Arch. Pol. Fish. (2014) 22: 159-165 DOI 10.2478/aopf-2014-0016 RESEARCH ARTICLE Spawning behaviour and the softmouth trout dilemma Manu Esteve, Deborah Ann McLennan, John Andrew Zablocki, Gašper Pustovrh, Ignacio Doadrio Received – 05 November 2013/Accepted – 26 February 2014. Published online: 30 June 2014; ©Inland Fisheries Institute in Olsztyn, Poland Citation: Esteve M., McLennan D.A., Zablocki J.A., Pustovrh G., Doadrio I. 2014 – Spawning behaviour and the softmouth trout dilemma – Arch. Pol. Fish. 22: 159-165. Abstract. Morphological, ecological and molecular data sets nest digging behaviour-widespread in all the salmonines, do not completely agree on the phylogenetic placement of the including softmouths, they seem to be mal-adaptive. softmouth trout, Salmo (Salmothymus) obtusirostris (Heckel). Molecules posit that softmouths are closely related to brown Keywords: phylogeny, spawning behavior, underwater trout, Salmo trutta L. while some morphological, ecological video and life history traits place them in the most basal position of the Salmoninae subfamily between grayling (Thymallus) and lenok (Brachymystax). Here we add an additional source of data, behavioural characters based on the first reported Introduction observations of softmouth spawning. During spawning softmouth females present three important behaviours not The softmouth trout, also known as the Adriatic found in the other Salmo members: they continually abandon trout, Salmo (Salmothymus) obtusirostris (Heckel), is their nests, rarely staying on them for periods over nine a cold freshwater salmonid found naturally in only minutes; they expel different batches of eggs at the same nest five river drainages of the Adriatic Sea: the Vrljika, at intervals of several minutes; and they do not cover their eggs immediately after spawning. These three behaviours are Jadro and Krka in Croatia, the Neretva in Bosnia and intriguing for two reasons: 1) they are possible homologous to Herzegovina and Croatia and the Zeta in Montenegro behaviours found in grayling females; 2) when coupled to the (Snoj et al. 2008). Based mainly on colour variations, it is sometimes divided into four subspecies: Salmothymus obtusirostris oxyrhynchus + M. Esteve [ ], I. Doadrio (Steindachner 1882) from the rivers Neretva and Departamento de Biodiversidad y Biología Evolutiva Museo Nacional de Ciencias Naturales Vrljika (this study), S. o. salonitana (Karaman 1927) Jose Gutierrez Abascal 2, Madrid, Spain from the River Jadro, S. o. krkensis (Karaman 1927) e-mail: [email protected] from the River Krka and S. o. zetensis (Hadišèe D.A. McLennan 1961) from the River Zeta. Morphologically, the Department of Ecology and Evolutionary Biology, University of softmouth is deep bodied with relatively large scales. Toronto, 25 Willcocks St, Toronto, M5S 3G5 ON, Canada Its colouration includes various greens, pale yellows, J.A. Zablocki and light browns with lateral round black and red Trout Unlimited, 720 Tahoe St STE 1, Reno NV 89509, USA spots with white outlines. Its most obvious morpho- G. Pustovrh logical characteristic is an elongated snout and University of Ljubljana, Biotechnical Faculty, Department of Animal a small, fleshy subinferior mouth with small jaws and Science, Groblje 3, 1230 Domale, Slovenia 160 Manu Esteve et al. teeth. This distinct head morphology, which resem- us about the evolutionary history of softmouth trout? bles the morphology of archaic salmonids such as We began by collecting information from video re- grayling (Thymallus) and lenok Brachymystax lenok cordings of softmouth trout during spawning, which (Pallas), is responsible for its historically unstable has never been reported before. We then discuss the classification. From its initial description as Salar results comparing softmouth trout spawning behav- obtusirostris by Heckel (1851), it was moved to iour with the other Salmo members, in particular, Thymallus (Steindachner 1874) and Salmo and with the other salmonines, in general, to see (Steindachner 1882), before being placed in its own whether this information could provide clues to re- genus, Salmothymus (Berg 1908). Behnke (1968) solving this dilemma. later redefined the taxon as a subgenus of Salmo based on morphological characters, however a gen- eral consensus based on morphology has never been Materials and methods reached. For instance, Sanford (2000) based on an exhaustive osteological analysis of juvenile fish, un- equivocally placed softmouth in the genus Salmo. Study Area While other authors, also basing their studies on osteological characters, placed it within the The Neretva is a 230 km long karst river in the Salmothymus genus in the most basal position of the Dinaric Alps flowing through Bosnia and salmonine tree (Stearley and Smith 1993, Wilson Herzegovina and joining the Adriatic Sea in Croatia. and Li 1999). A molecular-based study by Snoj et al. Underwater recordings took place in the upper (2002); see also Sušnik et al. (2006), placed the Neretva during March 11, 2009 in a shallow gravel, softmouth with the Ohrid trout, Salmo ohridanus area located approximately 2 km upstream from (Steindachner) as the sister group to brown trout, Glavatièevo village (43.2952 N, 18.0843 E, 381 m Salmo trutta L., with a recommendation to move it altitude). Recordings were made with a Sony into Salmo. HDR-HC7 video camera inside an underwater hous- Several ecological and life history traits are not ing. The camera was placed approximately 0.7 me- congruent with the new phylogenetic position of this ters from an active nest identified as an elliptical area species. For instance, softmouths spawn in late win- with clean gravel containing at least one fish. Neretva ter-early spring, which can be viewed as intermediate softmouth trout presented no marked sexual dimor- between traditional fall (Salmo) and spring (graylings phism, so behaviour was used to infer sex: the fish and lenoks) spawners. Also like graylings they are that intermittently performing quivering (courtship benthic feeders and have a similar behavioural ecol- behaviour consisting of low amplitude and high fre- ogy usually spending their time in the middle of river quency body vibrations from head to tail) was identi- and always swimming in small groups of at least two fied as the male, while the female was the fish that or three individuals (Georgiev 2003, Mrdak et al. regularly performed nest digging (female turns on 2012). Softmouths never rest on the ground between her side and excavates a depression in the gravel by stones or plants as do brown trout, hiding only rarely, beats of her tail). and then only in large cavities (J. Schoeffmann – per- The Vrljika River originates northwest of the sonal communication). We have named this dis- town Imotski in Croatia. After crossing the Cro- agreement between molecules on one side and some atian-Bosnian, the name of the river is Matica and it morphology, ecology, and life history traits, on the goes underground near the town of Drinovci. It co- other the “softmouth trout dilemma”. Here we pres- mes out as the Tihaljina River, which is in its lower ent data from yet another source, behaviour, to fur- part known as the Trebiat River, the right tributary ther investigate this dilemma. In this manuscript to the Neretva River. Observations in the Vrljika were therefore, we ask what can spawning behaviour tell made during February 13-15, 2011 and February Spawning behaviour and the softmouth trout dilemma 161 7-10, 2013 in Proloac near Imotski (43.2712 N, 17. 1025 E, 270 m altitude). The 2011 recordings were taken from the riverbank using a tripod and a Sony HDR-HC7 video camera. The 2013 recordings were done underwater using Gopro cameras (www.gopro.com) mounted on stationary underwa- ter tripods. As in Neretva, the Vrlijika females did not show marked sexual dimorphism and behaviour was used to infer the sex. Figure 1. Female softmouth trout probing substrate with her anal fin. Results episodes alternating with probing (the female lies Three distinct behaviours respect to the other over her nest and presses her anal fin into the sub- Salmo members studied were identified as follows: strate to test its suitability; Fig. 1). Nest residence 1. Females make short and continuous visits to time was not quantified because of interruptions their nests (wandering). caused by camera battery failure. 2. Females perform several spawning acts in the Vrlijika 2012 same nest separated by only a few minutes. One female was recorded in the same redd being 3. Females do not cover their eggs immediately courted and guarded by the same male during three after spawning. consecutive days, fot a total of 15 hours and 30 min- Details on each of the rivers are as follows: utes of spawning activity (Table 1). The results con- Neretva 2011 firmed the pattern of female periodic nest visits One female was recorded continuously aban- recorded in the Neretva. Recordings from February doning her nest. She stayed for short periods of time 13 included twelve spawning acts and are presented during which she performed a total of 32 digging graphically in Fig. 2. Figure 2. Ethogram of female softmouth trout activity at the Vrljika during February 13, 2011. 162 Manu Esteve et al. Table 1 Recording history of one softmouth tout female in the 2011 Vrlijika River spawning season (time is expressed in minutes) Rt nt nv a nt nd n sp Feb 12 258 22 7 3.14 3 0 Feb 13 587 125 54 2.33 32 12 Feb 1 85 30.5 6 5.07 14 0 Total 930 177.5 67 2.65 49 12 Rt – (recording time): minutes recorded, nt – (nesting time): minutes female spent at the nest, nv – (number of visits): number of nest visits by the female, a nt – (average nesting time): average time female spent at nest, nd – (number of digs): number of nest building digging episodes, n sp – (number of spawnings): number of spawning acts performed by the female Vrlijika 2013 behaviours found in grayling females, and when cou- Three females were recorded digging their nests pled with the nest digging behaviour, they seem to be while accompanied by a variable number of males.