Biodiversity and Biogeography of Chthamalid Barnacles from the North-Eastern Pacific (Crustacea Cirripedia)

The biogeography and ecology of the species of Chthamalus present on the west coast of America are described, using data from 51 localities from Alaska to Panama, together with their zonation on the shore with respect to that of other barnacles. The species present were C. dalli, Pilsbry 1916, C. fissus, Darwin, 1854, C. anisopoma Pilsbry 1916 and four species in the C. panamensis complex. The latter are C. panamensis Pilsbry, 1916, C. hedgecocki, Pitombo & Burton, 2007, C. alani nom. nov. (formerly C. southwardorum Pitombo & Burton, 2007) and C. newmani sp. nov.). These four species were initially separated by enzyme electrophoresis. They could only be partially separated by DNA bar coding but may be separated using morphological characters.


Introduction
Small chthamalid barnacles are a major world-wide component of the rocky intertidal zone of tropical and sub-tropical shores, though only a few species penetrate into temperate latitudes. Darwin [1] described Chthamalus fissus from California, but otherwise it was to be assumed that the rest of the Pacific was inhabited by "varieties" of the then world-wide species Chthamalus stellatus. As more specimens reached the museums, Pilsbry [2] distinguished C. dalli of the north-east Pacific temperate region, leaving C. fissus restricted to the warm-temperate region of southern California. Pilsbry [2] also described a curiously asymmetrical species, C. anisopoma from the Gulf of California, (Sea of Cortez) and two new species from the Panama Canal Zone. One, C. imperatrix, has now been assigned to Microeuraphia, but the other, Chthamalus panamensis, has caused severe problems to naturalists because of its apparent wide distribution from Baja California to Peru and to the difficulty of separating it from C. fissus and C. dalli. The apparent lack of distinct morphological characters to separate C. fissus and C. panamensis, especially of differences in cirri, mouth-parts and their setation, led Southward & Newman [3] to suggest that C. fissus, C. panamensis and the Caribbean form of C. bisinuatus (since distinguished as C. proteus, Dando & Southward [4]) should be regarded as varieties of a single Chthamalus populations, either side of the Central American Gap near the Gulf of Tehuantepec, as at least 315,000 years BP. They additionally presented evidence of a more recent northward migration of C. panamensis into Mexico starting 160,000 years BP.
The current descriptions of species in the panamensis complex [2,12] do not cover all the four species known to be present, nor do they adequately describe the range of intraspecific variation. Thus it is still difficult to separate the species in the C.'panamensis' complex. In the present study we describe in detail the morphology of Chthamalus species in the North Eastern Pacific, including C. anisopoma and the four species in the panamensis complex, relate this to the results of enzyme electrophoresis and DNA barcoding, and provide keys to separate all the species of Chthamalus known from the North Eastern Pacific. The one new species described is the "southern" clade of C. southwardorum. We discuss the relationships between the species in the C. fissus group, give guidance on the best characters for identification on the shore and outline the geographical distribution and zonation of Chthamalus species from Alaska to Panama.

Sample Collections
Most of the barnacle samples were collected on the TEP Alpha Helix Expedition, during March-April, 1978 and in shore-based collections immediately after the expedition and in October 1978 by AJS (Fig 1). These collection stations ranged from 32°52.27' N in California to 8°56.37'N in Panama. Subsequently samples were collected from Naos Island (formerly Quarantine Island) Panama (8°55´N), the type locality for Chthamalus panamensis Pilsbry, and from Avenida Balboa, Panama City (8°58´N) by Dr. P. Glynn in 1979 and sent alive by air. Specimens from La Shalpa, El Salvador (13.33°N) were collected by K. de Reimer in 1978. Samples from Isla Sacrificios (TEPE78-14, 15°40.8´N 96°14.6´W) and Farfan Point were sent frozen, by air freight, by Dr. D. Hedgecock. Preserved samples from these sites were also available for morphological examination. Living samples from British Columbia were collected by P.R. Dando in 1980 and 2013. Slide-mounted and dried specimens, collected by Dora P. Henry (University of Washington) from Mexico, Guatemala, Costa Rica and Panama were examined by AJS. They are now housed at the National Museum of Natural History, Smithsonian Institution, USA The type specimens of Chthamalus panamensis Pilsbry were loaned by the Academy of Natural Sciences of Drexel University, Philadelphia. Type and paratypes of the new species described in the present study have been deposited in the Natural History Museum, London (NHMUK) and the Biodiversity Research Museum, Academia Sinica, Taiwan (ASIZCR). The collections used for morphological descriptions, allozyme and mitochondrial DNA studies are listed in Table 1. The descriptions of sites where specimens were examined for details of species distribution and zonation, also where collections were made for allozyme studies of C. dalli are given in S1 Table. Permission to work and collect in the coastal waters of Mexico, El Salvador, Guatemala, Honduras, Costa Rica and Panama during the Tropical Eastern Pacific Expedition 1978 cruise of the R.V. Alpha Helix was obtained by Scripps Institution of Oceanography. There were no restrictions on collecting intertidal barnacles at the date of collection at any site from which they were removed.

Nomenclature Acts
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix "http://zoobank.org/". The LSID for this publication is: urn:lsid:zoobank.org:pub:0-E2F984F-D577-431A-B1FD-DB10B370D383. The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central, LOCKSS, PlymSEA

Morphological Examination
Barnacles were dissected and the opercular plates and the arthropodal characters, including the cirri and mouth parts, were examined by light microscopy. Samples of C. panamensis and an undescribed species in Panama were further investigated using scanning electron microscopy, methods following Chan et al. [15]. Setal terminology follows Chan et al. [16,17].

Allozyme Electrophoresis
Living specimens, from Panama, California, the Baja Peninsula and British Columbia, were maintained at Plymouth in a "tidal" tank at 25°C and fed fresh plankton twice a week. Both the live and frozen barnacles were treated as described by Dando et al. [18]. Enzymes were extracted by soaking the animals, for 2-3 h at 2°C, in a buffer containing 200 mM triethanolamine, 2 mM disodium EDTA, 1 mM 2-mercaptoethanol and 0.25% (v/v) 1-phenoxy-2-propanol, adjusted to pH 7-0 at 20°C. This left the shells and arthropodal structures available for examination. Enzymes were separated by vertical starch-gel electrophoresis using the buffers listed in Table 2 and staining procedures previously described [4,18]. Allozyme mobilities were measured with respect to horse-spleen ferritin, given a nominal mobility of 100, which was added to the samples as a visible marker. The enzymes used were mainly associated with the glycolytic cycle and had previously proved useful in separating species of Chthamalus in cyprids (P. R. Dando, unpublished studies) and newly metamorphosed individuals [19], as well as in adult stages [6,18].
Specimens of Chthamalus montagui Southward, 1976 and C. stellatus were run side by side on the same gels with specimens of the taxa examined in this study. The mobilities for the enzymes in C. montagui and C. stellatus, examined in the course of this study, were within the ranges previously reported [18]. This comparison provided an additional check on the validity of the species and population comparisons based on allozyme mobilities. In addition,

Mitochondrial DNA Analysis
The samples used were preserved in 75% ethanol and examined by mitochondrial DNA analysis. Total genomic DNA was extracted from soft tissues using a Qiagen DNeasy blood and tissue kit (Hilden, Germany) following the manufacturer's instructions. The mitochondrial COI gene was amplified using the universal forward primer LCO1490 [20] in combination with the reverse primer COI-R4 [21]. The amplification reaction mix contained 2 μl template DNA, 12 μl Taq DNA Polymerase Master Mix Red (1.5 mM MgCl 2 ; Ampliqon, Denmark), 1 μM of each primer with ddH 2 O to a total volume of 20 μl. The PCR profile was as follows: 4 min at 94°C for initial denaturation, 30 or 35 cycles of 30 sec at 94°C, 30 sec at 48°C and 1 min at 72°C, and 5 min at 72°C for final extension. The concentrations of PCR products were first evaluated using gel electrophoresis. If the concentration of PCR products was too low for sequencing, a nested PCR was conducted. The PCR products from the primary amplification were diluted to 1% and used for a secondary PCR. The universal barcode primers LCO1490 and HCO2198 [20] were used for the nested PCR using the same conditions for the amplification reaction mix and PCR profile. The PCR products were sequenced on an ABI3730XL Genetic Analyzer with BigDye terminator cycle sequencing reagents (Applied Biosystems, Foster City, CA).
To compare the Chthamalus sequences with those from other regional studies, sequences from Wares et al. [14] and Deng & Hazel [22] were downloaded from GenBank (Table 3) for analysis. Sequences of C. southwardorum, C. panamensis and C. hedgecocki (Table 3) from Pitombo & Burton [12] were also used for comparison. All of the sequences were aligned using MUSCLE [23] implemented in MEGA 5.05 [24] using the default setting and adjusted visually. The best-fit nucleotide substitution model of COI was selected by jModelTest 0.1.1 [25]. Etymology. Pitombo & Burton [12] stated explicitly that Chthamalus southwardorum is named "in honor of Dr A.J. Southward, for his extensive contribution to the knowledge of chthamalids and his careful collecting during the Tropical eastern Pacific Expedition". Unfortunately, the suffix -orum indicates "for men or for man (men) and woman (women) together" (ICZN, 4 th edition, 1999 Article 31.1.2). Since there is clear evidence of a lapsus calami in the original publication (ICZN ART 32.5.1) the name C. southwardorum is not admissible under ICZN Art. 21.13 and has to be treated as an incorrect original spelling that should be corrected to southwardi, taking the authorship and date of Pitombo & Burton [12]. However, Chthamalus southwardi Pitombo & Burton, 2007 is a junior primary homonym of Chthamalus southwardi Poltaruka, 2000 and therefore an invalid name (ICZN Art 57.2) and so a new replacement name is required for C. southwardorum. In the present study, I (BKKC) rename C. southwardorum as Chthamlaus alani, again in honour of Prof. Alan Southward, but using his first name, for his contributions to barnacle distribution, ecology and taxonomy.
Description. Shell depressed, sutures between plates visible (Fig 2A and 2B). External surface white and smooth, sheath white, with ribs at the basal region, not extending to the apex (Fig 2A and 2B). Internal colour of shell white. Orifice oval, scutum and tergum articulated with a marked angle (Figs 2A and 2B and 3D). Tergoscutal flaps orange-brown when alive (Fig  2A and 2B), colour fades out when preserved in ethanol or formalin. Scutum triangular, basal margin about twice as long as tergal margin (Fig 3). Tergal margin with a wide and elevated articular ridge, ridge extending beyond the tergal margin, articular furrow wide. Occludent margin straight, without teeth. Adductor muscle pit oval and deep, not extending to the basal margin. Lateral depressor muscle scar deep and smooth (Fig 3). Tergum trapezoid, external  surface smooth near the apex, basal region with growth lines or striations, spur wide and not obvious, basal margin with 4 rostral depressor muscle crests (Fig 3).
Penis without basi-dorsal point. Distribution. Northern limit probably at about 29°N inside the Gulf of California and 24°4 0' N on the Pacific coast; southern limit unknown, but probably north of 15°N. The holotype and paratypes of C. southwardorum Pitombo & Burton, 2007 were collected in the San Cristovan river estuary, San Blas, Nayarit, Mexico (21°30' 54'' N, 105°15' 53'' W), attached to an oyster shell, low midlittoral zone. Pitombo and Burton stated that the distribution of C. southwardorum was from Bahía Magdalena, Baja California Sur, Mexico to Puerto Chicama, Peru, but they did not distinguish it from the new species C. newmani described below.  Diagnosis. Chthamalus with strongly asymmetrical scuta and terga. Orifice often filled by left or right tergum and scutum. Cirrus I distalmost segment of posterior ramus with 2-3 bidenticulate setae. Cirrus II with bidenticulate setae with basal guards on the 4 distalmost segments of the posterior ramus and with bidenticulate setae with basal guards on the two distal segments of the anterior ramus. White tergoscutal flaps in fresh specimens.

Chthamalus anisopoma
Description. Shell white, depressed conic, external surface strongly and irregularly ribbed (Figs 2A and 6A). Eroded specimens often have the basal region of the shell ribbe d but the apical region is smooth. Orifice elliptical and often closed by the left tergum and scutum (Figs 2A and  6A). Tergoscutal flaps have white edge (Fig 2A) (the white colour is still visible after preservation in formalin or ethanol). Scutum and tergum strongly asymetrical (Fig 6B-6D). Individuals can have the left scutum and tergum larger than the right scutum and tergum or vice versa (Fig 6B  and 6D). The larger scutum is triangular, with equal sides, the smaller one is narrower. The external surface of the scutum is smooth with growth lines near the basal margin (Fig 6C), occludent margin smooth, tergal margin with shallow articular ridge, not extending beyond the tergal margin. Adductor muscle scar oval, lateral depressor muscle scar oval and smooth. Tergum triangular to trapezoid, 3-4 lateral depressor muscle crests on basal margin (Fig 6B and 6D) Segment counts on cirri I-III were based on five specimens collected from La Paz Bay, Baja California (AJS-2). Segment counts on cirri IV-VI were based on a single specimen from La Paz Bay, Baja California (AJS-2) ( Table 4). Cirrus I: posterior ramus with 6 or 7 segments, anterior ramus 7-or 8-segmented, both rami with serrulate setae dominant, the distalmost segment of the posterior ramus with 2-3 bidenticulate setae (Fig 7A and 7B, Table 4). Cirrus II: posterior ramus 5-segmented, anterior ramus 6-or 7-segmented, serrulate type setae dominant on both rami. Posterior ramus with bidenticulate setae with basal guards on the 4 distalmost segments ( Fig 7C and 7D, Table 4), anterior ramus with bidenticulate setae with basal guards on the two distalmost segments (Fig 7C and 7E, Table 4). Cirri III-VI: similar in morphology, serrulate type setae dominant, anterior ramus similar in length to the posterior ramus. Cirrus III: anterior ramus 13 to 19-segmented, posterior ramus 11 to 16-segmented ( Fig 7F, Table 4). Cirrus IV: anterior ramus 17-segmented, posterior ramus 15-segmented ( Fig 7G). Cirrus V and VI: anterior ramus 18-segmented, posterior ramus 17-segmented ( Fig 7H). Intermediate segments of anterior rami of cirri IV-VI have 4 pairs of long serrulate setae and 1 pair of short serrulate setae (Fig 7F-7H).
Maxilla bilobed, with serrulate setae on both lobes ( Fig 8A). Maxillule notched, 2 large and 4 medium sized setae above notch, 16 setae below notch of cutting margin ( Fig 8B). Mandible has 4 major teeth, first tooth separated from the remainder, fourth tooth bi-dentate. Pecten long and straight, with 30 fine teeth, inferior angle tipped with two pointed teeth (Fig 8C and  8D). Mandibular palp rectangular, with serrulate setae on all margins (Fig 8E and 8F). Labrum V-shaped, with~10 small teeth on both sides of the cutting margin ( Fig 8G). The middle notch with dense setae (Fig 8H).
Penis without basi-dorsal point. Distribution. Present in the Gulf of California and on the Pacific coast of Baja California Sur to at least 23°27' N.
Remarks. C. anisopoma is sympatric with C. alani nom. nov. and C. hedgecocki on some shores. C. anisopoma can be easily identified in-situ by its strongly asymmetrical opercular plates, also by its white tergoscutal flaps when open under water (Fig 2A). Specimens examined. AJS-3, 15 specimens, intertidal shores, Point Lobos,Todos Santos, Baja California Sur, Mexico, 23°24.8' N, 110°57' W, 4 Nov 1978; TEPE78-14, 10 specimens, exposed rocks and sheltered boulders, Isla Sacrificios, Oaxaca, Mexico, 15°40.8' N, 96°14.6' W;   Diagnosis. Chthamalus with shell pink-purple to brown externally (white in older and eroded specimens). Outer side of cirrus I with an array of long serrulate setae (one on each segment) on the anterior ramus. Posterior ramus of cirrus I lacks bidenticulate setae. Cirrus II has bidenticulate setae with basal guards on one or two of the distalmost segments of the posterior ramus and the distalmost segment of the anterior ramus. Tergoscutal flaps dull orange-brown in colour, when alive. COI sequence (Gene Bank Numbers: KU356704-KU356705).

Chthamalus hedgecocki
Description. External shell surface pink-purple to white in eroded or bleached specimens, surface ribbed but ribs do not extend to apex (Fig 2C). Inside of shell pale brown to white. Orifice elliptical. Scutum and tergum slightly asymmetrical, scutum and tergum articulated transversely ( Fig 9A). In older specimens the outer surface of the scutum and tergum was eroded, horizontal growth lines visible on the basal region (Fig 9B-9F). Scutum triangular, tergal margin about half the length of the basal margin, articular ridge not, or seldom, extending beyond the tergal margin, articular furrow deep. Occludent margin without teeth, smooth. Basal margin slightly convex, with obvious lateral depressor muscle crest (Fig 9). Tergum trapezoid, scutal margin slightly convex, spur wide and blunt, carinal margin with 4-5 lateral depressor muscle crests (Fig 9B and 9E).    (Table 4). Cirrus I: posterior ramus 5-7 segmented, anterior ramus 7-to 9-segmented. Outer side of anterior ramus with a longitudinal array of long, serrulate setae, a single seta on each segment, posterior ramus setae serrulate, no bidenticulate setae (absent from all specimens examined, Fig 10A). Cirrus II: posterior ramus 5-to 6-segmented, anterior ramus 5-to 6-segmented, first or first and second most distal segments of the posterior and the most distal segment of the anterior ramus with bidenticulate setae with basal guards (Fig 10B and 10D, Table 4). Cirrus III: anterior and posterior rami similar in length, anterior ramus 13-to 19-segmented, posterior ramus 14-to 17-segmented, intermediate segments have 3 pairs of long setae and 2 pairs of short simple setae (Fig 10E, Table 4). Cirrus IV and V: anterior and posterior rami 16-segmented, intermediate segments of both rami carry 3 pairs of long serrulate setae and 2 pairs of short simple setae (Fig 10F and 10G). Cirrus VI: anterior and posterior rami 19-segmented ( Fig 10H).
Maxilla with serrulate setae (Fig 11A). Maxillule notched, two large setae above notch, notch wide and shallow, with 4-5 setae in notch, region below notch protuberant, with > 14 setae along the protuberant cutting edge (Fig 11B). Mandible has 4 major teeth, fourth tooth bi-dentate ( Fig 11C). Pecten long, with 24 very fine teeth, inferior angle with a pair of pointed teeth ( Fig 11D). Mandibular palp rectangular, with serrulate setae on superior margin and tip (Fig 11E and 11F). Labrum with V-shaped cutting edge, without sharp teeth on either side of the cutting margin, central part of the notch carries dense setae (Fig 11G and 11H).
Penis without basi-dorsal point. Remarks. Pitombo & Burton [12] reported that the outer surface of the tergum in C. hedgecocki posesses a weak longitudinal furrow, frequently only on one plate. In the present study of specimens from Point Lobos, Baja California, the tergum of C. hedgecocki is smooth externally and without any longitudinal furrows. The arthropodal characters of C. hedgecocki in the present study agree with the descriptions in Pitombo & Burton [12]. The shallow longitudinal furrow in at least one of the the terga of C. hedgecocki may be a variable character and may not be a diagnostic character for species identification. Diagnosis. Chthamalus with cirrus I having bidenticulate setae with or without basal guards on the distal segment of the posterior ramus and cirrus II having bidenticulate setae with basal guards on the distalmost two or three segments of the posterior ramus and two distalmost segments of the anterior ramus. External surface of tergum without tergal furrow. Tergoscutal flaps brown when alive. COI sequence (Gene Bank Numbers: KU356709-KU356722).
Description. The hard parts were described mainly based on the holotype, supplemented with information from paratypes. Shell depressed, white to pale brown, composed of six plates, radii absent. Orifice elliptical (Fig 2D, 2E and 2F). Inside of shell white to pale brown. Scutum and tergum symmetrical, with tergoscutal flaps brown in colour when alive (Fig 2D, 2E and 2F), dull brown when preserved in formalin or ethanol. Scutum triangular, outer surface of scutum with upper part eroded, with growth lines as striation patterns on the lower basal region.  Inner side of scutum white, tergal margin with wide articular ridge and deep articular furrow. Basal margin slightly convex, lateral muscle scar present as 1 or 2 pits. Occludent margin straight without dentations, adductor muscle scar oval and deep (Fig 12). Tergum narrow, height of tergum about 1.5 times greater than the width, tergal margin about 0.6 of the length of the basal margin (Fig 12). External surface eroded with growth lines visible near the basal margin, inner side white and smooth, scutal margin strongly concave, carinal margin convex, basal margin with 4 rostral depressor muscle crests. Spur wide and bluntly convex (Fig 12).
Maxilla, two lobes are separated by a notch without setae, serrulate setae on both lobes (Figs 16A and 17A-17C). Maxillule cutting margin notched, with two large setae above notch, five setae below notch, inferior angle protuberant, with dense setae (Figs 16B, 17D and 17E). Mandible has 4 major teeth, first tooth separated from the remainder, the third and fourth teeth bidentate, pecten with 14 very small teeth, inferior angle tipped with a pair of pointed teeth (Figs 16D and 16E and 17G-17K). Mandibular palp rectangular, superior margin with dense serrulate setae, inferior margin with long serrulate setae (Figs 16F and 17L, 17M and 17N). Labrum concave, with six teeth on each side of the cutting margin, central part of notch with dense setae (Figs 16G and 16H and 17P and 17Q).
Etymology. Named (by BKKC) in honour of Prof. William Newman, for his extensive contributions in barnacle taxonomy and his leadership of the TEP Expedition (see Southward [32]).
Remarks. Hedgecock [5] collected specimens which he designated "FB Chthamalus fissus" from Farfan Beach, Panama and found identical specimens at Isle del Tigre, Golfo de Fonseca. Using allozyme electrophoresis, he showed these to be a separate species from C. fissus of California and the species collected at Isla Sacrificios and Chamela Bay. Specimens from the Farfan    Beach collection of Hedgecock were examined by electrophoresis and by morphology in the present study and found to be C. newmani.  anterior and posterior rami bears bidenticulate setae with basal guards, anterior ramus of cirrus III is about 1.5 times longer than the posterior ramus. Tergoscutal flaps orange. COI sequence (Gene Bank Numbers: KU356706-KU356708).

Chthamalus panamensis Pilsbry, 1916
Description. Shell pink to purple in unbleached or uneroded specimens, otherwise white, with six plates (Figs 2E and 2F and 18A and 18B). Surface with ribs extending to the apex. Orifice elliptical. Scutum and tergum articulated slightly obliquely. Scutum triangular, basal margin twice the length of the tergal margin, external surface with horizontal growth lines, and the upper portion often eroded, tergal margin with shallow articular ridge, not extending beyond the tergal margin, occludent margin straight to slightly convex, with teeth on proximal 1/3 of the margin in uneroded specimens (Fig 18C and 18D). Adductor muscle pit oval, not extending to basal margin. Lateral depressor muscle pit deep and smooth (Fig 18D-18F). Tergum triangular to trapezoid, external surface with horizontal striations and a deep longitudinal furrow is visible in the basal region (Fig 18D-18F). Such a furrow can be observed from above through the orifice of uneroded, undissected specimens (Fig 18A and 18B). Inner surface of tergum smooth, scutal margin straight, spur flat and not obvious, depressor muscle crests 3-4 (Fig 18).
Penis without basi-dorsal point.
Distribution. Pacific coast of Central America from about 14°N to 8°15' N (southern limit unknown).
Remarks. Specimens of C. panamensis collected for the present study agreed with the syntypes of C. panamensis described by Pilsbry [2] in having a deep groove on the outer surface of    the tergum, which is the key diagnostic feature of C. panamensis. This morphological feature of the tergum was also reported in Pitombo & Burton [12].
The sympatric C. newmani sp. nov is distinguishable from C. panamensis in situ by its lack of deep tergal grooves. In addition, it has has brown tergo-scutal flaps while those of C. panamensis are orange. The flaps are visible in the living barnacle when the orifice is open under water (Fig 2E and 2F). Arthropodal character differences are: Cirrus I of C. newmani has bidenticulate setae, Cirrus I of C. panamensis does not. Cirrus III of C. newmani has equal rami. Cirrus III of C. panamensis has unequal rami. This feature of cirrus III in C. panamensis was not reported by Pilsbry [2] or Pitombo & Burton [12].

Distribution and Zonation of Chthamalus spp. in the Eastern Pacific
The distribution and intertidal zonation of the chthamalids at the sites studied, on the TEPE78 cruise and from other collections, are listed in in S1 Table, with particular reference to the zonation of other barnacle species and the species' geographical distribution is shown in Fig 23. At the northern sites, in Alaska, and British Columbia, no species in the Chthamalus 'fissus' group were present and only Chthamalus dalli was found. The range of Chthamalus dalli extends south to southern California, where it co-occurs with the true C. fissus and is confined to sheltered crevices on the most exposed rocks on the open coast.
Chthamalus anisopoma was mainly restricted to the Gulf of California with a few present on the Pacific coast of Baja California Sur. In the Gulf of California, C. anisopoma was concentrated above the zone of Tetraclita sp., from the main zone at mid-tide level (MTL) to high water spring tide level (HWS), where it was restricted to crevices and under overhangs.
Chthamalus alani was found on the eastern side of the Gulf of California from Bahia Kino (28°51.55' N 112°1.70' W) southwards. On the western side of the Gulf it was found in the La Paz region. Where Tetraclita sp. co-occurred with C. alani at more exposed sites, C. alani was restricted to crevices. On the Pacific coast of Baja California the distribution of C. anisopoma overlapped with that of C. hedgecocki, at least as far north as Pt. Lobos, 23°26' N, where C. alani was also found in sheltered locations.
South of the Gulf of California as far as the Gulf of Tehuantepec, most of the rocky coastal sites are very exposed. Chthamalus hedgecocki was the dominant chthamalid present, with C. alani occurring only in sheltered locations. There was an abrupt species change south of the Gulf of Tehuantepec, with C. panamensis and C. newmani replacing C. hedgecocki and C. alani, respectively.
Both Chthamalus alani and C. newmani were generally absent from shores that were exposed to heavy wave action. When present they were found in crevices and on the sheltered side of rocks. In contrast, C. panamensis and C. hedgecocki were absent from the most sheltered shores. All four species were most abundant between MTL and high water neap tide level (HWN), often being confined to crevices or shady areas at HWS.

Molecular Analyses
Species identification by enzyme electrophoresis of specimens collected from British Columbia, California and Mexico. Chthamalus specimens from north of Scripps Institution of Oceanography and Pt. Lobos, Todos Santos (both on the Pacific coast) were compared, by enzyme electrophoresis, with specimens from British Columbia, to the north, and from La Paz in the Gulf of California and from Isla Sacrificios, Oaxaca, Mexico to the south (Figs 1 and 23, Table 5 and in S2 Table). The enzyme abreviations used are given in Table 2 and allozymes were named by their mobilities with respect to ferritin, which was given a nominal mobility of 100.  Table 1 and S1 In British Columbia, C. dalli Pilsbry, a species in the challengeri sub-group [33], was the only chthamalid species present and there was no evidence of cryptic species, or significant genetic variation from the specimens of this species sampled there or in California (S2 Table). C. dalli was the most distinctive of the seven species identified along the Pacific coast of North and Central America. It had unique allozymes for Ark1, Ark2, Ald1 and En in all individuals examined and over 99% of specimens could also be identified as a result of unique allozymes for Gpdh, Hk and Pk (Table 5, S2 Table).
Chthamalus fissus Darwin, was common in the Californian samples but was not observed in the specimens screened for electrophoresis from the other sites (S2 Table). The dominant species in the La Paz HWS sample. It was the most distinctive of the six species in the 'fissus' group examined, with four of the14 enzymes studied showing unique allozyme(s) in all individuals, Ald1 (80) En (128), Pgm (67, 74),) and Mpi (88, 100, 109), another two Mdh1 (83, 93) and Gpi (9,17,25,38) showed unique allozymes in over 96% of individuals ( Table 5). The allozyme relative mobilities are given in parentheses.
Chthamalus anisopoma Pilsbry was found in samples screened from Pt. Lobos and La Paz, being particularly abundant at the latter site, but comprising less than 4% of the chthamalids analyzed at Pt. Lobos (Table 6). This species had unique allozyme relative mobilities for Pgdh (115) and En (115) as well as characteristic allozymes, in the majority of individuals, for Gpi (62) and Pgm (76, 80, 87, 94 and 117) Table 5.
Two other chthamalids present were more similar morphologically to one another than to the three above species. The species at Isla Sacrificios and the dominant species at Pt. Lobos differed absolutely from the species found at La Paz in the mobilities of 2 of the 14 enzymes tested, Ark2 (16) and Ald1 (64, 100). The species found at La Paz had Ark2 (42,47,66,68,80), and Ald1 (75, 108). In addition the allozymes of Mdh2, Pgk Pgm and Gpi, could be used to separate > 98% of specimens into one of these two species (Table 5, S2 Table). The dominant species at Pt. Lobos, comprising 93% of the chthamalids present among the specimens sampled by electrophoresis, was absent among the individuals examined from La Paz, while the La Paz species constituted less than 5% of the chthamalids examined from Pt. Lobos (Table 6). The Pt. Lobos and La Paz species both differed in the allozyme mobilities of 5 enzymes, Gpi, Pgm, En, Pgk and Ark2, from C. anisopoma and C. fissus (S2 Table). The dominant chthamalid species from Pt. Lobos was identical with 20 frozen specimens, sent by Dr. Hedgecock, collected from Isla Sacrificios (TEPE-14). Since these latter chthamalids were found to be of the same taxon as that found at Bahia Chamela (BC), TEPE78-21, 19°35.0'N, 105°08.5'W [5], the Pt. Lobos species equates to the BC population of Hedgecock and C. hedgecocki of Pitombo & Burton [12] that the latter authors also identified from TEPE-14 specimens (Appendix 1 in [12]). C. hedgecocki could be distinguished from all the other TEPE chthamalids by the presence of the unique allozyme Ark2 (16) in all specimens examined. The dominant species in the La Paz HWS sample was Chthamalus alani nom. nov. (= C. southwardorum Pitombo & Burton 2007). It could be distinguished from the other five TEP chthamalids in the 'fissus' sub-group by the absence of the specific allozymes that characterise these five species and from the closely related C. newmani by the presence of the allozyme Gpdh 80 in all C. alani individuals. The distribution of the species in the Mexican samples examined by electrophoresis, is given in Table 6.
Species identification by enzyme electrophoresis of specimens collected from Panama. The chthamalids from Naos Island, Panama City and Farfan Point could be separated into two groups on the basis of the enzyme mobility patterns for three of the 16 enzymes tested, namely Mdh-2, Gpdh and Ark-2 (Table 5 and S2 Table). Any one of these enzymes could be used to assign a specimen to one of two presumed species. In addition the allozymes of Gpi, Pgdh, Pk and Ald could be used to separate > 95% of specimens into one of these two groups (S2 Table).
An examination of the specimens in the two groups revealed that they also showed morphological differences. The species dominant at Naos Island had, in live specimens, an orange tergo-scutal tissue flap and a deep longitudinal tergal furrow, "from apex to spur", as noted by Pilsbry [2] in his description of Chthamalus panamensis. The second species dominated at more sheltered sites, such as Avenida Balboa and Farfan Point. It lacked a tergal groove and live individuals had brown tergoscutal flaps. This species is named Chthamalus newmani sp. nov. in the present paper.
C. panamensis had characteristic allozymes for Mdh2 (46), Gpdh (80) and Ark2 (18,31,55). In contrast, the comparable allozymes for C. newmani were Mdh2 (72), Gpdh (78) and Ark2 (42,61,68). The specimens from Farfan Point were part of the sample examined by Hedgecock [5] and were referred to by him as FB. Thus Chathamalus FB in Hedgecock [5] is C. newmani. The distribution of the two Panamanian species, in samples examined by electrophoresis, is shown in Table 7. Comparison of Chthamalus species in the 'fissus' sub-group. Side by side electrophoretic comparisons of the enzymes of C. fissus Darwin, C. alani nom. nov., C. newmani sp. nov., C. panamensis Pilsbry, C. hedgecocki Pitombo & Burton and C. anisopoma Pilsbry with C. proteus Dando & Southward and C. fragilis Darwin from the Gulf of Mexico and C. dalli Pilsbry from California showed that all were quite distinct. Nei's statistics of genetic identity and distance [34] can be used to quantify the average differentiation between species based on allozyme data. These statistics (Table 8) clearly demonstrate the close relationship of C. panamensis with C. hedgecocki and of C. alani with C. newmani. No specimen showed a pattern for any of these enzymes that would suggest that it might be a hybrid between two species. This was additional evidence that the species are genetically distinct.
The observed mean heterozygosity for Chthamalus hedgecocki was 0.075 compared with 0.100 for C. alani and 0.093 for C. fissus and 0.139 for C. anisopoma. The mean heterozygosities for the Panamanian species were 0.077 for C. panamensis and 0.212 for C. newmani.
Mitochondrial DNA analyses. A total of 621 COI nucleotide sequences were obtained (Table 3). Of these, 136 nucleotides were variable and 98 nucleotides were parsimoniously informative. The best-fit nucleotide substitution models for the COI data set were TIM1+G. The genealogical topology showed three reciprocal monophylies, corresponding to C. panamensis and C. hedgecocki, C. newmani sp. nov. and C. alani (Fig 24). The evolutionary distances based on mean K-2-P distance were 2.1% within C. panamensis and C. hedgecocki, 0.6% within C. newmani, and 0.8% within C. alani. The evolutionary distances based on mean K-2-P distance between three lineages were 3.6% between C. alani and C. newmani, 8.2% between C. hedgecocki & C. panamensis and C. newmani, and 8.6% between C. hedgecocki & C. panamensis and C. alani.

Morphological Identification of Chthamalus Species in the North-Eastern Pacific
The taxonomy of Chthamalus species is often confusing, due to the high diversity of species and their similarities in morphology. In the present study it was found that all seven species of Chthamalus in the North-Eastern Pacific, C. anisopoma Pilsbry, 1916, C. alani nom. nov. (= C. nov.and C. panamensis Pilsbry, 1916, could be distinguished by: 1) the presence or absence of bidenticulate setae on cirrus I. However, given that in the related Atlantic species, C. proteus and C. fragilis, individuals vary in the presence or absence of complex seae on cirrus I and in the presence or absence of basal guards [4] these characters on cirrus I should be used with caution until more specimens from a variety of sites have been examined. 2) the number of segments bearing bidenticulate setae with basal guards on rami of cirrus II. 3) the colour of the tergoscutal flaps. 4) the tergum morphology. 5) the shape and number of teeth on the labrum (Table 9). Among the five species described in detail in this paper, Chthamalus panamensis and C. hedgecocki are morphologically similar, in that both species lack bidenticulate setae on the rami of cirrus I. They both have a longitudinal array of setae on the outer surface of the anterior ramus of cirrus I, as in C. fissus but unlike the four other Chthamalus species in the North-Eastern Pacific. C. panamensis is unique among the seven northern Pacific species in having a deep longitudinal furrow on the external surface of the tergum and in having the anterior ramus of cirrus III about 1.5 times the length of the posterior ramus. In the other species. the two rami of cirrus III are about equal in length.
Chthamalus anisopoma, C. newmani sp. nov and C. alani nom. nov. all have bidenticulate setae on the posterior ramus of cirrus I. C. anisopoma can be distinguished from C. alani and C. newmani by having bidenticulate setae without basal guards on the distalmost segments of the posterior ramus of cirrus I, bidenticulate setae with basal guards on the distalmostfour  Table 9).
Key for identification of Chthamalus species in the North-Eastern Pacific.

Molecular Analysis
We have shown that two sympatric chthamalid species of the 'fissus' group exist in the Gulf of Panama and that a related pair of species occur along the Mexican coast. These four species were found to be distinct by enzyme electrophoresis and morphology but two of them could not be separated by DNA barcoding. DNA barcoding is one of the most popular methods for species identification in recent years, especially for people who are not familiar with traditional morphological-based taxonomy. In this study, DNA barcoding revealed three reciprocal monophylies, corresponding to 1) C. panamensis and C. hedgecocki, 2) C. newmani sp. nov. and 3) C. alani. These four species could also be separated from, C. anisopoma, C. fissus and C. dalli (Fig 24). The DNA barcoding is appropriate to confirm the species status of C. newmani sp. nov. and C. alani nom. nov. from the other members of the 'fissus' subgroup. However, in the case of C. panamensis and C. hedgecocki in this study, the DNA barcoding is not suitable for species identification. Wares et al. [14] used DNA barcoding to identify Chthamalus species in the tropical eastern Pacific and then used 16S, nNAKAS and nEF1 genes for phylogenetic analysis. Their DNA barcoding results for C. hedgecocki and C. panamensis were similar to the pattern in the present study, in that the bootstrap value and posterior probability in their barcoding analysis was low (63-64 for C. hedgecocki; 73-87 for C. panamensis) but the bootstrap value and posterior probability for C. panamensis plus C. hedgecocki in the Neighbour-Joining and Bayesian methods were both 100, which is also similar to the pattern in our study. In our results, C. panamensis and C. hedgecocki, identified by morphology, were mixed in the same COI clade with low sequence divergence, which suggested that these two species are highly similar in their COI genes. Possible explanations include introgression, as these two species might have hybridised in nature, since they have been reported to be sympatric at two sites in Mexico, sampled in 2002 [12]. Molecular analysis of these two species needs further investigation, using more markers for analysis. Meyers et al. [11], based their analysis on the results of Wares et al. [14] and suggested that C. panamensis and C. hedgecocki separated relatively recently, about 400,000 years ago, and therefore it may be possible for them to hybridise. The above difficulties in establishing phylogenetic relationships, based on mitochondrial genes, between species in the Chthamalus 'fissus' subgroup are analogous to those between species of Mytilus in the Atlantic, Baltic and Mediterranean. In the latter genus, haplotype frequency distributions and mtDNA phylogenetic relationships were at variance with systematic hypotheses based on morphological and allozyme data due to hybridization and differential mtDNA introgression [35].
There are clear morphological and allozyme differences between the Chthamalus species found on the Pacific coast between Canada and Panama. All seven Chthamalus species in this region could be separated by their unique allozymes (Table 5). Nei's genetic distances for the allozyme data were 0.262 between C. hedgecocki and C. panamensis and 0.212 between C. alani and C. newmani. In contrast, the genetic distance between C.
hedgecocki vs C. alani was 0.743 and 0.691 in C. panamensis vs C. alani, with the distances between these two outer coast species and C. newmani being 0.737 and 0.785, respectively (Table 8). This confirms the close relationship between C. alani and C. newmani and between C. hedgecocki and C. panamensis.

Identification of Species on the Shore
Considerable plasticity in the shell morphology of cirripeds was recognised by Darwin [1]. Species of Chthamalus, especially those of the 'fissus' group, can be especially variable in shell morphology, hence the naming of Chthamalus proteus (Dando & Southward, 1980) from the Atlantic coast. On the Pacific Coast, C. fissus exhibits three morphs in opercular form [36] and similarly C. anisopoma exhibits two [37].
South of the Gulf of California, all Chthamalus mentioned in published ecological surveys, even from sheltered shores, have hitherto been called either C. fissus [38,39] or Chthamalus panamensis [40,41]. On the Californian coast, where C. dalli and C. fissus are sympatric [42], the two species are still not usually differentiated on the shore, e.g. [43,44], even though Miller et al. [45] described differences in shell coloration, opercular shape and shell smoothness in the two species from Hopkins Marine Station, Monterey Bay, that allowed the identification of 90%, or more, of specimens to their correct species on the shore. Miller et al. [45] also showed that there were slight differences in size, but not morphology, between the larval stages of the two species. C. fissus has also frequently been erroneously stated to be present in the Gulf of California [46].
Although specimens of Chthamalus on the Pacific coast of North and Central America can be identified to species by dissection or by enzyme electrophoresis or DNA sequencing, a guide is needed for identification in the field by shore ecologists. Individual plasticity in shell morphology, due to environmental factors, shell crowding, shell erosion, shell fouling and changes in shape with size, means that not all specimens can be identified in situ. However, the majority of adult specimens can be identified using the comparative features shown in Table 10 and Figs 2 and 25. These include shell colour, the presence or absence of a crenulated margin, the shape of the opercular opening, the presence of a tergal groove and the colour of the exposed part of the tergoscutal tissue flaps. The latter can be viewed after splashing the barnacles with seawater in order to encourage the barnacles to open. The most difficult species to separate, using shell characters, C. alani and C. newmani, may be distinguished by flap colour (Table 10, Fig 2). Two Atlantic coast species, Chthamalus proteus and C. fragilis, especially the former, could potentially colonize the Pacific coast in the future. These species both have brown shells, in non-eroded specimens, and brown tergoscutal tissue flaps with, usually, a lighter brown edge and can thus be distinguished from the Pacific coast species. A guide to identification of the west-Atlantic species is given by Dando & Southward [4].

Ecology and Distribution
Chthamalus species often occur in sympatric species pairs: one preferentially occupying more sheltered shores and the other found at more wave-beaten sites [47]. On the coast of California C. fissus is dominant in protected sites while C. dalli dominates on more wave-beaten shores [48]. This is also true for the species within the 'fissus' subgroup, with C. alani nom. nov. and C. newmani sp. nov. being the sheltered shore species and C. hedgecocki and C. panamensis being the equivalent exposed shore species. In Panama, Laguna [9] found that C. panamensis was the dominant barnacle of the upper rocky shore at both coastal sites and on offshore islands, but was only present in the higher salinity areas of estuaries.
The two sympatric species pairs, C. panamensis with C. newmani and C. hedgecocki with C. alani, showed a similar ecological separation to that shown by another sympatric pair in the NE Atlantic, Chthamalus stellatus with C. montagui [18,47,49]. In all three pairs the first mentioned species preferentially occupies the more open coast sites whereas the second species is found in more sheltered situations including estuaries and sheltered inlets. Similarly, the three open coast species had low mean enzyme heterozygosities (0.075-0.087) compared with 0.100-0.212 for the sheltered coast species. C. anisopoma, with a mean heterozygosity of 0.139, would come within the sheltered coast group and is confined to the Gulf of California and sheltered sites on the Pacific Coast of Baja California Sur. The higher mean heterozygosities in the sheltered-shore species are probably an effect of the more variable inshore environment, for example in temperature, salinity and turbidity, resulting in a more heterogeneous population.
A similar situation occurs in the Gulf of Mexico, with the equivalent species pair being C. proteus and C. fragilis (Dando & Southward, 1980). Like C. fragilis, C. alani and C. newmani (but not C. hedgecocki and C. panamensis) can be found on mangroves. Records of C. panamensis on mangroves in low salinity areas, e.g. [50], almost certainly refer to C. newmani. The distribution of the species at different tidal heights at the studied locations is detailed in SI Table. In the intertidal, Chthamalus hedgecocki and C. panamensis preferentially occur on wave-beaten rocks, with their major zone being from MTL to HWN, with lower densities up to EHWS. In contrast, C. alani and C. newmani were scarce at such sites, dominating the shore at sheltered sites between MTL and HWN, with somewhat lower densities up to HWS and to EHWS in sheltered crevices. The latter two species occur in sheltered locations, such as in crevices among the C. hedgecocki at Pt. Lobos, Baja California and inside the lagoon on mangroves at Estero el Zapote, El Salvador. They occur in sheltered locations, including those with higher turbidity and/or lower salinity than open coast positions, although apparently not being as tolerant of low salinities as Amphibalanus amphitrite, as suggested by the distribution of C. alani in Pichilingue Creek La Paz, although the salinity tolerance of the Chthamalus species has not been investigated. In the high intertidal, the Chthamalus spp. overlap with Microeuraphia imperatrix and M. eastropacensis, especially in sheltered depressions and crevices. However, the degree of exposure to waves and the presence of sheltered crevices and gulleys is a better guide than shore height to the distribution of the Chthamalus species.
We do not have enough detailed observations of the distribution of the exposed and sheltered coast forms where they both occur in sufficient numbers to examine the detailed distributions between the species pairs. In Panama City, on the shore protected from the prevailing SW swells by the San Felipe promontory, below the statue to Vasco Nuñez de Balboa in 1970, a site since infilled, the Chthamalus present at MTL-HWS consisted almost entirely of C. newmani. At Paitilla, 2 km to the west near the Club Union de Panama, the zone of Chthamalus lies above the LW zone of hydroids and bryozoans but below a several metre wide zone of Tetraclita stalactifera panamensis, Amphibalanus amphitrite, B. inexpectatus and B. tintinnabulum, i.e. below MTL [51]. It is likely that this Chthamalus zone consists of C. panamensis, although the sheltered-shore species has been reported alongside C. panamensis at Paitilla Point [12]. North of Naos Island, on the west side of the Avanida Amador causeway, a partially exposed site, both C. panamensis and C. newmani occured side-by-side in the main barnacle zone at MTL, in a ratio of approximately 3:1. In contrast, at the sheltered site of Farfan Point, on the western side of the entrance to the Panama Canal, C. newmani was dominant, with only occasional C. panamensis, from EHWS to below HWN.
Studies on intertidal zonation in the Golfo de Nicoya, Costa Rica showed a bimodal distribution of Chthamalus in the intertidal zone with chthamalids being present higher on the shore at the most estuarine site at Isla Yuca than at the open coast sites at Montezuma and Agujas [40], suggesting the presence of both sheltered and exposed-shore species. Isla del Coco has both sheltered and exposed shores [52] and at Wafer Bay there is also a bimodal zonation of chthamalids [53].
In the present study of samples collected between 1960 and 1980, we found no overlap between the northern species pair, Chthamalus alani nom. nov. and C. hedgecocki, and the southern pair, C. newmani sp. nov. and C. panamensis. The approximately 350 km long sand bar in the Gulf of Tehuantepec separated these two species pairs. As well as a lack of suitable habitat in the Gulf, the strong, dominating, winds are northerly [54] and opposing current systems [55,56] result in surface water flow out into the ocean. This south-trending current will take planktonic Chthamalus larvae away from the coast in the Bay of Tehuantepec, whether they arrive from northern or southern populations. They will be dispersed far from land and have no chance of return within their presumed short planktonic life span of 30-40 days [57]. Although Hedgecock [5] claimed to have found a few of the "BC form of C. fissus" (= C. hedgecocki) south of the Bay of Tehuantepec, in the Golfo de Fonseca, these were probaby a misidentification of C. panamensis, since Hedgecock did not examine any of the allozymes that we found were definitive characters for separating the two species nor did he examine morphological characters. Laguna [9] found C. panamensis, but not C. hedgecocki, at this location.
Within the Gulf of California north of the barrier islands, the only Chthamalus species present was C. anisopoma [22]. This species was mainly distributed at MTL, although height on the shore is partly determined by the rock type, as well as the degree of shade [58]. Along the eastern shore C. alani nom. nov. first appears from Bahia Kino southwards, a distribution since confirmed by samples collected in 2003 and 2006 [22]. C. hedgecocki first appeared on the eastern shore from Punta San Miguel, Sonora, southwards. It was found together with C. alani at Mazatlán, where C. alani was found in tide pools while C. hedgecocki occured outside pools in positions of greater exposure [12]. At that time, the western shore of the Gulf of California had not been well sampled north of the La Paz area (S1 Table) and the distribution of these species further north along this shore was uncertain.
The southern limit of the species in the Chthamalus panamensis complex, south of Panama, is uncertain. Pitombo & Burton [12] reported C. "southwardorum" from Peru (Punta Colon, Paita, and Punta Chicama, Trujillo). However, observations on samples collected south of Panama suggest that from Southern Ecuador southwards different, but related, species of Chthamalus occur (AJS unpublished studies). The northern limit of the C. panamensis complex is probably in California close to 32°52' N, where C. hedgecocki has been reported (the BC form in Hedgecock [5]). Here, C. fissus and C. dalli were the dominant Chthamalus species. The overlap between C. hedgecocki and C. fissus extends southwards along the length of the Pacific coast of the Baja Peninsula to Cabo San Lucas, 22°52' N [59]. C. alani is scarce along this section of coast due to the lack of sheltered shores, the known northern limit for this species being at Bahia Magdalena, 24°38' N.
Chthamalus fissus has a northern limit in the region of San Francisco, 37°35' N [59]. From north of San Franciso to the Aleutians only a single Chthamalus species, C. dalli, is found [59].
Our allele frequency studies on Chthamalus collected from a variety of habitats in British Columbia confirm that sympatric species of Chthamalus are absent from this region S2 Table. The density of Chthamalus on the shore along the Pacific coast has been considered to be largely determined by larval supply, so that areas with upwelling and with offshore currents have poor recruitment [39,[60][61][62][63]. On a local scale factors such as shading, rock type, surge channels, fresh water run off, predation and overgrowth can be important [38,39,44,56,64]. Observations on the development of the larvae of the individual species are needed, since Miller et al. [45] have shown that larval stages of C. fissus cannot develop fully at a temperature of 13°C, suggesting that the distribution of Chthamalus species along the coast may be partly determined by the water temperatures required for larval development.
The Chthamalus specimens from the coast of Mexico examined in the present study were collected mainly in 1978, with one sample from 1980. No specimens of C. panamensis were among these Mexican samples. Similarly Laguna [9,10] did not find C. panamensis at any site along the Mexican coast in any collection made between 1971 and 1980, including the TEPE-78 collections. The species distributions reported here (Fig 23, S1 Table) thus form a baseline against which future intertidal surveys can be compared. The densities and distributions of Chthamalus species have been shown to be particularly sensitive to changes in the environment [65,66].
Since 1980, there is evidence for the presence of Chthamalus panamensis at Mexican sites. Pitombo & Burton [12] found C. panamensis in 6 collections from Mexico made in 2002, at Punta da Mita, Banderas Bay, Nayarit; Punta Tenacatita, Jalisco and one from Pie de La Cuesta, Acapulco, Guerrero, but not in specimens from the 1978 TEPE collections made at Mexican sites, where only C. hedgecocki was found. This may indicate a recent northward expansion of C. panamensis in the tropical Eastern Pacific waters.
Most Chthamalus species are confined to the intertidal and they are not common fouling organisms on ships' hulls, although they have been found above the waterline on ships, barges and floating structures [67][68][69][70] as well as on floating plastic debris in mid-ocean [71]. Among species in the "fissus group", C. fragilis has colonized shores north of Cape Cod, possibly transferred by ships from Carolina in the 1800's [68]. C. proteus spread rapidly among Pacific islands from the west Atlantic post-1973 [69,72] and is expected to reach the Pacific coast of America, if it is not already present [68]. This species, together with C. panamensis and C. "southwardorum", has been found on vessels visiting the Pacific Coast ports of the USA [70].

Conclusions
Our results provide a pre-1980 baseline for the distribution of Chthamalus species along the Pacific coast of North and Central America from Cape Thompson in the subarctic to Panama in the tropics. Seven species were present within this range. C. dalli and C. fissus, as already known, reached their southern limits close to the Californa/Mexico border. Five other species inhabited the warmer regions from Mexico to Panama, each with a distinct distribution and habitat preference. Future intertidal surveys can compare results with the data presented here, since changes in the range and relative abundance of barnacle species are known to be useful early indicators of ecosystem change in response to environmental change. Colonization by species from other geographic areas, especially the Atlantic species C. proteus, may be expected to occur.
Supporting Information S1