Body size determines the number of scales in cyprinid fishes as inferred from hormonal manipulation of developmental rate By B. A. Levin 1,2 , A. A. Bolotovskiy 1 and M. A. Levina 1 1 Institute of Biology of Inland Waters, Russian Academy of Sciences, Borok, Russia; 2 Severtsov Institute of Ecology and Evolution, Russian Academy of Sciences, Moscow, Russia Summary In experiment with hormonal manipulation of developmental rate (alteration of thyroid status) of two cyprinid species, Abramis brama and Rutilus rutilus, we found that number of scales strongly correlates with body size at the moment of first scale appearance. If squamation development starts earlier and at a shorter body length, it results in reduction of scale numbers. If squamation development starts later and at a larger body length, the number of scales is increased. Calcu- lated distances between scales at the moment of their appear- ance are the same in the different experimental groups despite of acceleration or retardation of developmental rate and a corresponding change in the number of scales (decreasing and increasing). It suggests that size of body at the moment of scale initiation determines the number of scales. Induced hetero- chrony in initiation of squamation is one of the mechanisms, which is evidently responsible for intraspecies variability in scale numbers and might be considered a relevant reason for differences between close-related species in case of evolution- ary fixed heterochronies of scale initiation. Another plausible path for diversity in scale number is a change in distance between forming scales that seems to be more appropriate for species represented phylogenetically more distant lineages. Introduction Scales cover skin of the most of fishes and together with dermal denticles in chondrichthyans, scutes in acipenserids, and bony plates (like in sticklebacks or armored catfishes) constitute important part of dermal post-cranial skeleton (Sire et al., 2009) that protect the body of fishes and serve as calcium pools (Flick et al., 1986). Because scales are countable elements, they are widely involved in morphological descrip- tions of fishes and often used as a marker to distinguish different populations, subspecies or species. Some quantitative characters of squamation such as number of scales in lateral line, around caudal peduncle, in the predorsal region, and from lateral line to dorsal and to anal fins are required data in morphometrics and taxonomic work. Although various aspects of scale development, squamation pattern, histological diversity, epidermal-dermal interactions, origin and evolution of post-cranial dermal skeleton (Park and Lee, 1988; Sire and Arnulf, 1990; Sire et al., 1997, 2009; Sire and Akimenko, 2004; Witten and Huysseune, 2009) have been productively studied, the question of what determines vari- ability in scale number, has not been addressed. Because scales form late in ontogeny (Sire and Akimenko, 2004), some factors influencing development could contribute to the changes in their number. Environmental conditions such as temperature of water, salinity, and industrial pollution can alter number of scales (see brief survey in Levin, 2011) as can parasitism (Hubbs, 1927). Although those observations regis- tered effect of bias of number of scales under impact of certain factors, a mechanism, which is responsible for change in scale numbers was not revealed. Recently it was shown that the most influential factor altering the number of scales is thyroid hormones (Smirnov et al., 2006; Levin, 2010). Thyroid hormones (THs) are well known as regulators of developmental rate in fishes, and they are also essential for morphogenesis, metamorphosis, and regeneration (Leatherland, 1982; Brown, 1997; de Jesus et al., 1998; Power et al., 2001; Liu and Chan, 2002; Okada et al., 2005; Blanton and Specker, 2007). The drastic shift in number of scales in the common roach Rutilus rutilus was found as a result of heterochrony in timing of squamation under hor- monal (thyroid) manipulation (Levin, 2010). Experimentally induced heterochronies were associated with differences in body size at which squamation began. In addition some data suggesting correlation between number of scales and average body size of fishes in spawning stock (pleomerism) have been reported for coregonid fishes (Sva¨rdson, 1952). The present paper is an attempt to understand the contribution of body size to the determination of scale number in fish. For this purpose we use experimental data obtained for two cyprinid species via hormonal manipulation of developmental rate, an approach more often used in aquaculture (Inui and Miwa, 1985; Hirata et al., 1989; de Jesus et al., 1998 among others). Materials and methods Two cyprinid species, common roach Rutilus rutilus and bream Abramis brama were studied. For the former species we used published data (Levin, 2010). Eggs and sperm from running ripe bream were obtained from fish caught in the Rybinskoe Reservoir (Volga River) close to Borok, Yaroslavl Province, Russia. Stripped eggs were immediately fertilized (ÔdryÕ method in Petri dishes). Fertilized eggs were placed on glass plates and incubated in 5 L bowls using water from the Rybinskoe Reservoir. The progeny were reared in different treatment regimes beginning at 1 dpf: (i) in 1 ng ml )1 triiodothyronine, T 3 alkaline solution, (TH group); (ii) in 0.015% thiourea solution, a goitrogen that blocks the synthesizing activity of the thyroid gland and leads to decreased endogenous TH-level during 75 dpf, followed by a treatment period in clean (uncontaminated) water (THIO group); (iii) in clean (uncontaminated) water, control. All experimental test and control fish were reared under similar J. Appl. Ichthyol. 28 (2012), 393–397 Ó 2012 Blackwell Verlag, Berlin ISSN 0175–8659 Received: September 12, 2011 Accepted: March 07, 2012 doi: 10.1111/j.1439-0426.2012.02000.x U.S. Copyright Clearance Centre Code Statement: 0175–8659/2012/2803–0393$15.00/0 Applied Ichthyology Journal of