Homologues and homology and their related terms in phylogenetic systematics

In the field of phylogenetic systematics, the terms homology and homologue and their relationship to cladistic terms such as character, character state, synapomorphy and symplesiomorphy, as well as their relationships to each other, have been and are still discussed frequently. A recent re‐emergence of concepts of homology/homologue free of any reference to explanatory hypotheses prompted us to explore these concepts and their relationships to each other as well as to the concept of morpheme, as introduced recently. All concepts are examined with regard to their ontological status and their bearing in the epistemological process in morphology and phylogenetic systematics. To us, morphemes, homologues and in partem character states refer to things (concrete objects in the ontological sense). However, although morphemes are exclusively descriptive, the latter two represent objects of causal explanations. Homologue always refers to the things themselves, yet a character state also can be a property or the absence of a thing. In this context, a character as a transformation series of character states does not represent a thing but a natural kind. Character states of one character are connected by homology relationships, i.e. common descent. Synapomorphy and symplesiomorphy represent different states of a single transformation series. A nonexplanatory, purely descriptive, concept of homologues is contradictory to its original as well as the post‐Darwinian, evolutionary, concept which refers to causal relationships between parts of organisms and their correspondences in the archetype or ancestor, respectively. Character states, homologues and synapomorphies/symplesiomorphies can only be approximated in the form of hypotheses. We argue that the high value of the concept of homology and its related concepts for evolutionary biology should be maintained by acknowledging their explanatory nature and that dilution with nonexplanatory (even idealistic) definitions should be avoided.


Introduction
Descriptions of organisms and their parts as well as interpretation of these observations in terms of common descent are essential parts of phylogenetic systematics/evolutionary morphology. In this respect, various terms have been discussed and relationally established, including homology/homologue, character state/character and synapomorphy/symplesiomorphy. Herein, we aim to compare the relationships between these concepts and our recently introduced concept of morpheme (Richter and Wirkner, 2014;G€ opel and Richter, 2016).
It has been suggested that homologue(s), a term Richard Owen adopted from German scholars of different disciplines and coined in the context of comparative anatomy (Owen, 1843(Owen, , 1847, would be observable parts of an organism (Nelson, 1994: 104;Fitzhugh, 2016: 455). Recently, Vogt (2017) considered "comparative homology" as a "non-evolutionary comparative account of homology" following "a comparative descriptive style of reasoning that is concerned with the question of what is given here and now." "Comparative homologies . . . are anatomical structural kinds that are directly identifiable" (Vogt, 2017: 513); Vogt also used 'comparative homologues' in his text, which certainly fitted better to his arguments. In addition, homologues would correspond to the character states of the cladistic terminology and might also correspond to synapomorphy, a term introduced by Willi Hennig (Patterson, 1982;Nelson, 2011). We agree that homologues, character states and synapomorphy/symplesiomorphy often (the latter two, however, not always) refer to "things" (sensu Bunge, 1977;Mahner and Bunge, 1997), but we will show that these three terms refer to much more than just observations or descriptions thereof, but all include hypotheses on common descent. It is the morphemes, which are the objects of observation, which need to be conceptualized (i.e. hypothesized) as homologues, character states and synapomorphies/symplesiomorphies (synmorphy sensu Richter, 2017).

Ontology
Two fundamentals regarding ontology and epistemology need to be clarified as the basis.
1. Ontologically, at least in a materialistic scientific realism, concrete objects (mostly referred to as things or individuals) and abstract objects (i.e. classes or kinds of some sort) can be distinguished, with only the former bearing mindindependent existence (Bunge, 1977;Mahner and Bunge, 1997). The question of concreteness or abstractness is therefore of importance for understanding the relationship between these concepts. 2. Epistemologically, description of natural observations and phenomena, on the one hand, and explanation of their causes, on the other, are distinct processes in all natural sciences that should thus be separated by scientists in their accounts as well (Duhem, 1908;Hempel and Oppenheim, 1948). Consequently, entities that are objects of the description are not necessarily the same in the explanation; one famous example would be the spandrels of San Marco (Gould and Lewontin, 1979).
As for ontology, two concepts are essential in this context: "thing" and "natural kind". We are herein following the ontological framework of Mario Bunge (1919Bunge ( -2020e.g. Bunge, 1977;Mahner and Bunge, 1997).
Things are concrete objects, i.e. the real entities that make up the world around us. Things exist independently of any investigator's mind and bear properties (Bunge, 1977). A thing (and its properties) can be described but not defined 1 . In phylogenetic systematics and evolutionary morphology, for instance, individual organisms as well as populations are considered things (Mahner and Bunge, 1997) and so are parts of individual organisms, i.e. morphemes (see below; Richter and Wirkner, 2014;Gӧpel and Richter, 2016;Richter et al., 2020).
Natural kinds, however, are a particular sort of class. Classes in general comprise things that share at least one certain property, yet natural kinds comprise those things that share natural, lawfully related properties (Bunge, 1977;Mahner and Bunge, 1997). In systematics, for instance, taxa are considered natural kinds (Mahner and Bunge, 1997) with their unique common descent being the defining property (Gӧpel and Richter, 2016). The difference between things and natural kinds, and their relationship to phylogenetic systematics becomes most obvious when applied to speciesa debate held for decades with no end in sight (see Zachos, 2016, for a recent account on the "species problem"). In brief, species are either considered natural kinds like other (higher) taxa (Mahner 1993;Mahner and Bunge, 1997) or things, emphasizing ontological differences between species and supraspecific taxa (Ghiselin, 1987). Defendants of species as things refer to their nature as reproductive entity as well as their spatiotemporally restricted existence (Ghiselin, 1987, Zachos, 2016. Their opponents state, for instance, that species (like other taxa) encompass organisms sharing certain properties and are defined by these properties (Mahner and Bunge, 1997) and that the only supraorganismal things are populations as integral systems (Mahner, 1993; with organisms rather being parts of the population than members).
Clarification of the ontological status of the entities relevant in this context (e.g. homologue, character) will aid in understanding the differences of these concepts in the following.

Homologue and homology
One of the most elusive pairs of terms in evolutionary biology (and beyond) is certainly homologue and homology. Both terms refer to the Greek terms homos "same" and logos "relation, reason" (Harper, 2021). It is not our intention to go into all details of Owen's view on homologue and homology, but we refer to more comprehensive accounts (Rupke, 1994;Williams, 2004;Williams and Ebach, 2008). For us the ontological status of homologues is the focus of our interest here. In 1843, Owen gave the well-known definition of homologue, "The same organ in different animals under every variety of form and function" (Owen 1843: 379), a definition showing that homologues do not necessarily refer to things that are identical in form and function. He also used the term homologues for "corresponding parts in different animals being thus made namesakes" 2 (Owen, 1846: 173) which seems straightforward when comparing bones of different vertebrates, for example (Williams, 2004;Williams and Ebach, 2008;Rieppel, 2016). The concept of homology seems more complex, however, when Owen (1847: 175) described the different kinds of homology relationships; special, general and serial homology. Williams (2004: 196) described Owen's view as, "he considered homologues to be empty of meaning without reference to, and interpretation from, the archetype." This is in concordance with Fitzhugh (2016) who suggested that, for Owen, general homology presents a specific causal link from homologue to the archetype-as-cause, with special homology being rather auxiliary to general homology as the "higher" relation or law (Owen, 1847: 175, 199). In Owen's view, homologues are the "real physical entities" (Nelson, 1994(Nelson, , 2011, or things, whereas homology is the relationship between these entities (Nelson, 1994(Nelson, , 2011. In his seminal monograph on Richard Owen, Rupke (1994) interpreted Owen's view of the archetype as some kind of blueprint and the homologues result from this blueprint by some kind of "polarizing force" similar to the force involved in the process of crystallization. This implies certainly a causal relationship between archetype and homologues (Camardi, 2001; although unsatisfactory from our modern point of view), comparable to the causal connection of the existence of a "blueprint" of a Mercedes car and all of the actual Mercedes models driving around the world. This causal interpretation is, however, in contrast to Nelson (2011), who questioned the archetype as an explanation of the "visible" homologues. In our view, it is difficult to imagine that for Owen homologues were just descriptive statements on observations when he introduced the concept of homology. This is in line with him citing Geoffroy St. Hillaire's adoption of the term referring to the same underlying principle (although Owen objected to his claim of identical development as necessary criterion for parts to be homologues; Owen, 1847: 173). In fact, Owen's frequent use of homology and homologue in the same context contradicts the assumption of a substantial conceptual distinctness (as has been interpreted e.g. by Fitzhugh, 2016) of homologue and homology. Owen aims at "proposing a definite name for each distinct bone, declaratory of its special homology" (Owen, 1847: 173) and discusses the search for the "true homologue" (Owen, 1847: 182). Interestingly, in his criticism of the conflation of homologue/homology with analogue/analogy, he explains the "distinction of those relations" by giving the definitions of "analogue" and homologue (instead of "analogy" and "homology"; Owen, 1847: 174 ff.). To us, it is therefore implausible to assume a conceptual distinction between homologue and homology, and Owen appears to have seen homologue and homology as two sides of the same conceptual coin, i.e. thing and relationship. Furthermore, when Owen states that the correspondence of a part or organ (special homology) could "be determined by the relative position and connection of the parts" (Owen, 1847:175), this refers to the epistemology and not the ontology of the concept of homologue.
With Darwin and his successors, not the concept of "homology as causal relationship" but what is regarded as the cause has changed (see, e.g., Fitzhugh, 2006Fitzhugh, , 2016. Now, the nonmaterial archetype became a material reality (Williams, 2004). Darwin (1859: 435) wrote "If we suppose that the ancient progenitor, the archetype as it may be called, of all mammals, had its limbs constructed on the existing general pattern. . . ." Fitzhugh (2016: 456) pointed out that: "homology is not caused by common ancestry. Homology is [emphasis original] the hypothesis of common ancestry, or common cause, regardless of evolutionary connotation or otherwise." This corresponds very well to Remane (1952), in his discussion of his famous "criteria", stating that common descent is not part of the "definition" (i.e. characterization) but the "explanation" for homology 3 . Later, Remane (1955: 172) replaced "explanation" with "explication". This also is in agreement with Hennig's opinion (see also Richter, 2016). In his discussion of Remane's criteria, Hennig (1966: 94) states: 2 "A homologue is a part or organ in one organism so answering to that in another as to require the same name. . ." (Owen, 1866: XII).
But with respect to defining the concept "homology", all three of his "principal criteria" are only accessory criteria that we have to use because the real principal criterionthe belonging of the characters to a phylogenetic transformation series cannot be directly determined.
Certainly, the term "real principal criterion" is not very exact but Hennig's intention is clear.
Remane's criteria refer to homologues (see Williams, 2004;Williams and Ebach, 2008). Hennig's "real principal criterion", however, refers to homology as relationto the belonging to "a phylogenetic transformation series." We agree on the importance of distinguishing between homologue, the "real physical entity" sensu Nelson (2011), and homology, the relation (Nelson, 1989(Nelson, , 1994Fitzhugh, 2016) If, however, homology refers to a hypothesis of causal relationship connecting homologues, it is difficult to see how denoting homologues does not represent a hypothesis itself. It might be true that the reason to regard two things as homologues might be based on "similarity" as suggested by Fitzhugh (2016: 456) or "topographical relations, and/or connectivity" as suggested by Rieppel & Kearney (2002: 63) based on Owen (1847); however, allocations of homologues remain hypotheses (Lauder, 1994). Ontologically, homologues can only exist as homologues if the determining relationship between them is homology; if this is not the case, they are not homologues. Thus, what we denote as homologues are hypothesized manifestations of our homology hypothesis. In our view, both statements represent exactly the same hypothesis. It becomes clear that other concepts are needed to refer to parts of organisms that can be recognized without reference to homology or other causal explanations. Such entities are subject to mere observation and constitute the basis for comparative approaches to identify topographical correspondences as the first step towards homologization.

Morpheme and homologue
Morphemes, as introduced to the field of evolutionary morphology (descriptive units in morphology, see Richter and Wirkner, 2014;G€ opel and Richter, 2016;Richter, 2017;Richter et al., 2020), are all parts of an organism (or semaphoront) that can be recognized and described morphologically. The term is restricted to the level of description. They can be cohesive organ systems, single organs, parts of organs, tissues, portions of tissues, cells and subcellular components. This level of granularity, on which the morphemes of a description are identified and delineated, depends on the means of observation (e.g. naked eye vs. stereo microscope vs. transmission electron microscope; Richter and Wirkner, 2014;G€ opel and Richter, 2016). As an existing constituent of a single organism, each morpheme represents a thing (G€ opel and Richter, 2016) that can be observed and described and might thus be considered given here and now. This is not invalidated by the obvious fact that no description is theory-free as a consequence of perception and processing in the human mind. However, descriptions of morphemes should be as theory-neutral as possible, especially in regard to directly related explanatory hypotheses (Wirkner et al., 2017;Richter et al., 2020).
Epistemologically, our perception and description of morphemes as parts of organisms remain on the descriptive level and are mere observation statements on their form without account on putative explanation, such as evolution. It is only with homology assessment, i.e. the formulation of homology hypotheses, that a transition to the explanatory level is made. Only then can some morphemes of different organisms be hypothesized as homologues, i.e. as being connected via a relationship of (special) homology as parts of the same evolutionary transformation series (general homology) (Richter and Wirkner, 2014;G€ opel and Richter, 2016;Wirkner et al., 2017). In our view, this does not exclude that homologues also exist within populations or species, but this is not relevant for phylogenetic systematics. Therefore, morphemes are not the entities of any phenetic approach in systematics because they do not refer to any methodology in reconstructing phylogenies.
Morpheme and homologue might oftentimes refer to the exact same thing as part of the described organism, yet their perspective remains distinct (descriptive vs. explanatory). As pointed out in other accounts, this distinction (at least on the conceptual level) is valuable as not all morphemes can be evolutionarily individualized or are inheritable (Wagner, 2014: ch. 2;Richter and Wirkner, 2014). For instance, a single red blood cell is a morpheme and can be the object of a morphological description. Two single red blood cells from two individuals belonging to different species cannot, however, be individually homologized: they lack biological individuality and inheritability as prerequisites for homologization. Objects of homology hypotheses, i.e. homologues, thus are the entirety of red blood cells of these two species (Wagner, 2014). However, morphemes that can be individualized also can lack inheritability. For instance, numerous arteries (single arteries or arterial patterns) of the vascular system of the marbled crayfish (Procambarus virginalis Lyko, 2017) can be described as morphemes and differ between individuals (Scholz et al., 2021). However, because this species is isogenetic (Gutekunst et al., 2018), all of these cases of intraspecific variability are due to nongenetic variation and thus are not inheritable, and therefore cannot be homologized (Richter et al., 2020;Scholz et al., 2021). The difference of morpheme as unit of description and homologue as explanatory concept becomes especially clear in cases such as these.

Character and character state
At this point, we need to introduce a pair of terms crucial for cladistic analyses. Our concepts of character state and character are based on Hennig's "ideographic character concept" (Hennig, 1966;Grant and Kluge, 2004; the transformational homology of Patterson, 1982). Although Hennig might have considered systematics as nomothetic science with focus on the observation of "characters" (but now generally called character states), his concept of "transformation series" (now generally called characters) which is derived from his "deviation rule" with distinguishing plesiomorphic and apomorphic character states is certainly historical/evolutionary and therefore ideographic sensu Grant and Kluge (2004). Character states (which Hennig referred to as characters) are the inheritable entities that undergo transformation into other such entities, i.e. evolution from one character state to another (Hennig, 1966;Grant and Kluge, 2004). Characters (which Hennig referred to as transformation series) encompass all character states that originate from the same original character state in the common ancestor (Hennig, 1966). As such, a character is a natural kind defined by the common evolutionary origin (i.e. homology) of the character states which it encompasses (G€ opel and Richter, 2016). The incongruence of morpheme and character (i.e. transformation series) follows directly from the ontological status of our concepts.
The incongruence of morpheme and character state, however, needs more consideration. The most important difference between morpheme and character state lies in their object of reference. Morphemes refer to single organisms (although their description might be generalized on the species level), yet character states refer to the units of evolution, which might be populations or species (depending on the species concept) (Wagner, 2014: ch. 2;G€ opel and Richter, 2016). This is why polymorphous character states can occur in a species, but the polymorphy cannot be seen in the morphemes of a single individual organism. Furthermore, not each morpheme is represented by a (separate) character state, either due to lack of biological individuality (Wagner, 2014) or lack of inheritability (e.g. phenotypic plasticity; Richter and Wirkner, 2014;Scholz et al., 2021).
In some cases, all homologues might be represented by a single character state (i.e. the state "present" in neomorphic characters sensu Sereno, 2007). In other cases, however, homologues might well be represented by different character states (i.e. different character states in a transformational character sensu Sereno, 2007). In general, it is the hypothesis of homology that connects the different character states of a single character (i.e. transformation series; Hennig, 1966). However, identifying neomorphic characters (or more precisely their state "present") as well as the identical state of a transformational character in different organisms certainly also includes some kind of homology hypothesis (Szucsich and Wirkner, 2007). One example: the bones referred to as "humerus" in different tetrapod species are certainly homologues in Owen's sense and most probably in the interpretation of all his successors. The hypothesis of "character state identity" has been called "primary homology" (de Pinna, 1991;Brower and Schawaroch, 1996), although the term "primary homologue" might be a better fit in the case that the reference is made to actual things (and not properties, see below).
However, character states in transformational characters mostly refer to properties such as material composition, shape or colour, and thus are manifested in the properties of morphemes/homologues rather than their mere existence (Hennig, 1966: 89-90;Sereno, 2007). The humerus of one species being "short" in comparison to the humerus in another being "long" refer to the same homologue but represent different properties. In a transformational character "humerus, length", "short" and "long" represent the character states as property of the homologue "humerus". The hypothesis of "character state identity" is therefore more than identifying homologues because it includes properties of homologues as well.
In any case, "character state identity" hypotheses are not mere descriptions but always follow an intellectual process that we call character (state) conceptualization (Wirkner and Richter, 2010; see also Kearney and Rieppel, 2006). Based on the description of a morpheme, such as the hand of a human and of a chimpanzee, which are very similar but not identical, character states on different levels can be conceptualized. The "presence of the thumb" might be considered as one of these character states, which is scored the same in humans and chimpanzees (among others). But whereas in some (even many?) cases such hypotheses may seem trivial (almost being considered as facts), when dealing with, for example, the complex patterns of the circulatory system in arthropods (e.g. Wirkner et al., 2017), or with muscular patterns in arthropod limbs (e.g. Grams et al., in press), identifying "character state identity" can often be a challenge.
The challenge includes the possibility that a "character state identity" hypothesis is wrong (or not further corroborated in the present research cycle; see Richter, 2005) because, as part of a cladistic analysis, identical character states may turn out as not having evolved only once ("secondary homologous" sensu de Pinna, 1991). A character hypothesis also might be wrong if a certain character state turns out to not belong to the same transformation series. In addition, the character state "absent" is (in a strict sense) not homologous to other character states. This is one of the reasons why Hennig mostly avoided the term homology and instead introduced the pair of terms synapomorphy and symplesiomorphy, which explicitly also might refer to the absence of a thing (Hennig 1984).

Synapomorphy and symplesiomorphy
The last decade has seen a revival of the debate on the relationship between homologue and homology, and synapomorphy and symplesiomorphy (Scotland, 2010;de Pinna, 2012, 2014;Williams and Ebach, 2012;Nixon and Carpenter, 2012a, b;Assis, 2013;Farris, 2014). In our view, these terms/concepts represent different perspectives on the same phenomenoncorrespondence by common origin. Homology has no implication on the direction of transformation, whereas symplesiomorphy as "primitive" condition and synapomorphy as "derived" condition refer directly to phylogenesis, the real historical evolutionary process of speciation and transformation (Richter, 2017: 540). As an aside, Williams and Ebach (2012: 223) emphasized the difference to Nixon and Carpenter (2012a, b) in writing: "The key to their views is found in the definition of synapomorphy, which "is one kind of homology". Synapomorphy is better thought of as one or more homologs [sic]a homolog [sic] is a part." This was agreed with by Richter (2017: 540) in a footnote: "Indeed, the terms synapomorphy and symplesiomorphy do not pertain to general statements but refer to a special character state that might be either a synapomorphy or a symplesiomorphy in reference to a certain taxon. Homology refers to a general relationship between structures." That the terms synapomorphy/symplesiomorphy refer to homologues as things (but also to properties of them; see above) and not to relationships might, however, not entirely reflect Willi Hennig's intention, when he introduced the terms, but more importantly it also might include a conceptual contraction.
In a diagram, Hennig (1966: fig. 44) explained that "Similarity can be based on symplesiomorphy, synapomorphy, or convergence", which indicates that he uses the three terms indeed as kinds of relationships which exist between character states. In the diagram itself, however, synapomorphy and symplesiomorphy relate directly to the character states, shown by these states being given in brackets after the term [e.g. "synapomorphy (b')"].
"Synapomorphies" or "synapomorphous character states" are based on "synapomorphy". A certain character state might represent a "synapomorphy" or a "synapomorphous character state" of two or more taxa. The same holds true for symplesiomorphy in all its variants. Whereas synapomorphy/symplesiomorphy always refer to the identical character state, homologues might well refer to different character states (see above). From Hennig's use of the terms, it appears unlikely, however, that he would have been unlikely to have used the term homologues/homologies for complementary absences as suggested by Brower and de Pinna (2012:532) because homologues are things and absences are obviously not. The nonapplicability of homology/homologue to the absence of things was one of the reasons why Hennig introduced the terms synapomorphy/symplesiomorphy (cf. Hennig, 1966: 95; see also Hennig, 1984:38, translated in Richter, 2017. This also shows that, as opposed to homologues, synapomorphies/symplesiomorphies are not necessarily things (see M€ uller, 2003), which is in contrast to Nelson's (2011: 140) view. Synapomorphies are not necessarily "parts of organisms as evidence of relationship" but the same character state in two different taxa based on synapomorphy, which might well be the property of a homologue or even its absence.

Similarity vs. sameness
The terms "similarity" and "sameness" are tied to the concepts described above. Distinguishing these two terms is of crucial importance for their understanding.
In the case of morphology, comparison of (descriptions of) morphemes (see Richter and Wirkner, 2014;G€ opel and Richter, 2016;Richter, 2017) leads to the observation of both intraspecific and interspecific similarity in the features or properties of those morphemes. This "similarity" (or better, structural correspondence because the term "similarity" might imply a more subjective perception; see Riedl, 2000, Richter, 2005 as perception of resemblance still operates on the observational-descriptive level (Rieppel and Kearney, 2002;Rieppel, 2006). Upon the observations of similarity, hypotheses of sameness as historical/evolutionary identity are erected (Rieppel and Kearney, 2002;M€ uller, 2003). The hypotheses of sameness, however, can be of two kinds: (1) Hypotheses of character state identity encompass the notion of degree of difference in the observed morphemes to which the identical character state is assigned (Richter, 2017); (2) Hypotheses of character identity are hypotheses of evolutionary sameness and are assignments of different character states to the same transformation series (Hennig, 1966;Szucsich and Wirkner, 2007;Wirkner and Richter, 2010). The latter refers to the concept of homology sensu Hennig (1966), although other evolutionary concepts of homology do refer to the former or both (e.g. de Pinna, 1991;Szucsich and Wirkner, 2007). Nevertheless, all of these evolutionary, explanatory concepts of homology refer to sameness as historical identity of some sort (Rieppel and Kearney, 2002;Fusco, 2022). The sharp distinction of the descriptive and explanatory level and causal nature of the concept of homology was discussed by Nelson (1994Nelson ( , 2011, who emphasized that homology is not observable evidence but rather a phylogenetic relationship between character states. He did, however, treat synapomorphy as observable evidence (Nelson, 2011), although synapomorphy (as well as symplesiomorphy), as introduced by Hennig (1966), is also an explanatory concept and a special perspective on homology (i.e. sameness) rather than mere similarity of parts of organisms (for a more detailed account, see Richter, 2017).
"Similarity" and "sameness" refer to properties of the same parts of organisms. They do so, however, from different perspectives on different epistemological levels. Similarity exists between morphemes and is a perceptual part of the descriptive level. Sameness refers to character states and, thus, is an inferential, causal account operating on the explanatory level.
In this conceptual framework, similarity (or correspondence) clearly is the most important hint towards sameness, i.e. homology (M€ uller, 2003). However, not only can similarity exist without sameness in cases of convergence, but also sameness can exist with a high degree of dissimilarity (e.g. endostyle and thyroid). This shows that similarity and sameness are more epistemologically than ontologically connected. Although there cannot be wrong notions of similarity owing to its subjective perceptual nature (Riedl, 2000;Richter, 2005), there can, of course, be wrong hypotheses of sameness (i.e. homology).

Conclusion
In conclusion, we argue that both homology and homologue are inseparable explanatory concepts; homologues are things but bear their subsistence as homologues only due to homology as the relationship by which they are connected. It is therefore inconsistent to regard homologues as mere objects of perception because we may see the single objects but we cannot recognize their nature as homologues. Adding nonexplanatory concepts to these terms, in our view, waters down their meaningfulness and conflates the very distinct epistemological levels of description and explanation. It is to account for the mere objects of perception and morphological description that the term morpheme has been introduced (Richter and Wirkner, 2014;G€ opel and Richter, 2016), and its incongruence to the terms homologue and character state has been clarified herein. Incongruence also exists between homologue and character statethe latter being the inheritable trait that is subject to selection and transformationwhich can refer to either the things (homologues) themselves, or properties of these or even their absence (e.g. Wagner, 2014). The terms synapomorphy/symplesiomorphy are comparable to both homology and homologue depending on the context. The difference is semantic but not conceptual. The ontological status of the concepts discussed herein are summarized in Table 1.
Even more than 170 years after the terms homologue and homology were popularized by Owen, meaningful concepts of these terms and their relationship to each other are being discussed among systematists and such discussion still appears worthwhile. Although there is no primacy of the first (oldest) definition, it is indeed necessary to discuss these terms in the context of their historical origin and possible change of meaning to erect a conceptual and semantic framework for 21 st century systematics. We would like to point out the obviousness that this is not a matter of "right or wrong". To us, however, the specification of homology and homologue to the explanatory level, and ultimately the whole conceptual framework outlined in this manuscript, appears useful.