1,720,995 research outputs found
A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex
No full textBaumsteiger, Jason, Moyle, Peter B. (2019): A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex. Zootaxa 4543 (2): 221-240, DOI: 10.11646/zootaxa.4543.2.
FIGURE 1 in A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex
No full textFIGURE 1. Historic, present, and translocated ranges of Hesperoleucus taxa in California. The range of the Northern Roach in Oregon is largely omitted. Overall range of each taxon is based on watersheds, so are only approximate, and should not be used for precise locational information. Representatives of each taxon may occasionally occur in areas from which they were historically present but are now shown as absent, such as low elevation regions of the Central Valley. The Cuyama River (southern California) apparently supports a population of Hesperoleucus, but its origins and relationships are unknown so it is omitted from this map.Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach/Hitch (Cypriniformes, Cyprinidae, Hesperoleucus/Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on page 227, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus symmetricus subsp. symmetricus symmetricus (Baird & Girard 1854
No full textHesperoleucus symmetricus symmetricus (Baird & Girard, 1854) Central California Roach This subspecies is consistent with the descriptions and distribution of the species itself, minus the Red Hills region. According to Baumsteiger et al. (2017), population structure is available within the subspecies, but only one lineage (see below) stood out in the genomic analyses and therefore would rise to the level of a DPS by our estimation. However, the presence of population structure and morphological differences (Brown et al. 1992) within the subspecies indicates isolation/adaptation among locations that, if fully investigated, may warrant additional DPS designations. We would argue this should be investigated on a case-by-case basis to evaluate the level of isolation and potential threat to each population so as to not overuse the DPS designation within the U.S. Endangered Species Act. Kaweah Roach. The current population of Kaweah Roach is highly isolated in the Kaweah River system, at the southern end of the CA Roach range. Some morphometric characters are distinctive in Kaweah Roach, including a high prevalence of chisel-lip (Brown et al. 1992). Although found to be genetically (Aguilar & Jones 2009) and genomically distinct (Baumsteiger et al. 2017), these determinations were only discovered once analyses were restricted to CA Roach samples. Therefore, we believe Kaweah Roach is a unique evolutionary lineage within CA Roach, representing a distinct population segment (DPS) that should be managed accordingly. Description. Kaweah Roach are largely similar to the description for CA Roach, with slightly different morphometric and meristic mean counts reported (Brown et al. 1992). This includes a higher prevalence of chisellip, similar to Red Hills Roach. Distribution. This distinct population segment appears to be isolated in the Kaweah River above Kaweah Reservoir. Historically it may have occurred on the Valley floor, in places such as Tulare Lake or in the lower reaches of nearby river systems such as the Kings River. Such access is now denied by dams, diversions, and altered habitats. Status. This DPS is reasonably safe, at least in the short run (50 years). The upper half of the range within the Kaweah River is bounded by Kings Canyon National Park and as such receives special protection. The lower half, which flows mostly through private lands before ending up in Kaweah Reservoir, is less secure. Individuals here may be at risk from upstream movement of invasive species such as Largemouth Bass, Micropterus salmoides, a factor which will intensify with climate change and warmer temperatures.Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 228-229, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Lavinia
No full textLavinia and Hesperoleucus One conundrum within the Hitch–Roach species complex is uncertainty regarding whether species should be placed in one genus or two genera. While the complex was originally divided between Lavinia and Hesperoleucus, multiple examples of hybridization between the genera suggest they are more closely related than separate genera designations would normally indicate. For example, CA Roach are known to hybridize with Hitch in the Pajaro and Salinas rivers in Monterey Bay drainages (Miller 1945), in Coyote, Alameda, and Walnut Creeks in the San Francisco Bay region (Miller 1963; Leidy 1984, 2007, pers. comm.), and in Sacramento-San Joaquin drainages (Avise et al. 1975; Jones 2001). Hybrids were fertile in the Pajaro River; Avise et al. (1975) found 8% of Hitch examined to be F 1 hybrids and 5% to be backcrossed. Since it is unusual for two genera to be interfertile and because other endemic species in California do not appear to exhibit this behavior (Avise et al. 1975; Avise & Ayala 1976; Moyle & Massingill 1981; Moyle 2002), it has been proposed these two species belonged in a single genus. Aguilar & Jones (2009) found the two genera are closely related mtDNA lineages that could be considered to be one genus. Schönhuth et al. (2012), however, noted that while their mtDNA analyses supported this close relationship, their limited nuclear DNA results did not. Our recent study (Baumsteiger et al. 2017) found the strongest genomic signal coincided with splitting samples into two genera. We discovered, however, that samples from the Pit River (Northern Roach) were a confounding factor, having mixed historical ancestry with both genera. It was estimated in one analysis that approximately 80% of the Northern Roach genome clustered with Hitch whereas only 20% clustered with CA Roach. But all subsequent analyses found Northern Roach to be a distinct species, on par with currently identified species. Thus Northern Roach appears to be the result of an ancient hybrid speciation event (Mallet 2007) between ancestral Hitch and CA Roach but it is currently genomically different enough to warrant species status. It is our conclusion that despite hybridization between individuals in the different genera, the distinctiveness of each genus is still strongly maintained and prevalent throughout the genomes interrogated. If results were more admixed, the argument for a single genus would be more plausible. We recognize admixture between genera is uncommon but note that not all mutations lead to post-zygotic barriers (Abbott et al. 2013). Additionally, hybridization has been proposed between Roach/Hitch and other members of the subfamily Leuciscinae including Speckled Dace (Rhinichthys osculus), Arroyo Chub (Gila orcuttii), and the now extinct Thicktail Chub (Gila crassicauda) (Greenfield & Deckert 1973; Hopkirk 1973). Thus the frequency of hybridization suggests reproductive barriers are poorly developed in the endemic minnows of California. Secondly, although Northern Roach represent an ancient hybridization event and cluster primarily with individuals from the Lavinia genus (~80% but only in the admixture analysis), we cannot definitively say they are Lavinia. Northern Roach still strongly resemble CA Roach in appearance (see below), but this may simply be the result of natural selection for the CA Roach morphotype. Future phylogenetic analyses between all genera within the family may provide the solution but it remains, as yet, unavailable. Thus, without universal agreement between genomic analyses or improved criteria for taxonomically delineating hybrids, we default to the original genus assignment by Snyder (1913). Therefore, the species of Hesperoleucus are as described below.Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 223-224, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus parvipinnis Snyder 1913
No full textHesperoleucus parvipinnis Snyder, 1913 Gualala Roach Holotype: USNM 74466. Originally described from Gualala River, Sonoma Co., CA. Paratypes: CAS-SU 14903. Gualala Roach were first collected by Snyder (1908) who recognized them as Rutilus symmetricus but noted they bore “a distinctive local stamp by which they can be recognized without difficulty (p 175).” He later (1913) described them as H. parvipinnis. Murphy (1948) suggested these individuals constituted a subspecies not a species. Aguilar & Jones (2009) found them to be strongly genetically different from any other Roach along the California coast, suggesting they were more than a subspecies. Baumsteiger et al. (2017) found individuals within the Gualala system to be distinctive at the species level in all analyses, with pairwise F ST estimates between these individuals and Northern Roach to be the highest in the study (0.895). Thus the consensus reconfirms what Snyder (1913) proposed, that Gualala Roach are indeed a distinct species within Hesperoleucus. Description. Adult Gualala Roach are small (typically 50–80 mm) and bronzy in color, similar in general appearance to California Roach (H. symmetricus). However, they differ from other Roach species by having smaller scales (54–65 along the lateral line), shorter, more rounded fins, a shorter snout (in relation to head length) and a more robust body (Table 1). Gualala Roach have 7–8 dorsal rays (average 8) and 6–8 (average 7) anal rays (Hopkirk 1973). Snyder (1913 p. 66) described Gualala Roach as having “a light lateral stripe two scales wide extending from upper edge of gill opening to base of caudal and entirely above the lateral line; below is a somewhat wider dark stripe, which in turn is followed by several narrower and very distinct dark stripes which grow lighter ventrally.” Distribution. Gualala Roach are confined to the Gualala River and its tributaries in Sonoma Co. They are the most abundant fish species in the South and Wheatfield forks and in most headwater streams (Entrix 1992; DeHaven 2008). Occurrence is greatly reduced in the colder North Fork (Parker and Pool 1964; CDFG 1991) and in the mainstem below its confluence with the North Fork (Kimsey 1953; DeHaven 2008). Status. Historically, Gualala Roach were present throughout the Gualala River basin (Higgins 1997). Agency salmonid surveys indicate Roach may have increased in abundance due to habitat alterations favorable to warm water-tolerant species (Higgins 1997). Moyle et al. (2013) rate this species as “highly vulnerable” to extinction due to climate change. Statewide this is a species of Special Concern (IUCN status, Near-threatened) because of its limited distribution and increasing threats from vineyard and rural residential development, including introductions of non-native fishes (Moyle et al. 2015).Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on page 236, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus symmetricus subsp. serpentinus Baumsteiger and Moyle 2019, new subspecies
No full textHesperoleucus symmetricus serpentinus Baumsteiger and Moyle, new subspecies Red Hills Roach (Fig. 2) Holotype: Museum of Wildlife and Fish Biology, University of California, Davis, WFC3050, 51.8 mm SL, Horton Creek, Tuolumne County, CA, July 16, 2010. Paratypes. WFC3241–3244, 30.4—40.4 mm SL, same location and date as the holotype, The Red Hills Roach has strong support for a subspecies designation (e.g. Brown et al. 1992; Jones et al. 2002; Moyle 2002), especially given its increased isolation from other populations by New Don Pedro Reservoir (constructed in 1971), unique environmental constraints, and the prevalence (~97%) of the chisel-lip morphotype. Genetic and genomic analyses (Aguilar & Jones 2009; Baumsteiger et al. 2017) also found Red Hills Roach to be highly distinct, easily distinguishable from other CA Roach. In fact, our genomic study indicated a clear break between Red Hills Roach and all other CA Roach sampled in every analysis. With so many lines of support, subspecies status is clearly warranted. Diagnosis. Distinguished from other subspecies of H. symmetricus by their flattened body profile, smaller interorbital distance and fewer pectoral and pelvic rays (Jones et al. 2002; Table 1). The holotype has 8 dorsal rays, 8 anal rays, 12 pectoral rays, and 8 pelvic rays. Description. The Red Hills Roach is a genetically and morphologically distinct minnow adapted for living in small streams flowing through serpentine outcrops. They are typically small, less than 60 mm SL (Brown et al. 1992; Jones et al. 2002) and resemble CA Roach in their general appearance. Red Hills Roach exhibit an elevated frequency of “chisel-lip”, a keratinous projection on the lower lip used for scraping algae (Brown et al. 1992). However there is substantial temporal variation in frequency of the chisel-lip condition amongst CA Roach (Jones et al. 2002). Spawning coloration appears to be similar to that of CA Roach but is more intense. W. J. Jones (pers. comm.) describes it as follows: “The body is dark brown to brassy above, dark black lateral band, and brilliant white below, splashed with black blotches on the sides. Dorsal and caudal fins [are] dark olive-brown to reddish-brown, with the rays often deep-olive and with the nearly clear inter-radial membranes faintly flushed with brassy color; pectoral fins [are] yellowish with orange-red axils and very strong orange coloration at base; anal and pelvic fins [are] bright orange-red at the base with lessening coloration towards the rays. Cheeks and operculum with strong gilt reflections; strong orange coloration is found on the edges of the mouth (especially in males) with some blending into the upper mouth region. A lateral band is more strongly gilt than adjacent parts of body, thus often obscuring the lateral line. In females, the coloration is similar but less intense except for the orange coloration at the base of the paired fins that appears equally intense in both sexes. Males can be distinguished primarily by breeding tubercles on the top of the head.” Distribution. The Red Hills Roach is confined to Six Bit Gulch and its tributary streams: Amber Creek, Horton Creek, Minnow Creek and Poor Man’s Gulch in Tuolumne Co. (Jones et al. 2002). Six Bit Gulch, during rainy periods, flows into the western arm of New Don Pedro Reservoir on the Tuolumne River. In July 2010, Roach were observed in three discontinuous wetted reaches of Horton Creek, which covered approximately 500 meters in total wetted length (P. Moyle, unpublished observations). However, only the lower reach, which extends about 200 meters upstream from the confluence with Six Bit Gulch, appeared to be perennial (as indicated by lush growth of sedges and other riparian vegetation). A natural fish barrier approximately 1.2 km upstream from the confluence likely inhibits Roach from accessing upper Horton Creek. Roach have also been observed in Six Bit Gulch where it is forded by Six Bit Ranch Road and in a pool in Roach Creek. Status. Red Hills Roach merit protection as a threatened or endangered subspecies and rate as an IUCN status of Critical Concern (Moyle et al. 2011, 2015). They have an extremely limited distribution in a very harsh environment and less than 1,000 individuals persist in isolated summer pools (Jones et al. 2002). Their persistence is threatened by fire, depleted stream flows, lack of protection on private lands, and most notably, from invasive fishes. While some habitat is protected in the Red Hills Area of Critical Environmental Concern, much is on private land along a road and remains unprotected. Red Hills Roach has persisted only because of its physiological capacity to endure the extreme environmental conditions found in these small exposed streams. Etymology. Both the common and subspecies name reflect the nature of the rocks through which these small creeks flow. The Red Hills region is named for the color of the soils created by weathering one of the largest outcroppings of serpentine rocks in the Sierra Nevada. Serpentine soils are high in magnesium and other minerals and create conditions in which few species can live. The landscape has a sparse but highly endemic flora (Anacker & Harrison 2012, Harrison 2013) and streams are small and largely unshaded.Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 229-230, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus venustus Snyder 1913
No full textHesperoleucus venustus Snyder, 1913 Coastal Roach Holotype: USNM 74476. Originally described from Coyote Creek, Santa Clara Co., CA. Paratypes: CAS-SU 22492, 24282; USNM 75335. Snyder (1913) originally split this species into three species: H. navarroensis (from the Navarro River), H. venustus (from the Russian River), and H. subditus (Monterey region). These populations were later considered to be subspecies within the single species CA Roach, as suggested by the unpublished study of Murphy (1948). Aguilar & Jones (2009) did not find strong support for these species or subspecies with their mtDNA analysis although their microsatellite analysis found some structure among regions. As mentioned above, Baumsteiger et al. (2017) found clear evidence for populations in all coastal draining watersheds belonging to a single species. Subspecies were also detected but more broadly distributed than previously thought. Watersheds extending from the Russian River north to the Eel River (excluding the Gualala River) clustered as one subspecies and watersheds south of the Russian River (Tomales Bay to the Monterey region) clustered as a second. Description. Based on previous morphological work, the Coastal Roach species has a mixture of characters (Table 1). For a cyprinid, the head is large and conical, the tail deeply forked, the eyes large, and the mouth subterminal, slanting at a downward angle. However, the body can be stout or slender, the snout short or more pointed, the caudal peduncle narrow or thick, and the fins small and rounded or long and narrow. Individuals are typically less than 120 mm total length. The dorsal fin is short (7–10 rays) and positioned behind the insertion point of the pelvic fin. The anal fin has 6–9 rays. Pharyngeal teeth (0,5—4,0) have curved tips which overhang grinding surfaces of moderate size. Gill rakers number 8–11. The scales are small, numbering 47–63 along the lateral line and 32–38 before the dorsal fin. Individuals are usually dark on the upper half of their bodies, ranging from a shadowy gray to a steel blue, while the lower half of the body is much lighter, usually a dull white/silver color. A light lateral stripe approximately two scales wide extends from upper edge of the gill opening to the base of tail and entirely above the lateral line. Below this line is a somewhat wider dark stripe and then several narrower and very distinct dark stripes which grow lighter ventrally. Distribution. Coastal Roach are restricted to coastal watersheds, largely west of the coastal mountain range. The northernmost native population is in the Navarro River, although introduced Coastal Roach exist in the Eel River. Excluding the Gualala River (see below), watersheds south of the Navarro River do not support any Roach populations until the Russian River. Populations are then found intermittently from the Russian River south to the Tomales Bay region. Scattered Coastal Roach are found in streams around San Francisco Bay but are once again absent from coastal watersheds to the south until the Salinas and Pajaro river watersheds of the Monterey region. The population in Soquel Creek (Santa Cruz Co.) is probably introduced from the Salinas River, while the origin of the population in the Cuyuma River (San Luis Obispo and Santa Barbara counties) is uncertain but may be native (Moyle 2002). Coastal Roach are widespread in the Eel River, from an introduction in the early1970s. This introduction was noted by Moyle (2002) and Jones (2002) and is now confirmed by Baumsteiger et al. (2017) and A. Kinziger (pers. comm. 2017). Studies show that Eel River fish originated from the Russian River mainstem and not the recently connected East Fork (via the Potter Valley Project - Power et al. 2015). Status. Due to its wide distribution throughout northern California coastal drainages, the species appears to be in little danger of extinction. However, subspecies or distinct population segments (see below) may be at risk within their particular watershed(s).Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 230-231, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus mitrulus Snyder 1913
No full textHesperoleucus mitrulus Snyder, 1913 Northern Roach Holotype: USNM 74474. Originally described from Drew Creek, Lake Co., OR. Paratypes: CAS-SU 22491; UMMZ 139019. Northern Roach were first collected in 1898 by C. Rutter as Rutilus symmetricus (Rutter 1908). Rutter (1908: 139) stated “We have but few small specimens of this form, the longest being but 3 inches long. They were taken in North Fork Pitt (sic) River near Alturas and at the mouth of Joseph Creek, several hundred miles from where any other specimens of symmetricus have been taken. The form may prove to not to be symmetricus, but we cannot identify it otherwise with the material at hand.” In 1913, Snyder described the Goose Lake drainage fish as Hesperoleucus mitrulus. The first inclusion of Roach from the Pit River as part of H. symmetricus was by Hubbs et al. (1979). Genetically, Aguilar & Jones (2009) found that individuals from these locations were distinct from all other CA Roach. Our genomic study (Baumsteiger et al. 2017) also found all individuals collected from these locations to be distinct at the species-level in every analysis. Description. Northern Roach are small (adult size typically 50–100 mm TL) bronzy cyprinids, similar in appearance to the California Roach (Moyle 2002; Fig. 3). They have a robust body, deep caudal peduncle, short snout and short rounded fins. Northern Roach differ from CA Roach in having short rounded fins and “cup-like” scales (Snyder 1913; Table 1). Snyder (1908) examined 20 fish from Drew Creek, Oregon and all individuals had 8 dorsal rays and 7 anal rays. Snyder found that males had longer, larger fins than did females, especially pectoral fins; the sexes could be differentiated by the ratio of pectoral fin length to body length. Distribution. The original collections by Rutter (1908) consisted of the North Fork of the Pit River and the mouth of Joseph Creek. Snyder (1908) collected widely in the upper Pit River, Goose Lake basin of California and Oregon, and in the Summer, Abert, Harney and Warner basins of Oregon but only found Roach in tributaries to Goose Lake, Lake Co., Oregon. The first inclusion of individuals from the Pit River as part of this species was by Hubbs et al. (1979). While no mention is made of a range extension for the taxon, it is assumed this change was precipitated by the 1934 collection of 19 roach in the North Fork Pit River near Alturas, Modoc County (unpublished field notes and collections of Carl Hubbs at the Univ. of Michigan as reported in Reid et al. 2003). In the most comprehensive sampling of the Pit system in California to date, Moyle and Daniels (1982) found Roach at only 8% of 261 collection sites. Northern Roach were found in only three widely separated drainages: (1) Ash–Rush–Willow Creek drainage, Lassen/Modoc Co., (2) Bear Creek, tributary to the Fall River, Shasta Co. and (3) Beaver Creek, Lassen Co. All locations are above Pit River Falls (Shasta Co.) which divides the Pit River Basin into upper and lower drainages. The falls are a barrier to fish movement (Kinziger et al. 2016). Historically, the waterfall was the northern range limit for some Sacramento River fishes, such as Tule Perch, Hysterocarpus traskii (Moyle 2002). Roach below the falls had unimpeded access to the Sacramento River system and are assumed to be H. s. symmetricus. However, genetic and morphometric studies have not been conducted and relationships remain uncertain. Status. Northern Roach have been uncommon in California ever since they were first collected in 1898, although by that time much of the Pit River drainage had been heavily altered by grazing, agriculture and logging (Moyle & Daniels 1982). Subsequently, alien fish species that compete with or prey on Northern Roach (e.g. Green Sunfish, Lepomis cyanellus) became widely introduced. Their absence from some streams, however, may be the result of specific habitat requirements. Northern Roach prefer spring pools and swampy stream reaches, unlike the intermittent stream habitats where other species of Roach are found (S. Reid, pers. comm.). Thus in Lassen/Modoc Co., Roach are found in small numbers, inhabiting the weedy margins of streams and, in one case, an isolated spring pond (Moyle & Daniels 1982; S. Reid, pers. comm.). However a fairly recent survey of fishes in the Oregon portion of the Goose Lake watershed found Northern Roach to be relatively abundant in six small creeks (Scheerer et al. 2010). Moyle et al. (2015) rated the Northern Roach as a Species of Special Concern in California and the International Union for the Conservation of Nature (IUCN) lists their status as High Concern. Additionally, these Roach are highly vulnerable to extinction in the next century from climate change effects on their small streams (Moyle et al. 2013).Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 234-235, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
Hesperoleucus symmetricus
No full textHesperoleucus symmetricus (Baird & Girard, 1854) California Roach Lectotype: USNM 191, present designation. This species was originally described as Pogonichthys symmetricus by Baird & Girard (1854) from specimens collected in the San Joaquin River at Fort Miller, Fresno Co., CA. Paralectotypes: USNM 440410; ANSP 5330–33; MCZ 1961; MNHN 0000–0353. The benchmark California Roach is based on Baird & Girard (1854). Individuals primarily occupy streams/ rivers flowing into California’s Central Valley, most of which are isolated to some degree. Groups of California Roach appear to become easily isolated from one another and adapted to local conditions, particularly following anthropomorphic changes to the landscape (Brown et al. 1992). Many of these isolated groups are distinguishable both morphologically (Brown et al. 1992) and genetically (Aguilar & Jones 2009; Baumsteiger et al. 2017). In fact, Baumsteiger et al. found every sample location of CA Roach within our study was distinctive, including support for a new subspecies and a distinct population segment (see below). Baumsteiger et al. (2017) also found California Roach from the Central Valley to be genomically distinct from those along the coast. All analyses show a clear break between these two species. Our conclusion is CA Roach sensu stricto should be limited to individuals from Central Valley drainages and the following formal description reflects this new delineation. Description. CA Roach (H. symmetricus) are small, stout-bodied cyprinids with narrow caudal peduncles and deeply forked tails. Adult individuals usually have total lengths less than 100 mm, although fish up to 150 mm TL have been observed (Moyle 2002). The head is large and conical, with large eyes and a mouth that is slightly subterminal and slants downward. Individuals in some locations, especially those in streams of the Sierra Nevada, develop a keratinous plate on the lower jaw, referred to as a “chisel lip”. The dorsal fin is short (8–10 rays) and is positioned behind the insertion point of the pelvic fin while the anal fin is also short, with 7–9 rays (Table 1). Pharyngeal teeth (0,5,4,0) have curved tips, which overhang grinding surfaces of moderate size. CA Roach are dark on the upper half of their bodies, ranging from a shadowy gray to a steel blue, while the lower half of the body is much lighter, usually a dull white/silver color. Scales are small, numbering 47–62 along the lateral line. Individuals tend to exhibit general (non-nuptial) sexual dimorphism (Snyder 1908; Murphy 1943). California Roach are highly variable in most of their characteristics across their range. Brown et al. (1992) measured 14 morphometric variables from eight watersheds around the San Joaquin Valley, including the southern end of the range (Table 1). Using these variables, a discriminant analysis correctly assigned 76% of the individuals to the watershed from which they were collected, suggesting locations were sufficiently isolated from one another to have local adaptations or phenotypic divergence. ……continued on the next page Distribution. Individuals were historically found in most tributaries to the Sacramento and San Joaquin Rivers, including tributaries in the extreme southern San Joaquin Valley (e.g. Kern River), as well as in fringe habitats (backwaters with dense riparian cover) along the main rivers (Moyle 2002; Fig. 1). Fish were rarely found in streams above 1000 m in elevation, presumably restricted by high gradients and natural barriers. In the Tuolumne River, the upstream limit was historically (and still is) Preston Falls at 885 m elevation. The historic distribution in the upper Sacramento River is poorly understood. Today CA Roach are absent from many streams as the result of habitat change and invasions of alien predators. For example, Moyle & Nichols (1973) and Brown & Moyle (1993) found they were absent from the Fresno River (Fresno Co.) and other tributaries to the San Joaquin River. They were also once common in the mainstem San Joaquin River but are absent today. Moyle (2002) documented their consistent presence in the river at Friant (Fresno Co.) until the construction of Friant Dam in the 1940s. CA Roach are absent from most of their historic range in the Cosumnes River watershed due to the invasion of piscivorous Redeye Bass, Micropterus coosae (Moyle et al. 2003). In the upper Yuba River watershed, they are largely absent as the result of 19 th century hydraulic mining, with the exception of one small tributary to the South Fork Yuba, Kentucky Ravine (Gard & Randall 2004). The range of CA Roach has also been expanded, presumably through ‘bait bucket’ introductions by anglers. For example, Hetch-Hetchy Reservoir (1,162 m) on the upper Tuolumne River, well above a series of natural barriers, supports a large pelagic population (P. Moyle, unpublished observations). This is unusual because CA Roach do not generally live in reservoirs, especially those with predatory fishes. Interestingly, individuals from this location were found to be admixed with Hitch (Baumsteiger et al. 2017), a factor which may explain their persistence. Soquel Creek and the Cuyama River in southern California support presumed introduced populations, although their origins are also uncertain (Moyle 2002). It is possible that the ancestor of the widely-distributed CA Roach gave rise to most other Roach species and subspecies but this has not yet been investigated phylogenetically. The ability of these fish to persist in small, often intermittent tributaries presumably led to their colonization of adjacent drainages in a number of areas through capture of interior headwater streams by erosion or tectonic movements (Snyder 1908, 1913; Murphy 1948; Moyle 2002). Because they are relatively intolerant of saline waters, dispersal to coastal streams could not have occurred through ocean waters. However, connections at low elevations may have been possible in some cases when sea levels were lower and estuaries joined the mouths of rivers (Moyle 2002). Similarly in the giant San Francisco Estuary, exchange between salt-water intolerant populations likely occurred during flood years when river outflows were high enough to create freshwater lenses in the surface waters of the estuary (Ayres 1854; Snyder 1905; Murphy 1948; Leidy 2007). During high water periods, fish may also have been able to disperse through flooded marshes on the fringes of the estuary. Today such opportunities for dispersal are largely absent (Moyle et al. 2012). Status. Moyle et al. (2015) list California Roach as a Species of Special Concern, with an IUCN status of Near–threatened. If CA Roach were a single interbreeding taxon, there would be little danger of immediate extinction because they occur in many streams over a wide area. However, the discovery that CA Roach comprise multiple lineages, some of which are under immediate threats, is of conservation concern. After all, the small, isolated populations that are the most likely to be extirpated also tend to be the most distinctive (Brown et al. 1992). Emerging appreciation for the variation within the taxon (Moyle et al. 1989; Brown et al. 1992; Jones 2001; Jones et al. 2002; Aguilar & Jones 2009; Baumsteiger et al. 2017) has highlighted the need to preserve populations endemic to specific watersheds. On a broader scale, Moyle et al. (2013) regarded the entire taxon as highly vulnerable to extinction in the next century due to massive changes to smaller streams brought on by climate change, the continuing effects of anthropogenic land and water use, and the spread of invasive species.Published as part of Baumsteiger, Jason & Moyle, Peter B., 2019, A reappraisal of the California Roach / Hitch (Cypriniformes, Cyprinidae, Hesperoleucus / Lavinia) species complex, pp. 221-240 in Zootaxa 4543 (2) on pages 224-228, DOI: 10.11646/zootaxa.4543.2.3, http://zenodo.org/record/261777
- …
