ISSN 20790864, Biology Bulletin Reviews, 2015, Vol. 5, No. 5, pp. 405–414. © Pleiades Publishing, Ltd., 2015. Original Russian Text © M.V. Vinarski, 2015, published in Zhurnal Obshchei Biologii, 2015, Vol. 76, No. 2, pp. 99–110.

The Fate of Subspecies Category in Zoological Systematics. 2. The Present M. V. Vinarski Omsk State Pedagogical University, nab. Tukhachevskogo 14, Omsk, 644099 Russia Dostoevsky Omsk State University, ul. Adrianova 28, Omsk, 644077 Russia email: [email protected] Received November 12, 2013

Abstract—The current approach to the use of the subspecies category in zoological systematics is an integra tive one. This approach necessarily requires genetic confirmation of validity of the subspecies defined by mor phological data. This makes it possible to separate the subspecies that really exist as separate monophyletic population groups, thereby clearing the system of many “phantom” taxa established in the course of noncrit ical use of the subspecies concept. However, a detailed analysis of intraspecific variation by molecular taxon omy methods in many cases reveals rather a complicated divergence pattern that cannot be adequately described in terms of the classic scheme of species and its subspecies. In light of the irregularity of intraspecific divergence rate with the involvement of molecular and morphological characters, it is proposed to use an extended system of subspecies taxa in the description of extra complicated situations. In addition to the sub species, categories such as allospecies, morphotype, and morphospecies may be used; operational definitions are suggested for these categories. The systematics of the Palearctic great pond snails (the complex Lymnaea stagnalis s. lato) is used as a case study. The provisional system of this group developed by the author is based on morphological and phylogeographical data. The series of subspecies categories of different levels make it possible to maximally completely reflect the intraspecific variation of the great pond snails and, to some extent, the process of their genetic divergence and establishment of their distribution range. DOI: 10.1134/S2079086415050072

The second part of the paper deals with modern approaches to the usage of the subspecies category in zoological systematics, as well as problems of the so called microsystematics, i.e., the systematics operat ing at the lowest level of categories, such as subspecies and infrasubspecies ranks (Mayr, 1982). MODERN THEORY OF SUBSPECIES AND ITS APPLICATIONS Laying aside the phylogenetic species concept (PSC) and all of the projects of rankfree taxonomy, which omits subspecies, we can say that the subspecies concept as it is represented in the current literature has not considerably changed since the classic views of the evolutionary synthesis epoch. Mayr, who defined sub species in the 1940s as “a geographically defined aggregate of local populations which differ genetically and taxonomically from other subdivisions of the spe cies” (1942), repeated the very same definition 50 years later (O’Brien and Mayr, 1991). From the current point of view, the key word in the definition of 1942 is the word genetically. Taking into account the bitter experience of the past, the current supporters of biological species concept place emphasis on that the fact the “good” subspecies should display not only and not so much phenotypic uniqueness as the genetically

fixed distinctions in several independent characters, phylogenetic isolation (monophyly), and certain eco logical adaptations (Shvarts, 1980; Barrowclough, 1982; O’Brien and Mayr, 1991; Thorpe et al., 1991; Avise, 2000). Adherence to this criterion will allow for surmounting of the crisis in subspecies systematics, restoration of the “reputation” of subspecies among taxonomists, and a “spring clean” doing away with “bad” subspecies (Zink, 2004). Here, subspecies monophyly should be understood in a broad sense, that is, including the state of paraphyly. A strict mono phyly (holophyly according to Hennig) is hardly feasi ble at the level of subspecies, since a gene flow between individual subspecies exists, at least, potentially, and the criterion of reciprocal monophyly fails to work (Patten, 2010; Remsen, 2010). The absence of genetic specificity in a subspecies suggests that it is invalid (Panteleev, 1992, 2000). The morphological characters of subspecies should be irre versible, which makes them actual units of evolution (Shvarts, 1980). Reversible changes, no matter how deep they are, reprersent homeostatic processes at a population level rather than the microevolutionary changes, according to Shvarts. The phantom subspe cies lacking geographical specificity not only supple ment the list of synonyms but also actually mislead the conservation policy (Zink, 2004), since the taxonomic

405

406

VINARSKI

formalization of subspecies helps to draw the attention of both scientists and the legislators making decisions on the legal protection of taxa for their conservation (O’Brien and Mayr, 1991; Winston, 1999). The genetic specificity of a subspecies can be con firmed both analytically and experimentally, by trans ferring representatives of one subspecies to the habitat of another (Panteleev, 1992). The latter case directly verifies the irreversibility of subspecies characters. The “evolutionary dimension” of subspecies (sub species as a stage in speciation) is combined with its “ecological dimension,” which regards the subspecies as a type of species adaptation to a specific environ ment (Shvarts, 1980). This is traceable in the views of O. Kleinschmidt on subspecies as intraspecific entities adapted to different environmental conditions. These two aspects neither exclude nor imply one another, since every subspecies is not necessarily an incipient species. The emergence of subspecies is believed to be asso ciated with historical reasons, namely, specific fea tures in the establishment of species distribution ranges, their drastic reductions during glaciations, and subsequent spreading from refugia (Thorpe, 1987). This is the fundamental difference between subspecies and clines: the latter are determined by the current ecological situation and adaptive population response to the impact of environmental factors changing along spatial gradients (Thorpe et al., 1991). Another mech anism contributing to establishment of “good” sub species is the formation of insular isolates. The rate of genetically verified taxa among such subspecies is higher in a statistically significant manner than for continental ones, at least among birds (Phillimore and Owens, 2006). The current molecular systematics provides addi tional possibilities for separating subspecies. First, this is the long known rough correlation between the taxo nomic rank of the compared taxa and the genetic dis tance between them (Borkin and Litvinchuk, 2008). Geographically separated population groups display ing a noticeably lower genetic distance, as compared with the distance between definitely “good” species, may well be regarded as subspecies. For example, the species within a genus of freshwater mollusks belong ing to the family Lymnaeidae are separated by a genetic distance (computed based on cytochrome b gene) of 8.3 to 15.3%. On the other hand, subspecies of these snails display a genetic distance of only 2% (Vinarski et al., 2012b). This approach can be helpful when resolving frequently occurring “species or sub species” dilemmas of practical taxonomy. For exam ple, there is an “eternal question” on the status of the pig and human ascarids, which some authors believe to be species and others consider subspecies (Macko, 1983). An evident weakness of genetic distance as a crite rion is that it is basically impossible to determine its cutoff value for distinguishing between species and

subspecies. Therefore, this parameter is applicable only within the integrative taxonomy, i.e., along with other characteristics enhancing identification of taxo nomic rank. Phylogeography as one of the actively developing areas in molecular phylogenetics also does not exclude the existence of subspecies (Avise, 2000). Although allopatric phylogenetic lineages are usually regarded as “cryptic species,” it is possible that two populations have considerably genetically diverged yet retain the ability to unrestrictedly exchange genes in the zone of their secondary contact. Such populations may be regarded as subspecies (Mallet, 1995; Jorgensen et al., 2013). Recall that the reproductive isolation in the PSC is irrelevant to the essence of the species cate gory; thus, PSC supporters even in the above situation will not separate subspecies by referring to the unique ness of the evolutionary fate of these populations, which itself is sufficient to regard them as “good” spe cies (Cracraft, 1992). Thus, the use or rejection of the subspecies cate gory at present is to a considerable degree determined by the theoretical views of the individual taxonomist, making it contextually dependent and, provided the total dominance of PSC and similar concepts, des tined for gradual extinction in the future. A harbinger of this is the complete absence of a subspecies section in some of the newest taxonomy guidelines (Wiley and Lieberman, 2011; Wheeler, 2012). MICROSYSTEMATICS AND THE DIVERSITY OF SUBSPECIES CATEGORIES If systematics is regarded as a science that studies and describes biological diversity, it should be kept in mind that that biodiversity is a complex phenomenon that does not boil down only to taxonomic diversity. There are other aspects, in particular, morphological (Pavlinov, 2010), functional (Petchey and Gaston, 2002), and phylogenetic (Winter et al., 2013) diversity. Many of them appear at an intraspecific level and fall within the scope of systematics. A taxonomist requires a toolkit that makes it possible to describe various biodi versity aspects and express them in nomenclature. Unfortunately, the acting International Code of Zoolog ical Nomenclature (International Commission..., 2000) actually limits the intraspecific level by only one cate gory, subspecies. The numerous difficulties and failures in practical use of the theoretically rather clear concept of geo graphical subspecies described above are, according to my understanding, partly of a psychological nature, being determined by the position of subspecies as the formally terminal rank in the zoological systematics, inadvertently making this category more significant than it actually deserves. Species in nature are heterogeneous, as is suggested by numerous manifestations of morphological and genetic polymorphism, geographical variation, and so BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

THE FATE OF SUBSPECIES CATEGORY

on. The attempts of taxonomists to order this hetero geneity and cope with it using the available tools led to a question about the divisibility limit, i.e., a certain minimal level of taxonomic resolution that makes fur ther division senseless. The Ancient natural philoso phy when considering the divisibility boundary of material objects gave birth to the concept of the atom as the limit of any divisibility and an absolutely indivis ible entity “lacking any emptiness” inside. What is the atom for taxonomists? Presumably, an objective divis ibility limit is an elementary population (here we omit the facts that the boundaries of such a population in Nature are frequently hardly determinable and that a distinct population structure is far from being evident for many “lower” organisms). From a practical stand point, it is also clear that it is absurd to describe each population as a special taxon (Starobogatov, 1996). However, it is possible to attempt to create a system of categories of the most inferior rank to designate all of the manifestations of intraspecific variation that are more or less significant in the eyes of a taxonomist, be it geographical or nongeographical variation. In the Russian animal systematics, works by A.P. Semenov TianShanskii (1910) and L.S. Berg (1916) were clas sic examples of this approach. This is not an ideal solu tion. The abundance of infrasubspecific categories and their rather inconsistent use made this system cumber some, while Latin names of the most inferior rank were numerous and difficult to remember. The proposal to regard the species as the terminal taxon (Terent’ev, 1968), which was implemented in PSC (Cracraft, 1992), was the alternative. However, the infrasubspecific categories began disappearing from practical systematics even earlier, in the mid 20th century. Presumably, the main reason here was the decision of the International Commission on Zoo logical Nomenclature to abandon the regulation of any infrasubspecies entities since 1960 (Mayr, 1953). It should be emphasized that the Code does not prohibit their use but only refuses to establish rules for their treatment. Thus, subspecies became the terminal taxon subject to regulation by the Code. This makes it a sort of the taxonomic “atom,” although the subspe cies category itself cannot bear any special meaning. Subspecies is no more than a intercalary taxonomic rank without any obligatory usage. A genus should not be obligatory divided into subgenera; similarly, a spe cies is not obligatory polytypic. In the hierarchy of supraorganismal biological sys tems, the position of subspecies is in between an ele mentary population and a fullfledged species. An objective definition of the upper and lower boundaries of this “in between” is most likely infeasible, interfer ing with avoidance of subjectivity in subspecies taxon omy (Fitzpatrick, 2010; Winker, 2010). Even the “rule of 75%,” which is the most popular for defining sub species, is just a useful convention. The reference to the evolutionary dimension, that is, the interpretation of subspecies as incipient species, BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

407

does not save the situation. The course of evolution is unpredictable. In observation of an extant subspecies, it is impossible to forecast whether it will become a fullfledged species or not. The latter variant is likely to be more frequent. A subspecies may disappear not only by extinction but also by merging with another subspecies in the zone of their secondary contact (Pat ten, 2010). The evolutionary significance of a subspe cies is assessable only post factum, when its individual history has come to its end. Subjectivity in defining the subspecies rank, as well as manifold errors in the practical use of the subspecies concept, still does not make the concept itself sense less (Winker, 2010). In my opinion, the subspecies as a special rank in the Linnaean hierarchy should be not only preserved but also supplemented with a number of other subspecific categories, since modern system atics combines morphological methods with molecu lar genetic tools and does not always give an unambig uous result. The molecularization of taxonomy has detected a more complex biodiversity structure at the below species level than has been previously assumed, and the simple scheme of “species and its subspecies” has emerged to be evidently insufficient. Within the biological species concept, this problem is solved in part by introducing the superspecies category formu lated in the 1950s–1960s (SylvesterBradley, 1954; Amadon, 1966). A superspecies is a monophyletic group of allopatric species (allospecies) that are closely related, and phenotypically very similar but still too separate to be regarded as subspecies of the same species. The term allospecies has been intro duced to denote a boundary situation when the species or subspecies dilemma cannot be unambiguously resolved (Amadon, 1966). This is a terminal “prespe cies” stage in allopatric speciation, when there is no way to determine objectively whether speciation has been completed. With their nonoverlapping distribu tion ranges, allospecies are effectively isolated by dis tance; however, this does not directly imply that they are reproductively isolated in a physiological or behav ioral manner. According to Amadon, Canadian and European beavers, as well as the European bison and American buffalo, are examples of allospecies. The process of speciation is continuous; it can be further split to separate the taxa that correspond to more or less isolated stages of divergence. In particu lar, Amadon and Short (1976) have defined one more subspecies category, namely, megasubspecies, which corresponds to a subspecies or a cluster of subspecies known or assumed to approach the status of a separate species. Megasubspecies itself may be polytypic and comprise two or more subspecies. These taxa are phe notypically more separated than“common” subspe cies, but these categories cannot be distinctly distin guished (Amadon and Short, 1976). Presumably, this is the explanation why the megasubspecies category has failed to survive.

408

VINARSKI

Considering the species as a process of establish ment rather than an already established entity, that is, in its evolutionary dynamics, we frequently observe that independent groups of characters display nonuni form divergence rates. Phenotypic divergence may fall behind or keep ahead of genetic divergence; mito chondrial DNA mutates at a higher rate than nuclear DNA (Ballard and Whitlock, 2004); and so on. Molecular reductionism, a popular trend nowadays, tends to ignore morphology and reduce the overall his tory of a species to neutral genetic evolution. This is very operational but oversimplifies the actual subspe cies structure of biodiversity. Using selectively neutral genetic markers as the sole divergence indicators of (which is a frequent situation now), we are “geneti cally shortsighted” by completely ignoring the factor of natural selection (Remsen, 2010). An asynchrony in divergence rates may provide an explanation for the numerous cases in which pheno typically highly separated subspecies are genetically indistinguishable (Zink, 2004), as well as in the oppo site situations, in which phylogeographical analysis detects allopatric interspecific groups that are sepa rated by a large genetic distance yet are phenotypically indistinguishable (cryptic molecular lineages). Several reviews demonstrate that the average congruence between systems based on morphology and genetics may be considerably lower than 100%. For example, only 63% of the animal genera defined based on mor phology are verified by genetic tools (Jablonski and Finarelli, 2009). The absence of genetic support inter feres with the work of the taxonomist; however, this is not sufficient to state that the corresponding taxon is invalid. Even if one accepts that genetic systematics has worked correctly (and this is not always the case; see Groeneberg et al., 2011), it is incorrect to a priori regard genetic characters as more significant than the phenotypic ones (Lee, 2004). The ideal situation is when the molecules and phenotypes sing in unison; otherwise, it is more reasonable to include the maxi mum information about the studied organisms into the system rather than limiting ourselves to genetic data only. When studying biodiversity, the morpholog ical and genetic approaches are linked by Bohr’s com plementarity principle and thus, taken separately, give only a simplified vision of the actual intraspecific structure. In my opinion, it is desirable to return to a more fractional system of subspecies categories but one that is more operationally defined than in the classics (SemenovTianShansky, 1910) and that takes into account the specificity of molecular characters. An increase in the assortment of such categories will hardly cause any revolution in practical systematics, at least because a complex intraspecific structure is detectably only for a small number of highly variable species with broad distribution ranges. In the remain ing cases, there is no need in this expanded system. In fact, this is an implementation of an old idea by Sim

pson (1945, p. 23), who wrote: “It is desirable that all distinguishable groups should be distinguished (although it is not necessary that all enter into formal classification and receive names)”. I would like to illustrate this statement by an exam ple encountered when working on an “integrative” systematics of the freshwater mollusks of the family Lymnaeidae. This is the commonly known great pond snail Lymnaea stagnalis (L., 1758), a widely abundant Holarctic species and object of long debates by taxon omists. The elaboration of the microsystematics for this species is a case study in which the idea of diversity of subspecies categories finds its practical application. MICROSYSTEMATICS OF THE COMPLEX Lymnaea stagnalis S. LATO A characteristic of the species Lymnaea (Lymnaea) stagnalis s. lato is the unusual plasticity of its shell shape and its proportions on the background of a rela tive constancy in the structure of its reproductive sys tem (Hubendick, 1951). Although the proportions of the copulative organ, which is regarded as a significant taxonomic character, may differ within the subgenus Lymnaea s. str., the qualitative structure of this great pond snail organ is uniform (Kruglov, 2005). None theless, taxonomists of the past distinguished at least twenty varieties within the species L. stagnalis s. lato based on the differences in their shell habitus (Wester lund, 1885); moreover, some authors even went so far as to consider them independent species (Locard, 1893). The standpoint of Hubendick (1951) that, in the absence of qualitative distinctions in the structure of reproductive system, the conchological differences and the taxa derived from them should be ignored was recognized in the malacology of the mid20th century. Until now, most of the specialists have regarded L. stagnalis as a single species that is not divided into subtaxa despite its high intraspecific variation (Jackie wicz, 1998; Glöer, 2002). The exception is the system proposed by Kruglov and Starobogatov (1985), which divides L. stagnalis s. lato into six independent species (Fig. 1) differing in the shell characteristics and pro portions of the copulative organ. As has been shown, at least two of these species, L. stagnalis s. str. and L. fra gilis, are reproductively isolated (Davydov et al., 1981). All of the systems mentioned above were based exclusively on morphological data. A detailed analysis of the morphological and allozyme variations in the L. stagnalis s. lato from Ukrainian aquatic bodies gave results contradicting “…the concept of two sympatric species, L. stagnalis and L. fragilis (Mezhzherin et al., 2008, p. 339). Mezhzherin et al. (2008) proposed a new scheme comprising two allospecies connected by a narrow zone of gene introgression. These allospe cies have not been formally described and lack any taxonomic names. BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

THE FATE OF SUBSPECIES CATEGORY

(a)

(c) (a)

409

(b) (b)

(c)

(f)

(d) (e)

(d)

(e)

Fig. 1. Shells of the species belonging to the subgenus Lym naea s. str., distinguished in the system proposed by N.D. Kruglov and Ya.I. Starobogatov (European taxono mists consider them within the species L. stagnalis s. lato): (a) L. stagnalis s. str.; (b) L. fragilis; (c) L. doriana; (d) L. bodamica; (e) L. araratensis; and (f) L. media. Scale bar, 1 mm.

Fig. 2. L. fragilis shells from different parts of the Palearc tic: (a) Tyumen oblast, a waterbody in the town of Labyt nangi; (b) Omsk oblast, a waterbody near the village of Mezhdurech’e; (c and d) Pskov oblast, Venyato Lake; and (e) Montenegro, Skadar Lake. The individuals belong to L. stagnalis (a and b) Asian, (c and d) European, and (e) Balkan genetic lineages. Scale bar, 1 mm.

A study of genetic variation among the L. stagnalis s. lato from the Palearctic waterbodies, from France in the west to Transbaikalia in the east, was conducted by the author in collaboration with German colleagues (Vinarski et al., 2012a), and it has demonstrated the complex phylogeographical structure of this species, which includes at least three independent genetic lin eages. One of these lineages is abundant in eastern Europe, Siberia, and western and central Asia; another inhabits western and central Europe; and the third has a small distribution range in the south of Europe (Balkans and Italy). The morphological dif ferences in the structure of shell, radulae, and repro ductive system have been undetectable, although the genetic distance between the lineages corresponds to the value between “good” pond snail species that are separated both genetically and morphologically. It is quite possible to agree with Mezhzherin et al. (2008) and regard such lineages as allospecies in Amadon’s definition (Amadon, 1966). However, the situation is complicated, since mor phological analysis of the shell variation makes it pos sible to distinguish phenotypic groups within the com plex L. stagnalis s. lato that lack individual distribution ranges and are not likely to be the products of ecolog ical variation (ecophenotypes). For example, the auriculate great pond snail form, known as L. doriana (Bourguignat, 1862), L. lacustris (Studer, 1820), L. bodamica (Miller, 1873), and L. media (Hartmann, 1840), is mainly found in large lakes of northern and

central Europe but has been also recorded in the Altai and even Tuva (Kruglov, 2005). This form is frequently regarded as a specific adaptation to the surf zone of large lakes; however, the mollusk samples from the collection of the Zoological Institute (Russian Acad emy of Sciences) demonstrate that great pond snail variants with a shortened shell spire are also found in other biotopes, particularly in channels. On the other hand, the auriculate form is absent in the examined large lakes of the Southern Urals, with their stony lit toral and surfbeaten shores; all of the great pond snail individuals there have a normally developed shell spire.

BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

Another form, with a characteristic high and subu late spire and poorly swollen body whorl (Fig. 1b), has been observed in all three molecular lineages. It is met in waterbodies of western and eastern Siberia, Mongo lia, eastern Europe (Pskov oblast), Ukraine, the Bal kan Peninsula, Sweden, and so on; however, the indi viduals with this phenotype do not form a separate genetic cluster and independently appear in different molecular lineages (Fig. 2). The subulate morph of the great pond snail has been repeatedly described under different names, such as L. producta (Colbeau, 1859), L. raphidia (Bourguignat, 1862), and L. subulata (Westerlund, 1871). Kruglov and Starobogatov (1985) believe that the most proper name for this species is L. fragilis (L., 1758), which is used in the Russian litera ture.

410

VINARSKI

Table 1. Discriminant analysis of the conchological variation of the “minor” species belonging to the subgenus Lymnaea s. str. Group (predicted) L. fragilis L. stagnalis L. doriana L. media L. araratensis

Rate of correct identification, % 87.1 91.9 84.8 64.0 100.0

Group (actual)* L. fragilis

L. stagnalis

202 10 0 0 0

30 180 1 0 0

L. doriana

L. media

0 2 39 9 0

L. araratensis

0 0 5 16 0

0 4 1 0 5

Average, 85.6% * Species of the mollusks were identified using the keys by N.D. Kruglov (2005).

It is known that the individuals belonging to various “minor” species within the complex L. stagnalis s. lato can syntopically coexist without giving intermediate or transition forms (Davydov et al., 1981). Discrimi nant analysis confirms morphological isolation of these species (Table 1), although the “clouds” they form in multidimensional space partially overlap (Fig. 3). The average values of the shell morphometric indices in the representatives of such species under syntopic conditions differ significantly, which, according to Starobogatov (1996), is indirect evidence of their independence. However, in the absence of sta tistically significant genetic distinctions, we have to refer to them as morphotypes (Vinarski et al., 2012a) that are similar to the discrete morphotypes detected in other mollusk species (Weigand et al., 2012). What is the significance of such morphotypes for systematics? From the standpoint of molecular taxo nomic reductionism, this significance is almost zero. Failing to form monophyletic clusters on cladograms, having no isolated distribution ranges, and displaying no statistically significant differences for standard genetic markers, such groups of phenotypically similar individuals have almost no chances of being recog nized in modern geneoriented systematics. However, I would like to once again emphasize that the standard genetic markers, which were also used in our work (Vinarski et al., 2012a), are selectively neu tral. Since the differences in the shell shape and pro portions in the pond snails are of evident adaptive sig nificance and are determined by the differences in their living conditions (Starobogatov, 1967), it cannot be excluded that the morphotype establishment is not reflected in the divergence of these markers, which follows the pattern of a neutral evolution. The genes responsible for reproductive isolation (the socalled speciation genes; Wu and Ting, 2004; Nosil, 2011) are also hardly identical to these markers. That is why only whole genome sequencing of the mollusks belonging to different morphotypes can finally confirm or reject the statistically significant genetic differences between them.

Assessment of the rates of L. fragilis and L. stagnalis morphotypes in the waterbodies of Western Siberia has demonstrated an evident regular pattern in their geo graphical distribution. L. stagnalis is undoubtedly prevalent in the south of this region with its share decreasing northward to be completely replaced by L. fragilis to the north of 64° N (Fig. 4). I believe that this is good indirect evidence of the existence of cer tain genetic differences between these morphotypes, which, however, are to be determined. The coexistence of several morphotypes within a local mollusk fauna of a certain habitat contributes to the functional diversity of the freshwater community, which is an independent biodiversity aspect supple mentary to the taxonomic diversity (Petchey and Gas ton, 2002). At a superspecific level, the diversity of life forms, which in pond snails is subject to latitudinal variation (Nekhaev, 2011), corresponds to the func tional diversity. The classification of Lymnaeidae life forms has been elaborated (Kruglov, 2005); the mor photypes also deserve special classification. Thus, several situations have been observed within the complex L. stagnalis s. lato that require special cat egories for their designation. The allopatric groups that are phenotypically indis tinguishable but noticeably divergent from a genetic standpoint (at the level of good species) can be referred to as allospecies united into a superspecies following Mezhzherin et al. (2008). For the first time, the super species category was applied to the pond snails by Hubendick (1951), who separated the superspecies L. auricularia s. lato, comprising several geographical races. Later, taxonomists (Jackiewicz, 1998; Kruglov, 2005) did not support this idea; however, it again becomes topical in the epoch of molecular taxonomic boom. The allospecies living in northern and central Europe (the type locality for the Linnaean species Helix stagnalis) should retain its original name, Lym 1

naea [stagnalis] stagnalis. According to the priority rule, the southern European allospecies should be referred to as L. [stagnalis] raphidia (Bourg.). Finally, BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

THE FATE OF SUBSPECIES CATEGORY

411

5 4 3 2

Root 2

1 0 –1 –2 –3 –4 –5 –6 –6

–4

–2

Fragilis

0

2

Stagnalis

6

4 Root 1 Doriana

8

Media

10

12

Araratensis

Fig. 3. Distribution of the individuals belonging to “minor” species of the complex Lymnaea stagnalis s. lato in the space of the two first canonical axes.

the third, eastern European–Asian allospecies, by all accounts, should be named L. [stagnalis] acicularis (Westerlund, 1894), since this is the oldest of the appropriate names proposed for the “great pond snail” taxon with a type locality in Asia (north of western Siberia). The morphotypes in this classification are pheno typically separate groups with undetected genetic dif ferences lacking isolated distribution ranges, i.e., they able to live syntopically. In our case, the status of mor photypes is ascribed to L. fragilis (the subulate form with an cone, which would be better named L. producta according to nomenclature reasons), L. lacustris (also including other forms with an auricu late shell: L. bodamica, L. doriana, and L. media), and L. araratensis (the form with convex whorls of the spire and comparatively small size, with the shell height not exceeding 30–35 mm). Presumably, a “typ ical” L. stagnalis (Fig. 1a) should be considered a spe cial morphotype. Finally, the provisional system of the Palearctic “great pond snails” takes the following form: —Superspecies Lymnaea stagnalis s. lato; —Allospecies Lymnaea [stagnalis] stagnalis (includes the morphotypes L. stagnalis f. stagnalis, L. stagnalis f. producta, and L. stagnalis f. lacustris); 1 The

square brackets in trinomial Latin name of allospecies are used in accordance to Amadon’s recommendation (Amadon, 1966). BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

50

60

70

80

90

100

110

70

Surgut

60 Novosibirsk

Yekaterinburg

Astana

50

Scale 1 : 25000000 60

70

80

Fig. 4. Ratio of fragilis and stagnalis morphotypes in local faunas of western Siberia. The area of black sector is pro portional to the fragilis morphotype.

412

VINARSKI

Table 2. Characterization of the lower taxonomic catego ries discussed in the paper Type of geographical distribution Genetic distance

allopatry or parapatry

sympatry (syntopy)

Absent or very short*

Subspecies

Morphotype

Considerable**

Allospecies

Species

Not determined

Morphospecies

* In pond snails, this correspond to a distance of 0–2% accord ing to cytochrome b gene. ** In pond snails, this correspond to a distance of 8% and more according to cytochrome b gene.

—Allospecies Lymnaea [stagnalis] raphidia (includes the morphotypes L. raphidia f. stagnalis, L. raphidia f. producta, and, probably, L. raphidia f. lacustris); and —Allospecies Lymnaea [stagnalis] acicularis (includes the morphotypes L. acicularis f. stagnalis, L. acicularis f. producta, L. acicularis f. lacustris, and L. acicularis f. araratensis). The morphotypes producta and stagnalis are repre sented in all allospecies; the morphotype lacustris has not been reliably discovered in L. [st.] raphidia; and the morphotype araratensis, in the allospecies L. [st.] raphidia and L. [st.] stagnalis. This scheme is not only provisional, but also prag matic. The diversity of subspecific categories reflects different manifestations of the intraspecific diver gence, and the classification takes into account two factors—the type of geographical distribution and genetic distance (Table 2). The reader can note that the proposed system does not take into account the factor of reproductive isola tion. This is because the allospecies are allopatric in principle and are by definition well isolated by dis tance (Amadon, 1966). The experiments on hybrid ization of the species L. fragilis and L. stagnalis (Davy dov et al., 1981) should not be discounted, but their contradiction to the allozyme data was not explained by Mezhzherin et al. (2008). On the other hand, as Kornyushin (2002) noted, these results still await con firmation by an independent research team, while the autogamy of Lymnaeidae interferes with the interpre tation of hybridization results. Presumably, the allospecies can preserve reproductive compatibility and freely hybridize in the case of a secondary contact. As for the subspecies category, it is possible to define it for Lymnaeidae and related groups as allopat ric population groups that display morphological specificity but are separated by a very short genetic dis tance. The subspecies thus interpreted are known among the pond snails (Vinarski et al., 2012b), but it is impossible to distinguish the taxa of such rank for the superspecies L. stagnalis.

The scheme described above considers the cases with available data on the genetic distances between the compared groups of individuals. The category of morphospecies is proposed for forms that display phe notypic differences but are still genetically unexam ined. In doing so, we emphasize that the species status needs independent confirmation by molecular char acters. If a morphospecies is shown to be genetically discrete or not, it can be ascribed to one of the above categories with the already established operational cri teria (Table 2). CONCLUSIONS The history of development and practical applica tion of the subspecies category demonstrates that, despite all of the mistakes and excessive use, it reflects a certain natural reality and should hardly be rejected, as is proposed by the supporters of PSC and rankfree taxonomy. However, modern animal systematics, with its molecular genetic toolkit, detects previously hid den aspects of intraspecific diversity, which requires new solutions for their description and expression using the tools of systematics. An intricate picture of intraspecific divergence manifests itself in the forms that are beyond the frame of subspecies concept. The fact that varieties and other subspecies categories were discarded from the International Code of Zoological Nomenclature is most likely associated with the absence of operational criteria for their unambiguous application and the vagueness of their content. Molec ular taxonomy possesses the tools to more precisely define the meaning ascribed to particular terms and guarantee that none of the biodiversity aspects are missed by systematics. ACKNOWLEDGMENTS The work was supported by the Russian Founda tion for Basic Research, project no. 140401236 and Ministry of Education and Science of the Russian Federation, project no. 6.1957.2014/k. The author thanks his German colleagues, Katrin Schniebs (Dresden) and Peter Glöer (Hetlingen), for longterm collaboration in freshwater mollusk systematics and for the beneficial discussion of several issues related to the taxonomy of the complex L. stagnalis s. lato, as well as I.O. Nekhaev and an anonymous reviewer for their valuable criticism. REFERENCES Amadon, D., The superspecies concept, Syst. Zool., 1966, vol. 15, pp. 246–249. Amadon, D. and Short, L.L., Treatment of subspecies approaching species status, Syst. Zool., 1976, vol. 25, pp. 161–167. Avise, J.S., Phylogeography: The History and Formation of Species, Cambridge, MA: Harvard Univ. Press, 2000. BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

THE FATE OF SUBSPECIES CATEGORY Ballard, J.W.O. and Whitlock, M.C., The incomplete natu ral history of mitochondria, Mol. Ecol., 2004, vol. 13, pp. 729–744. Barrowclough, G.F., Geographic variation, predictiveness, and subspecies, Auk, 1982, vol. 99, no. 3, pp. 601–603. Berg, L.S., Ryby presnykh vod Rossiiskoi imperii (Freshwater Fishes of Russian Empire), Moscow: Tipogr. Ryabush inskikh, 1916. Borkin, L.Ya. and Litvinchuk, S.N., Genetic distances and speciation in Amphibians, in Mater. III s”ezda Gerpeto logicheskogo ova im A.M. Nikol’skogo “Voprosy gerpe tologii” (Proc. III Congr. of the Nikolsky Herpetologi cal Society “Problems of Herpetology”), St. Peters burg, 2008, pp. 41–52. Cracraft, J., Species concepts and speciation debates, in The Units of Evolution: Essays of the Nature of Species, Ereshefsky, M., Ed., Cambridge, MA: MIT Press, 1992, pp. 93–120. Davydov, A.F., Kruglov, N.D., and Starobogatov, Ya.I., Laboratory breeding of two forms of Lymnaea stagnalis and problem of systematics of subgenus Lymnaea s.str. (Gastropoda, Pulmonata), Zool. Zh., 1981, vol. 60, no. 9, pp. 1325–1337. Fitzpatrick, J.W., Subspecies are for convenience, Ornitol. Monogr., 2010, vol. 67, pp. 54–61. Hubendick, B., Recent Lymnaeidae. Their variation, mor phology, taxonomy, nomenclature, and distribution, Kung. Svensk. Vetensk. Handl., Ser. 4, 1951, no. 3. International Commission on Zoological Nomenclature, International Code of Zoological Nomenclature, Lon don: Int. Trust Zool. Nomenclature, 1999, 4th ed. Glöer, P., Die Süβwassergastropoden Nord und Mitteleu ropas: Bestimmungschlussel, Lebenweise, Verbreitung, in Die Tierwelt Deutschlands. 73. Teil, Hackenheim: Conchbooks, 2002. Groenenberg, D.S.J., Neubert, E., and Gittenberger, E., Reappraisal of the “Molecular phylogeny of Western Palaearctic Helicidae s.l. (Gastropoda: Stylommato phora)”: When poor science meets GenBank, Mol. Phylogenet. Evol., 2011, vol. 61, pp. 914–923. Jablonski, D. and Finarelli, J.A., Congruence of morpho logicallydefined genera with molecular phylogenies, Proc. Natl. Acad. Sci. U.S.A., 2009, vol. 106, pp. 8262– 8266. Jackiewicz, M., European species of the family Lymnaeidae (Gastropoda, Pulmonata, Basommatophora), Genus, 1998, vol. 9, no. 1, pp. 1–93. Jorgensen, M.C., Sierwald, P., and MasonGamer, R.J., A review of subspecies recognition in polydesmidan mil lipedes (Diplopoda) with a revision of the subspecies of Euryuridae (Xystodesmoidea), Zool. Scr., 2013, vol. 42, no. 3, pp. 317–326. Kornyushin, A.V., Species composition of bivalve mollusks of Ukraine and their protection strategies, Vestn. Zool., 2002, vol. 36, no. 1, pp. 9–23. Kruglov, N.D., Mollyuski seveistva prudovikov (Lym naeidae, Gastropoda, Pulmonata) Evropy i Severnoi Azii (Osobennosti ekologii i parazitologicheskoe znachenie) (Mollusks of Family Lymnaeidae (Gastropoda, Pulmo nata) from Europe and Northern Asia: Ecological Fea tures and Parasitological Significance), Smolensk: Smolensk. Gos. Pedagog. Univ., 2005. BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015

413

Kruglov, N.D. and Starobogatov, Ya.I., Methods of experi mental hybridization and some results of its applica tions in the taxonomy of Lymnaeidae, Malacol. Rev., 1985, vol. 18, pp. 21–35. Locard, A., Coquilles des Eaux Douces et Saumatres de France. Description des Familles, Genres et Especes, Lyon: A. Rey, 1893. Lee, M.S.Y., The molecularization of taxonomy, Invertebr. Syst., 2004, vol. 18, pp. 1–6. Macko, J.K., On the subspecies in parasites with respect to helminthes, Folia Parasitol., 1983, vol. 30, pp. 107– 116. Mallet, J., A species definition for the modern synthesis, Trends Ecol. Evol., 1995, vol. 10, pp. 294–299. Mayr, E., Systematics and the Origin of Species, from the Viewpoint of a Zoologist, Cambridge: Harvard Univ. Press, 1942. Mayr, E., Principles of Systematic Zoology, New York: McGrawHill, 1969. Mayr, E., The Growth of Biological Thought, Cambridge, MA: Belknap, 1982. Mezhzherin, S.V., Garbar, A.V., Korshunova, E.D., and Zhalei, E.I., Morphological and genetic variability of the great pond snail Lymnaea stagnalis s. l. (Gas tropoda, Lymnaeidae) in waterbodies of Ukraine, Vestn. Zool., 2008, vol. 42, no. 4, pp. 339–345. Nekhaev, I.O., Geographic distribution features of subge nus and life forms of Lymnaeidae (Gastropoda, Pulmo nata) in Eastern Europe, Biol. Bull., 2011, vol. 38, no. 4, pp. 406–411. Nosil, P., The genes underlying the process of speciation, Trends Ecol. Evol., 2011, vol. 26, no. 4, pp. 160–167. O’Brien, S.J. and Mayr, E., Bureaucratic mischief: recog nizing endangered species and subspecies, Science, 1991, vol. 251, no. 4998, pp. 1187–1188. Panteleev, P.A., A concept of subspecies in systematics, Usp. Sovrem. Biol., 1992, vol. 112, no. 3, pp. 350–358. Panteleev, P.A., Role of temperature in geographic specia tion (an example of Rodents), Russ. J. Ecol., 2000, vol. 31, no. 2, pp. 69–73. Patten, M.A., Null expectation in subspecies diagnosis, Ornitol. Monogr., 2010, vol. 67, pp. 35–41. Pavlinov, I.Ya., Morphological diversity: general trends and characteristics, Sb. Tr. Zool. Muz. Mosk. Gos. Univ., 2010, vol. 49, pp. 343–388. Petchey, O.L. and Gaston, K., Functional diversity (FD), species richness and community composition, Ecol. Lett., 2002, vol. 5, pp. 402–411. Phillimore, A.B. and Owens, I.P.F., Are subspecies useful in evolutionary and conservation biology? Proc. R. Soc. London, Ser. B, 2006, vol. 273, pp. 1049–1053. Remsen, J.V., Subspecies as a meaningful taxonomic rank in avian classification, Ornitol. Monogr., 2010, vol. 67, pp. 62–78. SemenovTianShansky, A.P., Taxonomic borders of spe cies and its divisions: experience of precise systematiza tion of lower systematic units, Zap. Akad. Nauk., Fiz Mat. Otd., 1910, vol. 25, no. 1, pp. 1–29. Shvarts, S.S., Ekologicheskie zakonomernosti evolyutsii (Ecological Patterns of Evolution), Moscow: Nauka, 1980.

414

VINARSKI

Simpson, G.G., The principles of classification and a clas sification of mammals, Bull. Am. Mus. Nat. Hist., 1945, vol. 85, pp. 1–350. Starobogatov, Ya.I., Reconstruction of the system of fresh water mollusks (Pulmonata), Tr. Zool. Inst., Akad. Nauk SSSR, 1967, vol. 42, pp. 280–304. Starobogatov, Ya.I., A species in the theory and nature, in Sovremennaya sistematika: metodologicheskie aspekty (Modern Systematics: Methodological Aspects), Mos cow: Mosk. Gos. Univ., 1996, pp. 165–182. SylvesterBradley, P.C., The superspecies, Syst. Zool., 1954, vol. 3, no. 4, pp. 145–146. Terent’ev, P.V., In defence of a species as the final taxon, Zool. Zh., 1968, vol. 47, no. 6, pp. 887–900. Thorpe, R.S., Geographic variation: a synthesis of cause, data, pattern, and congruence in relation to subspecies, multivariate analysis, and phylogenesis, Bull. Zool., 1987, vol. 54, pp. 3–11. Thorpe, R.S., Brown, R.P., Malhotra, A., and Wuster, W., Geographic variation and population systematics: dis tinguishing between ecogenetics and phylogenetics, Bull. Zool., 1991, vol. 58, pp. 329–335. Vinarski, M.V., Schniebs, K., Glöer, P., and Hundsdoerfer, A., Preliminary phylogeography of Lymnaea stagnalis: are there cryptic species of the great pond snail in XI Int. Congr. on Medical and Applied Malacology, Abstracts of Papers, Rio de Janeiro: Soc. Bras. Malacol., 2012a, p. 66. Vinarski, M.V., Schniebs, K., Glöer, P., Son, M.O., and Hundsdoerfer, A.K., The steppe relics: taxonomic study on two lymnaeid species endemic to the former

USSR (Gastropoda: Pulmonata: Lymnaeidae), Arch. Mollusk, 2012b, vol. 141, no. 1, pp. 67–85. Weigand, A.M., Götze, M.C., and Jochum, A., Outdated but established? Conchologically driven species delin eations in microgastropods (Carychiidae, Carychium), Org. Diversity Evol., 2012, vol. 12, no. 4, pp. 377–386. Westerlund, C.A., Fauna der in der Palaarktischen Region Lebenden Binnenconchylien. V. Fam. Succineidae, Auri culidae, Limnaeidae, Cyclostomidae, und Hydrocenidae, Lund: Ohlsson, 1885. Wheeler, W.C., Systematics: A Course of Lectures, Chiches ter: Wiley, 2012. Wiley, E.O. and Lieberman, B.S., Phylogenetics: Theory and Practice of Phylogenetic Systematics, Chichester: Wiley, 2011. Winker, K.M., Subspecies represent geographically parti tioned variation, a gold mine of evolutionary biology, and a challenge for conservation, Ornitol. Monogr., 2010, vol. 67, pp. 6–23. Winston, J.E., Delimiting Species: Practical Taxonomic Pro cedure for Biologists, New York: Columbia Univ. Press, 1999. Winter, M., Devictor, V., and Schweiger, O., Phylogenetic diversity and nature conservation: where are we? Trends Ecol. Evol., 2013, vol. 28, no. 4, pp. 199–204. Wu, Ch.I. and Ting, Ch.T., Genes and speciation, Nat. Rev Genet., 2004, vol. 5, pp. 114–122. Zink, R.M., The role of subspecies in obscuring avian bio logical diversity and misleading conservation policy, Proc. R. Soc. London, Ser. B, 2004, vol. 271, pp. 561– 564.

Translated by G. Chirikova

BIOLOGY BULLETIN REVIEWS

Vol. 5

No. 5

2015