide primers was applied. All vaginal isolates were grouped into 5 clusters in␣comparison with a set of 21 reference strains based␣on the initial ARDRA results, which was then confirmed by ribotyping. Finally, the strains were subjected to PCR analysis with eight different species-specific primer pairs, which allowed most of␣them to be classified as belonging to one of␣the␣following species: Lactobacillus crispatus, Lactobacillus fermentum, Lactobacillus gasseri, Lactobacillus helveticus and Lactobacillus plantarum. In conclusion, this study suggests that the most straightforward identification strategy for vaginal lactobacilli would be grouping by ARDRA or ribotyping, followed by PCR specific primers identification at species level.     

GEOGRAPHICAL CORRELATION OF MORPHOLOGICAL VARIATION IN SILVETIA COMPRESSA (FUCACEAE,FUCALES, PHAEOPHYCEAE)

Silvetia compressa ( J. Agardh) Serrão et al. is a common member of the upper intertidal fucoid community on the Pacific coast of America from Humboldt County, California, to Punta Baja, Baja California, Mexico. A relatively narrow range of morphological variability is exhibited by most mainland populations, regardless of latitude, but some mainland populations and all insular populations participate in a complex pattern that we have attempted to analyze. A few populations on the Monterey Peninsula in which the fronds are atypically delicate were described by Setchell & Gardner as f. gracilis, to which was assigned a population from Santa Catalina Island. After comparing populations from various parts of the range of the species, including all of the Channel Islands, we conclude that two subspecies may be recognized. In subsp. compressa, which includes f. gracilis as a growth form and occurs chiefly on the mainland, the frond is robust with long tapered receptacles. In the variant subspecies, which is chiefly insular but also occurs on the coast of northern Baja California, the typical frond has slender axes as in f. gracilis, but is more densely branched and has short ellipsoidal receptacles. Comparison of nucleotide sequences from the ITS regions of rDNA revealed an identical pattern for subsp. compressa from Baja California and central California, including populations assignable to f. gracilis. By contrast, the pattern for the variant subspecies differed by 2 bp (0.3%) from that of subsp. compressa.