Printer Friendly

OLYMPIC TORRENT SALAMANDER (RHYACOTRITON OLYMPICUS) OVIPOSITION SITE WITH NOTES ON EARLY DEVELOPMENT.

We uncovered the 1st Rhyacotriton olympicus oviposition site ever found on 10 August 2016. Torrent salamanders (genus Rhyacotriton), endemic to the Pacific Northwest of North America, have concealed oviposition sites that are rarely encountered. Few data exist on Rhyacotriton oviposition, but torrent salamanders seem to be unique among stream-breeding salamanders in that eggs are deposited unattached to either a substrate or each other (MacCracken 2004; Hayes and Quinn 2005; Karraker and others 2005; Thompson and others 2011). Among the 4 recognized torrent salamander species, descriptions of eggs and oviposition sites are limited to a few observations for 3 of the species: Cascade Torrent Salamander, R. cascadae (MacCracken 2004); Columbia Torrent Salamander, R. kezeri (Nussbaum 1969; Russell and others 2002; Thompson and others 2011); and Southern Torrent Salamander, R. variegatus (Karraker 1999; Karraker and others 2005). Egg and oviposition site descriptions are lacking for the Olympic Torrent Salamander (R. olympicus) (Bury and Jones 2005; Hayes and Jones 2005). Here, we describe the 1st known oviposition site for the Olympic Torrent Salamander, provide preliminary data on the pre-hatching egg development, and compare these data to available descriptions of oviposition sites for other torrent salamander species and amphibians that deposit eggs in concealed sites.

METHODS

Field Site

We discovered the oviposition site described in this paper opportunistically in the course of fieldwork on a landscape-scale study. The structure of this site led us to search for torrent salamanders on 29 June 2016. Our search revealed 8 adult Olympic Torrent Salamanders, including 2 gravid females, leading us to suspect that the stream-adjacent seep in which they were found could harbor an oviposition site. During a subsequent visit to the seep on 10 August 2016, we first discovered eggs judged those of a torrent salamander. This find led us to carefully excavate the core flow area of the seep, which had an elliptical footprint measuring about 42 cm along the slope axis and 30 cm along the stream-parallel axis, and was roughly 18 cm deep. We returned to the site to collect data on embryonic development on 6 September and 11 October 2016. On the latter date, we installed a Hobo Tidbit (*) UTBI-001 datalogger to record seep water temperature at half-hour intervals. The datalogger was buried approximately 10 cm into the substrate, tethered to an immovable root, and removed on 16 January 2017. We measured canopy cover with a spherical densitometer and provide a description of vegetation near the oviposition site.

Laboratory

We collected 5 of the eggs from the oviposition site for developmental study and macrophotography, 3 on 10 August 2016 (1 with twin embryos) and 2 more on 6 September 2016, 1 of which was reared to hatching. Macroscopic observations were made using a binocular compound Leica DM1000 microscope with a C-Mount 0.40X DFC 450 Camera and with a Mustcam (*) UM016 1080P HDMI Digital Microscope. The Leica scope was used for observations and photography of early-stage embryos, mostly with the 1.25X objective, whereas the Mustcam (*) was primarily used for observations and photography of the egg reared to hatching. With the Leica scope, we used the Leica Application Suite 4.4.0 software to measure embryos from the images. For select imaging and observations, we removed some outer jelly capsular layers from some eggs because debris on their outer surface obscured visualizing details of the embryos. We assumed that whole eggs and near-round embryos approximated spheres to estimate their volume. We preserved embryos from which some of the capsular layers were removed in 95% ethanol for future genetic analysis.

To rear a single egg to hatching, we created an incubator by placing a small (3.8 L) cooler inside a large (49.2 L) one. The smaller cooler was the rearing chamber, in which water depth was maintained at ~5 cm. We added 2 Hobo Tidbit (*) UTBI-001 dataloggers to monitor water temperature at 30-min intervals beginning on 8 September 2016. Water for rearing was obtained from the oviposition site stream and an untreated well. We placed an aquarium air pump in the rearing chamber, running continuously to oxygenate the water. The incubator was placed indoors in an unheated, uninsulated shelter. We regulated temperature partly through placement of ice jugs in the larger cooler when ambient temperature was deemed too warm (~13[degrees]C), and with a space heater placed inside the shelter but outside of the incubator when the temperature was judged too cold (~7[degrees]C).

Analysis

We provide means, standard deviations (SD) and ranges to describe selected variables. For analyses, we used Time Series package on JMP[R] version 12.0.0. In particular, we examined the time series of temperature difference between the seep and incubator. We developed the best ARIMA model for that time series, which had an autoregressive order and moving average order of 3 for both (the differencing order was 0). We tested whether this time series differed significantly from a random walk around zero using an Augmented Dickey-Fuller test (ADF).

RESULTS

Study-Site Description

The site is located at an elevation of 189 m along a 1st-order tributary (sensu Strahler 1952) that is part of the headwaters of an unnamed 3rd-order tributary of the Clearwater River, Jefferson County, Washington State, about 16 km northeast of the small community of Clearwater. This stream lies within a non-fish-bearing headwater basin that was clearcut harvested in February through March 2009. The oviposition site was in a stream-adjacent seep (UTM Zone 10, 409656.8E, 5277549.9N, WGS84), 80 m downstream from the stream origin and 1.4 m upslope from the intersection of this lst-order stream and a 2nd-order stream. The seep is located within a 22-m second-growth riparian buffer with 91% canopy cover and a roughly NW aspect (azimuth 320[degrees]). Dominant overstory trees were Western Hemlock (Tsuga heterophylla) with lesser amounts of Douglas-fir (Pseudotsuga menziesii) and Red Alder (Alnus rubra). Salmonberry (Rubus spectabilis) and Sword Fern (Polystichum munitum) dominated the understory; Pacific Golden Saxifrage (Chrysosplenium glechomaefolium) and Five-stamened Mitrewort (Mitella pentandra) dominated the groundcover (Fig. 1A).

Oviposition Site, Egg Description, and Development

On 10 August 2016, we found 10 unattached eggs 8 to 18 cm from the seep surface, scattered throughout a pocket of colluvium composed largely of gravel intermixed with sand and cobble (Fig. IB), and supported by a 25-mm diameter A. rubra root. Each egg was separated from the others by distances of 1 to several centimeters. All eggs except 1 were found at 2 depths: the first 4 eggs at around 8 cm; the remaining eggs were found at depths up to 18 cm with all but 1 at 16 to 18 cm in depth. The saturation zone of the seep, fed from a crack in the bedrock on the slope wall, spread across an area about 0.13 [m.sup.2] and was bounded by bedrock. The 10 eggs averaged 9.5 mm [+ or -] 0.85 mm SD in outer diameter (range: 8.0-11.0 mm). The 11 ova (1 egg had 2 embryos) averaged 4.5 mm [+ or -] 0.53 mm SD in diameter (range: 4.0-5.0 mm) exclusive of jelly layers. A relatively large capsular space surrounded all embryos, extending across almost 60% of the capsular diameter (Fig. 1B), and was estimated to comprise over 5 times the volume of the embryo. Fine debris adherent to the outer surface of the eggs obscured clear imaging of embryos (Fig. 1B). When first discovered, embryos were near-round when viewed perpendicular to the plane of the embryonic axis (Fig. 1C) and ovoid when viewed in the plane of the embryonic axis, but differentiation had progressed as evidenced by presence of the head and at least 19 somites along each side of the vertebral axis (Fig. 1C, Fig. 1D). The ovoid-round asymmetry of embryonic and perpendicular axes, respectively, is evident in the egg with 2 embryos (compare Figs. 1E, Fig. 1F). Dissection of the capsules of the 3 eggs obtained on the initial 10 August date revealed no less than 5 jelly layers surrounding each embryo.

We returned to the site on 6 September 2016 and collected 2 additional eggs. One was dissected and preserved, the other reared in the lab. The embryonic differentiation at the oviposition site had advanced, in particular, heads and tails of those embryos extended well beyond the core embryo/yolk, and reduction of their yolk was substantial, giving them an elongate shape (Fig. 2A). The heart and primordial eye and gill arch region on the head were also evident (Fig. 2A).

During our site visit on 11 October 2016, we found evidence of disturbance from high water flow, and the seep substrate at the oviposition site had been considerably reorganized. Search of the seep failed to reveal any of the remaining 5 eggs.

Captive-Reared Development through Hatching

Developmental data obtained after 11 October 2016 came exclusively from the incubator-reared egg obtained on 6 September 2016. Prior to hatching, the incubator temperature averaged lower and was more variable than for the oviposition site (incubator: 9.5[degrees]C [+ or -] 1.6[degrees]C SD [range: 5.2-14.7[degrees]C]; oviposition site: 9.9[degrees]C [+ or -] 0.6[degrees]C SD [range: 7.1-10.6[degrees]C]; Fig. 3). However, the time series of the temperature difference between the oviposition site and the incubator did not differ significantly from a random walk around zero (ADF = -2.22; AD[F.sub.critical value] = -2.86).

During the incubation of the lab-reared egg, we made observations with the naked eye and a hand lens several times a week. When changes in the eggs development were visible, we inspected the egg with a digital microscope. Here, we describe the observable morphological markers in the development of the incubator-reared embryo (Table 1). At the beginning of captive rearing, the embryo was unpigmented. Pigmentation first appeared in the eyes, then in the skin. The number of melanophores, responsible for skin pigmentation, progressively increased in the dorsolateral region of the abdomen. By 12 November 2016, patterning was apparent in the dorsolateral region, with clusters of melanophores ringed around areas lacking melanophores, creating distinctive "spots" (Fig. 2B). We first observed a heartbeat relatively early on (11 October 2016), concurrent with near-complete pigmentation of the eyes. The gill and limb buds were also becoming visible at this time. The anterior limb buds were becoming visible before the posterior limb buds. As our observations continued, the limb buds slowly increased in size prior to the appearance of digit primordia; these first appeared on the anterior limbs, and <1 wk later on the posterior limbs. Yolk regression was first accompanied by development of a mid-yolk groove and the appearance of a large blood vessel on each side of the yolk body. Ramification of a visible blood vessel network over the yolk body took less than a week. By 12 November 2016, the mid-yolk groove was obscured and the costal grooves were beginning to become evident.

On 4 January 2017, the captive-reared embryo hatched. At hatching, the larva was 14.7 mm snout-to-vent length (SVL), had a 10.6-mm tail, and its head length and head width were 4.6 mm and 2.5 mm, respectively. The melanophore clusters, responsible for creating the distinctive "spots" first visible in the late stage embryo, gave the impression of prominent stripes along the dorsolateral region of the body trunk (Fig. 2E). Melanophores were also denser on the fronto-parietal region on top of the head (Fig. 2E, Fig. 2F). All ventral surfaces lacked melanophores and were essentially white except for a faint yellow coloration resulting from unabsorbed yolk showing through the abdomen.

DISCUSSION

Oviposition in the Olympic Torrent Salamander is similar to that observed for other species of torrent salamander in that the eggs are neither attached to a substrate nor one another [MacCracken (2004) for the Cascade Torrent Salamander; Nussbaum (1969), Russell and others (2002), and Thompson and others (2011) for the Columbia Torrent Salamander; and Karraker and others (2005; Nancy Karraker, University of Rhode Island, Kingston, RI, pers. comm.) for the Southern Torrent Salamander.] Both Nussbaum and others (1983) and Altig and McDiarmid (2015) state that lack of attachment is a distinguishing feature for torrent salamander eggs. Our find underscores this assertion for the genus.

The individual deposition of torrent salamander eggs creates ambiguity in determination of clutch size in the field. In 2 cases, investigators determined that oviposition sites represented multiple clutches (Table 2). Using the mean of 9 ova obtained from gravid females across the range of all 4 species (Stebbins and Lowe 1951; torrent salamanders were regarded as 1 species at that time), Nussbaum (1969) concluded that 32- and 75-egg oviposition sites involved deposition by at least 3 and 8 females, respectively. However, Nussbaum (1969) also noted that the oviposition site at which he found 32 eggs (now attributed to the Columbia Torrent Salamander) consisted of 2 groups of 16 eggs located 25 cm apart. Moreover, he added to the ambiguity about the number of eggs at this oviposition site by later characterizing it as 34 eggs, based on his including 2 additional torrent salamander eggs found in the stomach of a 54-mm SVL giant salamander (Dicamptodon spp.) larva buried 30 cm from this oviposition site. Later, Nussbaum and Tait (1977) estimated that clutch sizes from gravid females (now Cascade and Southern Torrent Salamanders) ranged, respectively, from 2 to 14 and 5 to 16. The 2 groups of eggs that Nussbaum (1969) found may have represented 2 clutches. This possibility becomes even more plausible given that the Columbia Torrent Salamander, with a smaller ovum size (Table 2), may be more likely to have a larger clutch size.

We raise the issue of the difficulty in distinguishing separate clutches in the field for 2 reasons. First, descriptions of oviposition sites have not typically included the spatial relationship among individual eggs or groups of eggs, which may be the only field marker available to differentiate clutches. For example, Altig and McDiarmid (2015) described torrent salamander oviposition as grouped unattached single eggs, but their generic usage of "grouped" is ambiguous. Second, though we think it likely that described oviposition sites with 11 or fewer eggs represent 1 clutch, genetic data are necessary to verify that assertion. Additionally, genetic data in combination with a description of the spatial relationship of eggs would contribute to the understanding of the potential spatial distribution of eggs belonging to 1 clutch. Unfortunately, we were unable to perform such genetic analysis and are therefore unable to prove our assumption that this was a single clutch.

Descriptions of animal size at hatching are sparse for torrent salamanders. However, our sole captive-reared hatchling was similar in size to the few other reports for Rhyacotriton. Specifically, it was slightly smaller (14.7 mm SVL) than the smallest larvae reported for the Cascade Torrent Salamander (15.8 mm SVL) in the Columbia River Gorge and slightly larger than the smallest larvae reported for the Southern Torrent Salamander in coastal Oregon (13.5 mm SVL; Nussbaum and Tait 1977). It was also within the size range of larvae regarded as hatchlings for the Southern Torrent Salamander (14.0-16.0 mm SVL) in northwestern California (Tait and Diller 2006). However, our captive-reared hatchling was much smaller (only 61% of the mean length) than 12 Columbia Torrent Salamander larvae (24.1 mm SVL, range: 22.5-25.8 mm) characterized by Nussbaum (1969) as newly hatched. Difference from the latter is unlikely to reflect rearing conditions because our captive-reared animal had just hatched and had not yet fed.

The size variation of larvae recorded by Nussbaum (1969) may reflect a lack of information on the time since hatching or variation in egg sizes from which they developed, which may be species-, region-, or condition-specific (Table 2). More data on hatchling sizes will be needed to evaluate the large size of the individuals defined as hatchlings by Nussbaum (1969). All other reports of individuals classified as hatchlings were very close to 15 mm SVL. As with hatchling size, the pigmentation of our captive-reared hatchling also matched other recorded observations.

Data on the coloration and pattern for hatchling Rhyacotriton is even more limited than for size at hatching. Based on "very young larvae" obtained over the range of all 4 species, Stebbins and Lowe (1951) reported that they have nearly white venters, identical to our captive-reared hatchling. Nussbaum (1969) also stated that hatchlings have their dorsal melanophores arranged in distinct rings, which form 2 rows along the back; our captive-reared hatchling also had this pattern (Fig. 2E). Nussbaum (1969) added that this longitudinal ringed arrangement was strikingly similar to that in hatchlings of the Northern Dusky Salamander (Desmognathus fuscus) and the Allegheny Mountain Dusky Salamander (D. ochrophaeus), which are plethodontids not particularly closely related to torrent salamanders (Pyron and Wiens 2011). Similar patterns are observed among a number of desmongnathine plethodontids (Altig and McDiarmid 2015), all of which possess stream-dwelling larvae typically found in low-flow habitats, so this may reflect a habitat-linked convergent pattern.

Pre-hatching development of our captive-reared animal was likely faster than natural development in the field, based on other torrent salamander reports. The eggs we found were at Harrison (1969) stage 21, and assuming our 1st survey (29 June 2016) preceded oviposition, we estimate development from oviposition to hatching (4 January 2017) to be less than 189 days, but more than 147 days. However, water temperature in the lab averaged 9.5[degrees]C and at the field site during captive rearing averaged 9.9[degrees]C, above average mid-October to mid-November water temperature. The temperature range above the mean was greater in the laboratory than at the field site, which may have disproportionately accelerated development (DuShane and Hutchinson 1941; Arrighi and others 2013). Nussbaum (1969) estimated time from oviposition to hatching to be around 200 d for eggs in late tail-bud stage (Harrison [1969] stage 31) that were collected on 14 December 1968 from a spring that had 8.3[degrees]C water temperature, reared in the laboratory at 8.0[degrees]C, and hatched between the 2 to 31 May 1969 interval. Karraker (1999) reported that 8 eggs from an 11-egg oviposition site, which were left to develop in a stream enclosure, hatched 193 to 256 d after discovery; temperature data were lacking. Clearly, better temperature data coupled to developmental patterns are needed to elucidate the basis of variation in torrent salamander pre-hatching development. The fact that daily temperature fluctuations can markedly influence both developmental rate and morphology (Arrighi and others 2013) underscores this need. Although not statistically significant, the difference in temperature fluctuated between the oviposition site ranging from 7.1[degrees]C to 10.6[degrees]C and the incubator ranging from 5.2[degrees]C to 14.7[degrees]C over the incubation period (Fig. 3).

We speculate that the relatively large capsular volume in eggs of the Columbia Torrent Salamander (see Fig. 1 in Thompson and others 2011) and the parallel condition described here for the Olympic Torrent Salamander (Fig. 1B) is diagnostic of torrent salamanders and pivotal to their ecology. Nussbaum and others (1983) found 6 jelly layers of varying thicknesses around individual eggs of torrent salamanders. We recorded 5 jelly layers, but our observations, macroscopic and made without staining, could have easily missed a thin capsular layer. We suggest that multiple jelly layers contribute to large capsular volumes and pre-adapt torrent salamanders to their dynamic oviposition habitat. Torrent salamanders tend to occur and are frequently more abundant in stream-network headwaters (Hunter 1998; Wilkins and Peterson 2000; Kroll and others 2008), which are dominated by colluvial processes (Montgomery and Buffington 1998; Gomi and others 2002). We suggest that the lack of egg attachment may potentially restrict torrent salamander oviposition to low-flow colluvial matrices frequently encountered in headwater streams. However, colluvium is often mobile and colluvial processes are dynamic (Gomi and others 2002), presenting a potential risk of compression and injury to eggs laid within its matrices, though torrent salamanders may to some degree be able to distinguish colluvial deposits that are more stable. The large capsular volumes may help protect eggs laid within colluvial matrices if substrate shifting occurs. Hence, focused investigation of the Rhyacotriton egg-colluvium relationship may be useful in informing habitat use in torrent salamanders.

Of the 10 oviposition sites described for the 4 species of torrent salamander, some features of their habitat merit comment. All torrent salamander oviposition sites found to date have been from low-flow habitats in seeps, springs, or low-order streams, frequently in relative proximity (<500 m) to stream origins (Table 3). Given their lack of egg attachment, torrent salamanders may select habitats in which eggs can avoid flow-induced scour in the main channel or higher-order streams. Disturbance of oviposition habitat has resulted in the flow-related loss of eggs (MacCracken 2004), so measuring flow in torrent salamander oviposition sites versus alternative lotic habitats may help define conditions that limit oviposition. Further, all except 1 of the oviposition sites described to date have been found in 2nd-growth managed forests (Table 3), though this may reflect the preponderance of forest studies in those landscapes. Inorganic substrates that characterize torrent salamander oviposition habitat may be more accessible in 2nd-growth forests because of less concealing wood. Further, low-flow, low-order stream habitats may also facilitate finding torrent salamanders because they are intrinsically more accessible in those habitats than in more complex higher-order streams. Hence, whether finding most torrent salamander oviposition sites in headwater areas reflects their relative abundance or greater accessibility is unclear. As pointed out by the late Robert Stebbins in a discussion with one of the junior authors (MPH), for species with cryptic oviposition sites, like torrent salamanders, those encountered may simply be the most accessible (pers. comm.). Hence, when characterizing oviposition sites for such taxa, one must consider how accessible oviposition sites may differ from those that are not.

Though a unique find, our discovery supports collective reports of other Rhyacotrition species in what constitutes suitable oviposition habitat, a low-flow environment in which the eggs are individually deposited within a concealing inorganic matrix. Rhyacotriton olympicus was the last Pacific Northwest stream-breeding amphibian for which an oviposition site had not yet been discovered. Future efforts at understanding Rhyacotriton oviposition should focus on refining the understanding of the range of oviposition site variability and the physical processes that may be important in maintaining suitable oviposition sites. We expect that searches in habitats produced by colluvial processes may be fruitful in this regard.

ACKNOWLEDGEMENTS

We thank CMER and the Forests and Fish Adaptive Management Program for funding support for the Type N Experimental Buffer Treatment Study, the Washington State Department of Natural Resources for access to the site during our fieldwork, D Schuett-Hames (Northwest Indian Fisheries Commission) for vegetation and physical data, and LA Campbell and the Otolith Lab (WDFW, Fish Program) for providing access to and assistance with camera microscope resources. Hai Keren (WDFW, Wildlife Program) assisted with the statistical analysis of the time series.

LITERATURE CITED

ALTIG R, MCDIARMID RW. 2015. Handbook of larval amphibians of the United States and Canada. Ithaca, NY: Cornell University Press. 345 p.

ARRIGHI JM, LENCER ES, JUKAR A, PARK D, PHILLIPS PC, KAPLAN RH. 2013. Daily temperature fluctuations unpredictably influence developmental rate and morphology at a critical early larval stage in a frog. BMC Ecology 13: DOI: 10.1186/1472-6785-13-18.

BURY RB, JONES LLC. 2005. Olympic Torrent Salamander (Rhyacotriton olympicus Gaige). In: Jones LLC, Leonard WP, Olson DH, editors. Amphibians of the Pacific Northwest. Seattle, WA: Seattle Audubon Society, p 66-69.

DUSHANE GP, HUTCHINSON C. 1941. The effect of temperature on the development of form and behavior in amphibian embryos. Journal of Experimental Zoology 87:245-258.

GOMI T, SIDLE RC, RICHARDSON JS. 2002. Understanding processes and downstream linkages of headwater systems: Headwaters differ from downstream reaches by their close coupling to hillslope processes, more temporal and spatial variation, and their need for different means of protection from land use. Bioscience 52:905-916.

HARRISON RG. 1969. Organization and development of the embryo. In: Wilens S, editor. New Haven, CT: Yale University Press, p 44-66.

HAYES MP, JONES LLC. 2005. Rhyacotriton olympicus (Gaige, 1917) Olympic Torrent Salamander. In: Lannoo M, editor. Amphibian declines: The conservation status of United States species. Berkeley, CA: University of California Press, p 880-882.

HAYES MP, QUINN T. 2005. Rhyacotriton kezeri Good and Wake, 1992 Columbia Torrent Salamander. In: Lannoo M, editor. Amphibian declines: The conservation status of United States species. Berkeley, CA: University of California Press, p 876-880.

HUNTER MG. 1998. Watershed-level patterns among stream amphibians in the Blue River watershed, West-Central Cascades of Oregon [thesis]. Corvallis, OR: Oregon State University. 97 p.

KARRAKER NA. 1999. Rhyacotriton variegatus (Southern Torrent Salamander): Nest site. Herpetological Review 30:160-161.

KARRAKER NA, OLLIVIER LM, HODGSON GR. 2005. Oviposition site of the Southern Torrent Salamander (Rhyacotriton variegatus) in Northwestern California. Northwestern Naturalist 86:34-36.

KROLL AJ, RISENHOOVER K, MCBRIDE T, BEACH E, KERNOHAN BJ, LIGHT J, BACH J. 2008. Factors influencing stream occupancy and detection probability parameters of stream-associated amphibians in commercial forests of Oregon and Washington, USA. Forest Ecology and Management 255:3726-3735.

MACCRACKEN JG. 2004. Rhyacotriton cascade (Cascade Torrent Salamander): Nest. Herpetological Review 35:367.

MONTGOMERY DR, BUFFINGTON JM. 1998. Channel processes, classification, and response. In: NAIMAN RJ, BILBY RE, editors. River ecology and management: Lessons from the Pacific Coastal Ecoregion. New York, NY: Springer-Verlag, p 13-42.

NUSSBAUM RA. 1969. A nest site of the Olympic Salamander Rhyacotriton olympicus (Gaige). Herpetologica 25:277-278.

NUSSBAUM RA, TAIT CK. 1977. Aspects of the life history and ecology of the Olympic Salamander, Rhyacotriton olympicus (Gaige). American Midland Naturalist 98:176-199.

NUSSBAUM RA, BRODIE ED JR, STORM RA. 1983. Amphibians and reptiles of the Pacific Northwest. Moscow, ID: University of Idaho Press. 332 p.

PYRON RA, WIENS JJ. 2011. A large-scale phylogeny of Amphibia including over 2800 species, and a revised classification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution 61:543-583.

RUSSELL KR, GONYAW AA, STROM JD, DIEMER KE, MURK KC. 2002. Three new nests of the Columbia Torrent Salamander, Rhyacotriton kezeri, in Oregon with observations of nesting behavior. Northwest Naturalist 83:19-22.

STEBBINS RC. 1951. Amphibians of western North America. Berkeley, CA: University of California Press. 539 p.

STEBBINS RC, LOWE CH. 1951. Subspecific differentiation in the Olympic salamander Rhyacotriton olympicus. University of California Publications in Zoology 50:465-484.

STRAHLER AN. 1952. Hypsometric (area-altitude) analysis of erosional topography. Geological Society of America Bulletin 63:1117-1142.

TAIT CK, DILLER LV. 2006. Life history of the Southern Torrent Salamander (Rhyacotriton variegatus) in coastal northern California. Journal of Herpetology 40:43-54.

THOMPSON CE, WALKER JM, WATERSTRAT FT, MCINTYRE AP, HAYES MP. 2011. Rhyacotriton kezeri (Columbia Torrent Salamander) oviposition site. Herpetological Review 42:406-408.

WILKINS RN, PETERSON NP. 2000. Factors related to amphibian occurrence and abundance in headwater streams draining second-growth Douglas-fir forests in southwestern Washington. Forest Ecology and Management 139:79-91.

CURTIS E THOMPSON (1), CHARLES E Foxx (2), REED OJALA-BARBOUR, AIMEE P MCINTYRE, AND MARC P HAYES

Washington State Department of Fish and Wildlife, Habitat Science, 1111 Washington St. SE, Olympia, WA 98501 USA; curtistliomfison2002@yahoo.com

(1) Current address: 16818 Sherman Street, La Conner, WA 98257 USA.

(2) Current address: Skokomish Department of Natural Resources, North 541 Tribal Center Road, Skokomish Nation, WA 98584 USA.

Submitted 30 October 2017, accepted 5 July 2018. Corresponding Editor: Chelsea Waddell.
TABLE 1. Developmental markers of a captive-reared Olympic Torrent
Salamander (Rhyacotriton olympicus) embryo observed over the interval
5 October to 2 December 2016.

Observation
date         Eyes            Heart        Yolk and body trunk

 5 October   Dark spots
11 October   Fully apparent  Beating; 80  Large lateral blood vessels
                             beats/sec    over yolk; mid-yolk
                                          groove visible
14 October                                Increase in blood vessels
                                          over yolk
 1 November
12 November                               Costal grooves appear:
                                          mid-yolk groove
                                          obscured
27 November
 2 December

Observation
date         Gills                   Appendages

 5 October
11 October   Gill buds               Small anterior and posterior limb
                                     buds
14 October
 1 November  Gills with circulating  Limbs buds more developed
             blood
12 November
27 November                          Digits of anterior limbs visible
 2 December                          Digits of posterior limbs visible

Observation
date         Skin pigmentation

 5 October
11 October   Melanophores appear
14 October   Melanophore density increased
 1 November
12 November  Dorsolateral lines of spot pattern become evident
27 November
 2 December  Dorsolateral spot pattern striking

TABLE 2. Observation dates, localities, egg characteristics, and
water temperatures at oviposition sites among species of torrent
salamander (genus Rhyacotriton).

Torrent                                Locality          Number
salamander species  Observation date   (State: County)   of eggs (n)

R. cascadae         14 August 2003     WA: Skamania       5
R. kezeri           14 December 1968   OR: Tillamook     32 (a)
                    28 September 1969  WA: Wahkiakum     75 (a)
                    16 July 2001       OR: Clatsop       10
                    26 July 2001       OR: Clatsop        7
                    26 July 2001       OR: Clatsop       11
                    26 July 2010       WA: Grays Harbor  10
R. olympicus        10 August 2016     WA: Jefferson     10 (b)
R. variegatus       28 September 1995  CA: Humboldt      11
                     4 October 1996    CA: Humboldt       8

Torrent             Capsular           Ovum
salamander species  diameter (mm)      diameter (mm)

R. cascadae         -                  ~3-4
R. kezeri           -                   -
                    -                   -
                    -                   3.8 [+ or -] 0.4
                    -                   4.1 [+ or -] 0.4
                    -                   3.9 [+ or -] 0.4
                    11                  4
R. olympicus         9.5 [+ or -] 0.9   4.5 [+ or -] 0.5
R. variegatus       -                   5.6 [+ or -] 0.4
                    -                  ~5

Torrent             Water
salamander species  temperature ([degrees]C)

R. cascadae         12.0
R. kezeri            8.3
                     9.1
                     8.9
                    11.1
                     9.7
                    12.5
R. olympicus         9.9 [+ or -] 0.1
R. variegatus        -
                    10.4

Torrent
salamander species  Source

R. cascadae         MacCracken (2004)
R. kezeri           Nussbaum (1969)
                    Nussbaum (1969)
                    Russell and others (2002)
                    Russell and others (2002)
                    Russell and others (2002)
                    Thompson and others (2011)
R. olympicus        This study
R. variegatus       Karraker (1999)
                    Karraker and others (2005)

(a) Likely multiple clutches. (b) 1 egg had 2 embryos so the clutch
size was actually 11.

TABLE 3. Habitat characteristics of oviposition sites among species
of torrent salamander (genus Rhyacotriton). Order of species
presentation matches Table 2.

               Locality          Elevation  Aquatic     Stream
Species        (State: County)   (m)        unit        order (a)

R. cascadae    WA: Skamania      740        stream (c)  2nd
R. kezen       OR: Tillamook     ~85        spring      1st
               WA: Wahkiakum     ~60        seep (f)    1st
               OR: Clatsop       439        spring      1st
               OR: Clatsop       195        stream      1st
               OR: Clatsop       256        spring      1st
               WA: Grays Harbor  157        stream      1st
R. olympicus   WA: Jefferson     189        seep        1st
R. variegatus  CA: Humboldt      235        stream      1st
               CA: Humboldt      158        stream      2nd

                                                 Canopy
Species        Notes                     Aspect  cover      Forest type

R. cascadae    473 m lorigin (d)         S       1007%      2nd-growth
R. kezen       at origin                 W          -       2nd-growth
               at origin? (g)            S          -       2nd-growth
               at origin                 NW        96%      2nd-growth
                75 m [down arrow]origin  N         99%      2nd-growth
               at origin                 E         94%      2nd-growth
                49 m [down arrow]origin  SE         0% (l)  2nd-growth
R. olympicus    80 m [down arrow]origin  NW        91%      2nd-growth
R. variegatus  in riffle                 NW        97%      2nd-growth
               step-run                  sw         -       old-growth

               Dominant
Species        trees (b)  Substrate

R. cascadae    DF, NF     Under cobble
R. kezen       -          Sandstone cracks (c)
               DF         Sandstone cracks (h)
               SS, WH     Sand and fines (i)
               WH, SS     Varied (j)
               SS, WH     Varied (k)
               WH         Varied (m)
R. olympicus   WH, DF     Gravel (n)
R. variegatus  CR, DF     Under boulder (o)
               CR, DF     Under boulder (p)

(a) Based on Strahler (1952). This is the order of the stream
associated with the aquatic unit if it is not the aquatic unit per
se. (b) CR = Coast Redwood (Sequoia sempervirens), DF = Douglas-fir
(Pseudotsuga menziesii), NF = Noble Fir (Abies nobilis), SS = Sitka
Spruce (Picea sitchenis), WH = Western Hemlock (Tsuga heterophylla).
(c) In thalweg of glide. (d) [down arrow] origin means below the
upstream origin of the stream. (e) Clast size of sandstone material
not indicated, however, discovery was made by "digging into the
mouth of spring and following the seepage current", so the clast
size was probably not large. (f) Described as a trickle. (g) Of
side-slope seep. (h) In a sandstone cliff beneath a large (2.4 X
1.2 X 1.2 m) boulder pried away from the cliff. (f) Beneath a small
(0.3 X 0.2 X 0.1 m) boulder in the thalweg of a small pool. (j) Under
a 17-cm deep layer of moss-covered coarse and fine gravel, sand, and
silt overlain with fine organic sediments. (k) Covered by a 15-cm
deep mixture of fine gravel and organic silt overlain by a thick
layer of moss and a decaying log. (l) Though canopy cover was 0%.
harvest slash dominated by wood 2-10 cm in diameter completely
covered the oviposition site. (m) Composite of gravel, silt and
small decayed wood pieces. (n) Intermixed with some sand and cobble.
(o) On large cobble. (p) 42 cm in diameter (longest dimension) over
loose gravel.
COPYRIGHT 2018 Society for Northwestern Vertebrate Biology
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Thompson, Curtis E; Foxx, Charles E; Ojala-Barbour, Reed; Mcintyre, Aimee P; Hayes, Marc P
Publication:Northwestern Naturalist: A Journal of Vertebrate Biology
Article Type:Report
Geographic Code:1U9WA
Date:Dec 22, 2018
Words:5418
Previous Article:REVISITING EVIDENCE FOR HISTORIC PRESENCE OF NORTHWESTERN DEERMOUSE (PEROMYSCUS KEENI) ON LANGARA ISLAND, HAIDA GWAII, BRITISH COLUMBIA.
Next Article:EFFECTS OF A SEVERE DROUGHT ON SUMMER ABUNDANCE, GROWTH, AND MOVEMENT OF CUTTHROAT TROUT IN A WESTERN OREGON HEADWATER STREAM.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters