Climacia Californica Chandler, 1953 (Neuroptera: Sisyridae) in Utah: Taxonomic Identity, Host Association and Seasonal Occurrence

Home , Climacia, Ephydatia, Ephydatia fluviatilis, Spongillidae

AQUATIC INSECTS 2019, VOL. 40, NO. 4, 317–327 https://doi.org/10.1080/01650424.2019.1652329

Climacia californica Chandler, 1953 (Neuroptera: Sisyridae) in Utah: taxonomic identity, host association and seasonal occurrence

Makani L. Fishera, Robert C. Mowerb and C. Riley Nelsona aDepartment of Biology and M. L. Bean Life Science Museum, Brigham Young University, Provo, UT, USA; bUtah County Mosquito Abatement, Spanish Fork, UT, USA

ABSTRACT ARTICLE HISTORY We provide a record of spongillaflies (Sisyridae) with an associated Received 30 December 2018 host sponge from a population found in Spring Creek, Utah. We Accepted 15 May 2019 monitored the population for the 2016 field season to identify the First published online insect and its associated host sponge and to establish the sea- 30 September 2019 ’ sonal period of the insect s presence in the adult stage. We also KEYWORDS evaluated the commonly used sampling techniques of sweeping Aquatic Neuroptera; and light trapping and made natural history observations. We Sisyridae; freshwater identified the spongillaflies as Climacia californica Chandler, 1953 sponge; Utah; and the sponge to be Ephydatia fluviatilis (Linnaeus, 1758). We col- Intermountain West lected 1726 adults, and light trapping proved to be the superior collecting method. The population was characterised by a two- week emergence peak occurring at the end of July to beginning of August followed by a steep decline. Other habitats within the state contained E. fluviatilis and should be sampled using light traps at times when peak abundances are predicted to further understand the distribution of C. californica across the state and the Intermountain West.

urn:lsid:zoobank.org:pub:F795281F-0042-4F5F-B15B-F6C584FACF2B

Introduction Sisyridae, or spongillaflies, and freshwater sponges (Porifera: Spongillidae) are linked together in a parasite–host association (Parfin and Gurney 1956; Steffan 1967; Resh 1976; Pupedis 1980). Sisyrids feed on freshwater sponge with stylets used to pierce and suck the contents from individual sponge cells (Cover and Resh 2007; Parfin and Gurney 1956; Pupedis 1980; Figure 1a). Freshwater sponges are found globally and have a wide range of morphological forms (Manconi and Pronzato 2008). Spongillaflies, having adapted to many of the same habitats as their hosts, are also found worldwide (Cover and Resh 2007; Parfin and Gurney 1956). Within North

CONTACT Makani Fisher [email protected] Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/naqi. ß 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.

Published online 30 Sep 2019 318 M. FISHER ET AL.

Figure 1. Live Climacia californica Chandler, 1953 from Spring Creek, Springville, Utah: (A) larva on host sponge; (B) adult on leaf.

America, there are two genera and six known species (Bowles 2006). Despite their occurrence in parts of western North America (Bowles 2006), sisyrids have only been recorded twice in the Intermountain West. First, a host association was made for them with the sponge Ephydatia mulleri (Lieberkuhn,€ 1856) in Idaho (Clark 1985). Second, a Utah record was made from a single female specimen collected at a light trap in 1979 from the state’s southern border (Bowles 2006). After this initial Utah record, sisyrids have been sought extensively throughout the state by both professionals and students, largely from Brigham Young University (BYU) and Utah State University (USU). These efforts were directed towards habitats with potential freshwater sponge hosts. Sponges have rarely been noted in Utah, but they do exist in some regions where they provide potential habitats for the insect. The junior author anecdotally observed them at Wellsville Reservoir in extreme northern Utah in the 1980s as well as in several side channels of Lake Powell beginning in 2012. Rader also identified Ephydatia fluviatilis (Linnaeus, 1758) in Utah AQUATIC INSECTS 319

Lake (Rader 1984; Rader and Winget 1985). Other researchers including Dennis Shiozawa, Russell Rader, Richard Baumann and various BYU aquatic entomology classes (personal communication, January 20, 2017) repeatedly searched many of these and other likely habitats throughout the state over the course of 35 years. None of those efforts yielded spongillaflies, but the second author (Utah County Mosquito Abatement, UCMA), observed sisyrid adults in mosquito light traps from Spring Creek, Springville, Utah in 2014 (Figure 1b). After repeatedly seeing sisryids in those traps, he alerted the junior author who began to survey the creek. In fall of 2014, BYU’s aquatic entomology class collected larvae and adult sisyrids from this creek in rather large numbers (Figure 1). After BYU students regularly collected specimens from this site in 2014 and 2015, we monitored the population for the 2016 field season. Records and dates have been published for North American sisyrids by Bowles (2006), but no populations have been consistently followed through a season. These collective efforts produced a new distributional record of spongillaflies for Utah and their associated sponge host. This record further expands their known distribution in the Intermountain West. We also compared the two commonly used sampling techniques of sweeping and light trapping (Bowles 2008) and observed the population’s seasonality. In particular, we inves- tigated the effectiveness of the sampling techniques, length of the adult flight season and the frequency and magnitude of emergence events. Compared to sweeping, we hypothesised light trapping to be superior as spongillaflies are strongly attracted to a variety of lights (Bowles 2008). In regard to their natural history, we expected the sisyrids to be active from May to November (Bowles 2006) and to be multivoltine (DeWalt, Resh, and Hilsenhoff 2010).

Material and methods We monitored sisyrids in Springville, Utah at the Utah Division of Wildlife Resources (UDWR) Fish Hatchery. The study site is located at 401005600N, 1113604300W at an elevation of 1395 m. Spring Creek flows through the site providing habitat for the sponges and sisyrids we examined. We identified suitable freshwater sponge habitat within a 100 m stretch of the creek and sampled it intensively. We found many larvae present on the sponge (Figure 1a) and collected both larvae and sponge for taxonomic purposes. Collected sisyrid larvae were stored in 70% ethanol. We attempted to identify the larvae using keys as detailed in the work by Pupedis (1980) and Bowles (2006). We sampled four of these sponges for identification by scraping their surfaces using soft tip forceps and placing the scrapings in 50 ml vials filled with 70% ethanol. We extracted their spicules using several techniques to aid in identification of the species. We first used the technique described by Smith and Pennak (2001) for extracting spicules using heated HNO3. We used a section of sponge tissue containing gemmules to ensure finding all types of spicules. This yielded both the megascleres and gemmoscleres (Figure 2) needed for proper identification. We also left sponge tissue in 10% vol potassium hydroxide overnight and found this to yield a good number of megascleres, but the gemmules did not reduce to gemmoscleres. We then 320 M. FISHER ET AL.

Figure 2. Ephydatia fluviatilis (Linnaeus, 1758) spicules extracted and cleared from host sponge collected at Spring Creek, Springville, Utah: (A) megascleres and gemmoscleres; (B) isolated gemmosclere.

explored treating the gemmules in cold HNO3 for several days, and many individual gemmoscleres were obtained using this approach. We mounted cleared megascleres, gemmoscleres and sponge tissue onto glass slides for species identification (Ricciardi and Reiswig 1993). We repeated this process for sponge we collected earlier the same year from the Provo River in Provo, Utah. This sponge did not have any sisyrid larvae, but its spicules were similar to those from Spring Creek. We compared both these sponge identifications to sponge previously identified from Utah Lake (Rader 1984; Rader and Winget 1985). This was done to identify other nearby local habitats with host sponge that may be suitable for the spongillaflies. We monitored the adult sisyrid population along Spring Creek and identified them to species using Bowles (2006) guide. We used wing veins and male genitalia for definitive identification (Figure 3) as female characteristics tend to be more variable (David Bowles, personal communication, January 21, 2016). The population was monitored by simultaneously using quantitative sweep sampling and light trapping. AQUATIC INSECTS 321

Figure 3. Taxonomically relevant parts of adult Climacia californica Chandler, 1953 from Spring Creek, Springville, Utah: (A) forewing; (B) cleared male genitalia with visible gonarcus complex, lat- eral view; (C) cleared male genitalia with visible gonarcus complex, dorsal view.

The sampling techniques were done together to directly compare the two. If a specimen was collected by one technique, it could not be collected by the other i.e., if specimens were effectively attracted to light, they would be taken up by the trap and sweep counts would be low and vice versa. For sweep samples we used a 45-cm collapsible net to sweep the vegetation along the north side of the creek each week a light trap sample was taken. We started at the edge of unmown vegetation 3 m east of the trap to 97 m beyond the trap, for a total of a 100 m sweep. We removed all sisyrid adults (Figure 1b) from the net and placed them into 70% ethanol. This process was repeated back to the starting point resulting in vegetation being swept twice for a total sweep of 200 m. We checked our net for sisyrids every 15 sweeps. The number of sweeps we did varied slightly from week to week due to available vegetation, but averaged 135 sweeps each way, 270 sweeps total. We used a New Jersey Light Trap (Mulhern 1942) that was provided by UCMA to also collect specimens. We deployed the trap continuously using a photoswitch to turn the light on each night and off each morning. We recovered the trap and sweep samples each Friday for the duration of the flight season and a few weeks beyond. We conducted a preliminary early season sweep around the trap site that indicated the adult flight season began near 20 May in 2016. We subsequently began our quantitative trapping and sweeping on 27 May 2016. We monitored the population 322 M. FISHER ET AL.

Table 1. Male and female counts of Climacia californica Chandler, 1953 from trap and sweep samples from 27 May 2016 through 18 November 2016 at Spring Creek, Utah. Date Trap Trap Trap Sweep Sweep Sweep Overall in 2016 males females total males females total total May 27 0 0 0 1 2 3 3 Jun 3 1 3 4 1 1 2 6 Jun 10 12 9 21 7 8 15 36 Jun 17 9 17 26 4 5 9 35 Jun 24 19 31 50 1 7 8 58 Jul 1 6 10 16 1 2 3 19 Jul 8 11 32 43 3 4 7 50 Jul 15 27 35 62 6 5 11 73 Jul 22 60 93 153 6 6 12 165 Jul 29 288 216 504 19 24 43 547 Aug 5 180 208 388 17 15 32 420 Aug 12 31 34 65 14 21 35 100 Aug 19 26 23 49 8 9 27 66 Aug 29 11 21 32 1 7 8 40 Sep 2 ––––– –– Sep 9 ––––– –– Sep 16 ––––– –– Sep 23 ––––– –– Sep 30 41 34 75 7 4 11 86 Oct 7 2 3 5 2 1 3 8 Oct 14 0 0 0 1 2 3 3 Oct 21 1 0 1 1 1 2 3 Oct 28 0 0 0 5 2 7 7 Nov 4 0 0 0 0 0 0 0 Nov 11 0 0 0 1 0 1 1 Nov 18 0 0 0 0 0 0 0 Totals 725 769 1494 106 126 232 1726

intensively, recording the number of sisyrids caught in each sweep and trap sample, until we consistently obtained low counts of essentially zero. We sampled the population 22 times as four light trap samples were also lost when they were not kept after mosquito counts were obtained, which produced a gap in our data. However, trends in numbers before and after those lost samples indicate that an assumption of uni- form change during that period is reasonable because of the large number of weekly samples we collected throughout the season. This large number of samples also provided enough consecutive data to conduct statistical analysis of sampling technique effectiveness and the adult population over time. The sex ratio of sisyrids has not been studied well (Andersen and Greve 1975). We sorted specimens by sex to investigate effectiveness of the sampling methods as well as to estimate a sex ratio (Table 1). There were a few specimens whose genitalia were broken off and sex could not be determined. Those specimens were not recorded in counts or used in statistical analysis. For analysis, we used a generalised linear model with a log link function with a normal distribution of the response variable (Proc GENMOD in SAS; SAS 9.3 SAS Institute, Cary, North Carolina, USA). We analysed the main effects of gear (sweep and trap), time and sex as well as their two-way and three-way interactions with an alpha level of 0.05. This was done using two levels for gear (sweep vs. trap) and sex (male vs. female) and the 22 sample times throughout the season (Table 2). The null hypothesis was that no sex-based preferences existed for either sampling technique and that both were equally effective and not effected by time. AQUATIC INSECTS 323

Table 2. Results from running the Spring Creek, Utah, Climacia californica Chandler, 1953 trap and sweep counts in a generalised linear model. Parameters DF Chi square p value Gear 1 18,098.2 <0.0001 Sex 1 46.55 <0.0001 Time 21 101,256 <0.0001 Gear sex 1 6.55 0.0105 Gear time 21 74,893.3 <0.0001 Sex time 21 1877.96 <0.0001 Gear sex time 21 2141.96 <0.0001 Results from the main effects of gear, sex, time, 2-way interactions, and 3-way interactions. We considered anything below a 0.05 alpha level as significant.

Vouchers of all sponge and spongillaflies identified are deposited at the Monte L. Bean Life Sciences Museum, Brigham Young University, UT, USA.

Results We identified the host sponge from Spring Creek as Ephydatia fluviatilis (Linnaeus, 1758) based on elongate megascleres and birotulate gemmoscleres (Figure 2). We identified the sponge from the Provo River to be the same species. These identifications also match the identification for the sponge found in Utah Lake (Rader 1984; Rader and Winget 1985). We identified the adult sisyrids as Climacia californica Chandler, 1953. The identification was straight forward, and we saw no change in forewing or genital characteristics (Figure 3) over the course of the season, indicating only one species present at the site. Although we used male characteristics for identification, female genitalia and forewing patterns were also consistent throughout the season. The larvae proved to be more difficult to identify. According to Bowles (2006), the larval characteristic of the number of setae, four large instead of three on the meso- and metathoracic sclerite (Figure 4), indicates that our larvae are Climacia chapini Parfin and Gurney, 1956. This characteristic was taken from the previous work on sisyrid larvae by Pupedis (1980). However, according to Pupedis (1980), this setal number identifies our larvae as C. californica bringing up some discrepancy between these two treatments and our specimens. Because of our confident adult identification and its consistency through the season, we consider both adult and larvae to be C. californica. In the 22 samples, we collected 1726 adults of C. californica: 769 trap females, 725 trap males, 126 sweep females and 106 sweep males (Figure 5; Table 1). From the generalised linear model (Table 2), we found that each of the main effects from the variables gear, sex and time were significant, all having two-sided p values < 0.0001. All two- and three-way interactions were also significant with the two-way interaction of gear and sex having a p value of 0.012 and all other interactions having p values < 0.0001. The adult population was higher than 0 for sweeping and trapping from 27 May to 18 November 2016 (Figure 5). With the exception of the beginning and end of the season, the light trap caught more sisyrids than sweeping. Male and female numbers fluctuated throughout the season, but both techniques caught more female specimens overall, reflecting a population with a slight female sex biased ratio. The 324 M. FISHER ET AL.

Figure 4. Meso- and metathoracic sclerites of Climacia californica Chandler, 1953 from larvae collected from Spring Creek, Springville, Utah: (A) whole larvae with placement of sclerites marked with a bracket; (B) close up of sclerites with 4 major setae clearly visible and 1 small minor seta highlighted with a circle and arrow on mesothoracic sclerite. season was also characterised by a two-week peak of abundance at the end of July and beginning of August, and the adult population diminished drastically and eventu- ally disappeared in November (Figure 5; Table 1).

Discussion The general area where the spongillaflies were recently discovered was surveyed in the 1970s and 1980s by Dennis Shiozawa and Russell Rader (personal communication, January 20, 2017). No spongillaflies were found from these surveys, and the reason for this recent finding of spongillaflies in such large numbers at the location is not clear. It could be due to introduced stowaways on fish or equipment into the fish AQUATIC INSECTS 325

Figure 5. Male and female counts of Climacia californica Chandler, 1953 from samples taken from 5 May 2016 to 18 November 2016 at Spring Creek, Springville, Utah: (A) sweep male and female sample numbers; (B) light trap male and female sample numbers. raising facilities of the UDWR Fish Hatchery at the stream, improved sampling or specimen identifications or the habitat becoming more suitable in some way to main- tain larger populations. Whatever the reason, we have now found an established population that provides a large additional record for the rarely caught sisyrids in the state of Utah and Intermountain West (Bowles 2006). Species of Ephydatia have previously been identified as hosts for C. californica within the Intermountain West (Clark 1985). We confirmed the sponge E. fluviatilis as a host for this species and that it is the only freshwater sponge currently known from Utah (Rader 1984; Rader and Winget 1985). The adult’s forewing and male genital characters of C. californica shown in Bowles (2006) provide us confidence that our species identification is correct. The confusion of whether the larvae support C. chapini or C. californica seems to come from a tran- scriptional error in Bowles (2006), which was based on Figure 1f from Pupedis (1980). With this apparent error and our adult identifications, we propose the larvae 326 M. FISHER ET AL. we collected to be C. californica and that larvae of this species have four major setae on the meso- and metathoracic sclerites, not three (Figure 4). The beginning and end of the season where few specimens were caught resulted in similar trap and sweep counts, but throughout the majority of the season the light trap caught more, at times hundreds more, than sweeping (Figure 5; Table 1). Similar to other neuropterans (Killington 1937), we observed no unique male or female preference to light in this female dominated population. We suggest that both techniques are accurate subsamples of the population with no preference biases, though light trapping is preferred, if possible, because of its ease of use and large capture rate. The active adult season from May to November is consistent with records published by Bowles (2006). Furthermore, it gives the whole picture showing C. californica adults being found consistently throughout the season. Also highlighted is a striking two-week peak at the end of July (Figure 5; Table 1). As both sexes increased equally, to peak at the same time, we interpret it to be a well-synchronised mass emergence before the onset of winter. This is likely for mating and laying eggs that hatch and develop into overwintering larvae (DeWalt et al. 2010). With this recently confirmed and thriving population of sisyrids, sampling should be conducted at other areas of Utah with special attention given to areas containing the host sponge to further delineate the distribution of spongillaflies in the state. As these insects are studied by few and are unfamiliar to most (David Bowles, personal communication, January 21, 2016), we also provide taxonomic aid clarifying and demonstrating key features for the identification of C. californica and its host sponge. It is likely that more C. californica will be found as more structured, directed and focused sampling takes place around aquatic habitats where the sponge occurs.

Acknowledgements We thank Utah County Mosquito Abatement for the use of their New Jersey light trap for sampling, for leading us to the Sisyridae population, and help in separating Sisyridae in each week’s sample. We are particularly grateful to David E. Bowles from the U.S. National Park Service and Missouri State University Department of Biology for confirming the identification of C. californica and discussions during the course of this study. We thank Mark Belk from the Department of Biology, Brigham Young University for help running the statistical models as well as with their interpretation.

Disclosure statement No potential conflict of interest was reported by the authors.

References Andersen, T., and Greve, L. (1975), ‘Neuroptera in Light-Traps at Osteroy, Hordaland’, Norwegian Journal of Entomology, 22, 123–128. Bowles, D.E. (2006), ‘Spongillaflies (Neuroptera: Sisyridae) of North America with a Key to the Larvae and Adults’, Zootaxa, 1357, 1–19. Bowles, D.E. (2008), ‘Spongillaflies (Neuroptera: Sisyridae)’,inEncyclopedia of Entomology, ed. J.L. Capinera, Dordrecht: Springer, pp. 3516–3518. AQUATIC INSECTS 327

Chandler, H.P. (1953), ‘A New Species of Climacia from California (Sisyridae, Neuroptera)’, Journal of the Washington Academy of Sciences, 43, 182–184. Clark, W. H. (1985). ‘First record of Climacia californica (Neuroptera: Sisyridae) and its host sponge, Ephydatia mulleri (Porifera: Spongillidae), from Idaho with water quality relation- ships,’ Great Basin Naturalist, 45, 391–394. Cover, M.R., and Resh, V.H. (2007), ‘Global Diversity of Dobsonflies, Fishflies, and Alderflies (Megaloptera; Insecta) and Spongillaflies, Nevrorthids, and Osmylids (Neuroptera; Insecta) in Freshwater’,inFreshwater Animal Diversity Assessment, eds. E.V. Balian, C. Lev^eque, H. Segers, and K. Martens, Dordrecht: Springer, pp. 409–417. DeWalt, R.E., Resh, V.H., and Hilsenhoff, W.L. (2010), ‘Diversity and Classification of Insects and Collembola’,inEcology and Classification of North American Freshwater Invertebrates (3rd ed.), eds. J.H. Thorp and A.P. Covich, Italy: Academic Press, pp. 587–657. Killington, F.J. (1937), A Monograph of the British Neuroptera (Vol. 2), London: Adlard and Son. Linnaeus, C. (1758), Systema Naturae per Regna Tria Naturae, secondum Classes, Ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, Locis (Vol. 2, 10th ed.), Stockholm, Sweden: L. Salvii. Manconi, R., and Pronzato, R. (2008), ‘Global Diversity of Sponges (Porifera: Spongillina) in Freshwater’, Hydrobiologia, 595, 27–33. Mulhern, T.D. (1942), ‘New Jersey Mechanical Trap for Mosquito Surveys’, New Jersey Experiment Station Circular Number, 421, 1–8. Parfin, S.I., and Gurney, A.B. (1956), ‘The Spongilla-Flies, with Special Reference to Those of the Western Hemisphere (Sisyridae, Neuroptera)’, Proceedings of the United States National Museum, 105, 421–529. Pupedis, R.J. (1980), ‘Generic Differences among New World Spongilla-Fly Larvae and a Description of the Female of Climacia Striata (Neuroptera: Sisyridae)’, Psyche: A Journal of Entomology, 87, 305–314. Rader, R.B. (1984), ‘Factors Affecting the Distribution of a Freshwater Sponge’, Freshwater Invertebrate Biology,3,86–97. Rader, R.B., and Winget, R.N. (1985), ‘Seasonal Growth Rate and Population Dynamics of a Freshwater Sponge’, Hydrobiologia, 123, 171–176. Resh, V.H. (1976), ‘Life Cycles of Invertebrate Predators of Freshwater Sponge’,inAspects of Sponge Biology, eds. F.W Harrison and R.R. Cowden, New York: Academic Press, pp. 299–314. Ricciardi, A., and Reiswig, H.M. (1993), ‘Freshwater Sponges (Porifera, Spongillidae) of Eastern Canada: Taxonomy, Distribution, and Ecology’, Canadian Journal of Zoology, 71, 665–682. Smith, D., and Pennak, R.W. (2001), Pennak’s Freshwater Invertebrates of the United States: Porifera to Crustacea (4th ed.), New York: Wiley, pp. 52–58. Steffan, A.W. (1967), ‘Ectosymbiosis in Aquatic Insects’,inSymbiosis (Vol. 2), ed. S.M. Henry, New York: Academic Press, pp. 207–289.