Morphological variation during post-embryonic development in the centipede Lithobius melanops: traditional and geometric morphometrics approaches

Two modes of post-embryonic development, hemianamorphosis and epimorphosis, show a distinct pattern among extant centipede (Chilopoda) orders. Although studies on post-embryonic development in Lithobiomorpha date back to the 19th and the 20th century, many ambiguities referring to nomenclature of their post-embryonic stages still exist. In this paper coherent terminology of the post-embryonic stages in Lithobius melanops, which could be applied to other lithobiomorphs, is proposed. Additionally, morphological variability of selected morphological traits was analyzed using traditional and geometric morphometric approaches. We recognized five anamorphic (anamorph 0 to 4) and five epimorphic stages (agenitalis, immaturus, praematurus, pseudomaturus and maturus). Measurement and count of certain morphometric characters, detailed description of genital appendages, shape and size variation of the forcipular apparatus, the cephalic capsule and the terminal legs are given. Moreover, for the purpose of geometric morphometric analyses we separated specimens of epimorphic stages into three groups (agenitalis-, praematurusand maturus-like) based on the level of differentiation of genital appendages. Sexual size dimorphism of the forcipular apparatus was observed only in the praematurus group. Also, significant inter-group forcipular size and shape differences are found between some tested groups. Furthermore, significant differences in size and shape were recorded for the cephalic capsule between all groups. Finally, significant size differences in ultimate legs are present between all epimorphic groups, while shape differences were detected only between agenitalis and maturus groups. Our results contribute to overcoming terminological disparities and provide guidelines for distinguishing stages via discrete and continuous changes during post-embryonic development of the anamorphic centipede.


Introduction
As in most arthropods, post-embryonic development of centipedes (Chilopoda) is organized into various stages, separated by moults. Two different modes of the post-embryonic development of these arthropods can be described through two different processes of moult-dependent development hemianamorphosis and epimorphosis (Haase, 1880; but see also e.g., Minelli & Sombke, 2011;Fusco & Minelli, 2021). The epimorphic centipedes are characterized by a full adult number of the body segments almost immediately after hatching, or more precisely, during the embryoid phase (Vedel & Arthur, 2009;Brena, 2014;Stojanović et al., 2015Stojanović et al., , 2020a. On the other hand, in the anamorphic centipedes the first several stages after hatching have incomplete segment number and achieve the final (adult) number by addition of segments through a series of moults (Minelli & Sombke, 2011). Precisely, in the anamorphic centipedes post-embyonic development can be specified as hemianamorphosis, where the individuals, after reaching the adult number of segments, retain a constant number of segments during the following moults, but body growth and features (e.g., secondary sexual characteristics) are becoming more pronounced (Verhoeff, 1905; see also Minelli & Fusco, 2013).
Both of the abovementioned developmental patterns have important roles in understanding the phylogenetic relationships post-embryonic development in centipede lithobius melanops Downloaded from Brill.com09/10/2023 06:57:39PM via free access within the class Chilopoda. For example, Haase (1880Haase ( , 1881 recognized chilopod post-embryonic modes as taxonomic features, and divided all centipedes into two subclasses, viz. Anamorpha (orders Scutigeromorpha, Craterostigmomorpha and Lithobiomorpha) and Epimorpha (orders Scolopendromorpha and Geophilomorpha). Moreover, a comparative evolutionary study by Edgecombe and Giribet (2007) showed that anamorphosis is a plesiomorphic characteristic which is shared between all mentioned anamorphic orders, while the monophyletic nature of Epimorpha was confirmed in subsequent studies (Fernández et al., 2016(Fernández et al., , 2018. There are two clearly distinguished phases during post-embryogenesis in the anamorphic centipedes: the "larval" phase (anamorphic phase) which is characterized by incomplete adult segment number, posterior segments and their appendages, and "post-larval" (epimorphic) phase with an adult-like appearance, clearly separated postpedal segments and final number of leg-bearing segments (e.g., Verhoeff, 1902Verhoeff, -25, 1905Andersson, 1979;Minelli & Sombke, 2011). Both phases consist of a different number of stages. In Lithobiomorpha, as the largest and the most diversified group of hemianamorphic centipedes, the terminology referring to names of different stages dichotomy between authors of the 19th and first half of the 20th century vs. authors of the last decades of the 20th century (and after). Indeed, there are several different approaches in the use of these terms (table 1). Adapted and widely accepted, Verhoeff's (1902-25) version of stage division implies in total five anamorphic and five epimorphic stages. The results of the following studies also suggested division of the anamorphic phase and increase in the number of the epimorphic stages (e.g., Brolemann, 1930;Murakami, 1958Murakami, , 1960bMurakami, , c, 1961aMurakami, , 1963. Additional divisions of Verhoeff's epimorphic stages implied that those stages are grouped together; based on the criteria which take into account a development of the genital apparatus rather than all morphological changes appearing during the duration of a certain post-embryonic stage (Verhoeff, 1902-25).
Although the stages that appear during the anamorphic phase of lithobiomorph post-embryogenesis can be clearly distinguished, because external morphological characters are easily recognized, the opposite pattern is present among epimorphic stages. This is linked with the remarkable variability of morphological characters within and between these stages (Stojanović et al., 2020b). To consider this intra-/inter-stage variability of the "post-larval" phase, it is necessary to use some special methods or techniques, such as statistical processing of the meristic and/or morphometric data. Morphometric approaches have already been used to analyze the morphological variability in centipedes, but such studies are still generally rare (Simaiakis et al., 2011;Lopez Gutierrez et al., 2011;Dugon et al., 2012;Siriwut et al., 2015;Baiocco et al., 2017;Zarei & Seifali, 2020;Peretti et al., 2022;Vujić et al., 2022).

Materials and methods
The specimens of L. melanops analyzed in this study were collected during a few consecutive days during the beginning of August 2019 by D. Stojanović  For species and stage identifications, meristic and metric data were collected from images taken with a Nikon smz 1270 binocular stereo microscope with an attached Nikon ds-Fi2 camera and Nikon ds-L3 camera controller, as well as a Carl Zeiss Stemi 2000-C binocular stereomicroscope with an AxioCam MRc camera and integrated Axio Vs40 software packages. The final images were processed using the Zerene Stacker software (for focal stacking) and Adobe Photoshop cs6.
To make it easier to see individual parts of the forcipular apparatus, such as the coxosternal teeth, porodonts, coxosternal median diastema and shoulder of forcipular coxosternite in the youngest stages (anamorph 0 and 1), the selected specimens were colored with toluidine blue before imaging (see fig. 3B and D). To analyze body length, we used number and arrangement of ocelli and coxal pores of 52 specimens in anamorphic stages and 179 specimens in epimorphic stages. However, antennal length and number of antennal  articles were analyzed in 49 specimens in anamorphic stages and 175 specimens in epimorphic stages, because antennae were damaged in some specimens. Also, the number and arrangement of ocelli and coxal pores were analyzed on both sides of the body in all specimens, while antennal length and number of antennal articles were analyzed mostly on both sides of the body, but also on one of the sides in damaged specimens. Only one antenna per specimen was analyzed in one specimen from A1 and A3, six specimens from A2 and ag, three specimens from A4, ps and M, and eight specimens from im. The representation of the arrangement of the ocelli and the coxal pores is harmonized with Bonato et al. (2010). Additionally, to investigate shape and size variation of the forcipular apparatus, the cephalic capsule and the terminal legs, separately, among three groups (Age, Praem and Mat), each of the structures was photographed with a reference scale using a Nikon smz 1270 binocular stereo microscope. Further, a certain number of landmarks and semilandmarks were positioned on the forcipular apparatus (30 landmarks), the cephalic capsule (2 landmarks and 12 semilandmarks) and the ultimate legs (25 landmarks) ( fig. 7). Position of

Developmental stages
In the present study, all of 231 collected specimens of L. melanops were separated by ontogenetic stages, and after that, photographed and measured. In total, five stages with different numbers of collected specimens were recognized per both post-embryonic developmental phases, namely: A0 (2 examples

Morphological character changes through the post-embryonic development
Variation of meristic and morphometric characters Variation of the forcipular apparatus, arrangement of ocelli and development of the genital appendages on the postpedal segments in females and males during post-embryonic stages are well documented by figs. 3-6, respectively. Also, descriptive statistics of the main morphological features (body and intact antennae length, number of antennal articles, ocelli, coxal pores and pair legs) at anamorphic and epimorphic stages are presented in tables 2 and 3, respectively.

Discussion
The post-embryonic development in L. melanops is described in detail on the aspect of morphometry and morphology of stages in the present study. The results of this study provided information on distinguishing the ontogenetic stages, meristic variability of the morphological structures, as well as size and shape variation in the selected structures at intra-and inter-epimorphic stages. Specifically, the five different stages have been characterized for both anamorphic (A0-A4) and epimorphic (agenitalis, immaturus, praematurus, pseudomaturus and maturus) developmental phases based on morphological figure 9 Centroid size differences of (A) The forcipular apparatus; (B) The cephalic capsule; and (C) The ultimate leg among epimorphic groups. The median with the first and third quartiles is shown (in boxes), together with the range of variation and outliers.
variation of the analyzed morphological structures. Also, sexual dimorphism was detected only in the forcipular apparatus in praematurus, where males have higher values of the forcipules cs than females. Centroid size was significantly different among all analyzed groups in the case of the cephalic capsule and the ultimate legs, whilst forcipular apparatus cs was significantly different between Age and Mat and between females Praem and Mat. Furthermore, our results revealed that the anterior part of the cephalic capsule is shorter and less protruded at Mat than in specimens at Age and Praem, whilst the forcipular coxosternite and the forcipular coxae are broader, and both the forcipular ungulum and the ultimate legs are more convex, in Mat than at Age. To the best of our knowledge, the present study represents the first attempt to explain post-embryonic stages as well as shape variation of the selected morphological structures at both intra-and inter-epimorphic stages using traditional and geometric morphometric approaches in Lithobiomorpha.

Developmental stages in Lithobiomorpha
Current knowledge of post-embryogenesis in the lithobimorphs is provided by numerous studies in Lithobius species, but also species from other genera (table 1; but see also Brocher, 1930;Roberts, 1956;Scheffel, 1961Scheffel, , 1969Eason, 1989;Lewis, 1965;Joly, 1966;Demange, 1967;Tobias, 1969;Andersson, 1978a;Wignarajah & Phillipson, 1977;Albert, 1983;Barber & Eason, 1986;Zulka, 1991;Zapparoli, 1998;Voigtländer, 2000). Broadly speaking, the total number of recorded post-embryonic stages varies in different lithobiomorph species (e.g., Andersson, 1979;Voigtländer, 2007). In the majority of the species, five anamorphic and a various number of epimorphic stages were reported (e.g. Andersson, 1979; for details, please see table 1). The oldest attempt to define the lithobiomorph post-embryonic stages began more than a century and a half ago, when Fabre (1855) recognized five post-embryonic stages in Lithobius forficatus (Linnaeus, 1758), based on a combination of the external morphological features (number of segments, leg pairs and ocelli), and named simply as "stadium" 1 to 5. At the beginning of the 20th century, Verhoeff (1902-25) reviewed terminology in detail, taking into account that Meinert (1872) summarized the first four of Fabre's stadiums under the term "pullus", and provided evidence that "stadium 5" is not only an adult stage, but composed of the three stages, named "juvenis", "junior" and "adultus". Further changes were proposed at about the same time by Haase (1880) and Latzel (1880). The views of both researchers were essentially the same, especially concerning Fabre's stadium 5. In particular, Haase divides the Meinert's pullus stage into two separate stages, viz. pullus (includes stadium 1 and 2), and "puer" stage (includes stadium 3 and 4), while Latzel accepted pullus without any changes. Also, Haase accepted the names of Meinert's terms ("juvenis", "junior" and "adultus") for the divided stadium 5, but Latzel adopted the terms of the last of Fabre's stages as "immaturus", "juvenis" and "maturus", respectively (with the note that Latzel's and Haase's "juvenis" stage actually does not indicate the same age category) (table 1).
After mutual comparison of all these approaches, Verhoeff (1902-25) pointed out that Meinert's, Haase's and Latzel's descriptions of lithobiomorph development had weaknesses, because they did not define stages which have fifteen pairs of legs. This fact and the discovery of three to five new stages of development prompted Verhoeff (1905) to distinguish hemianamorphosis from the true anamorphosis. Efforts to unify terminology and define the number of stages resulted in a proposal to divide development into 11 stages, with a detailed morphological description of each of them (Verhoeff, 1902-25). In the anamorphic phase, six stages (föetus, larvae prima, -secunda, -tertia, -quarta, and -media), and five epimorphic stages (agenitalis, immaturus, praematurus, pseudomaturus, and maturus) have been recognized. Verhoeff claims that, in general, the stage larvae media does not occur in lithobiids, but only in some species. However, some authors accepted this stage as a transitional stage, as a last occasional "larval" (Joly, 1966) or as the first "post-larval" (Scheffel, 1961(Scheffel, , 1969, but most of them did not recognize this stage (e.g., Eason, 1964;Andersson, 1976Andersson, , 1978aAndersson, , 1979Andersson, , 1980Andersson, , 1981aAndersson, , b, 1983Andersson, , 1990Barber, 2009) (table 1). Joly (1966) observed that the larva media stadium only represents 3-4% of his individuals and Andersson (1976) observed it only one time in Lithobius forficatus. It can be considered as simply an accidental slightly more developed larval stadium (Andersson, 1976).
The first acceptance of Verhoeff's terminology was provided by Chamberlin (1913Chamberlin ( , 1916Chamberlin ( , 1917Chamberlin ( , 1925a and Brolemann (1930). However, Brolemann (1930) made certain changes, such as adjusting the names of the anamorphic phase stages (foetus and larvae i-iv) and adding additional stages into the epimorphic phase (viz. unique stage pseudomaturus in i and ii, as well as stage maturus in i and ii). Thus, the number of recorded post-embryonic stages in lithobiids was thirteen (six "larval" and seven "post-larval" stages). Following studies on post-embryonic development in lithobiids mostly accepted these terminological recommendations without any or with only some minor adjustments (e.g., Murakami, 1958Murakami, , 1960bEason, 1964). The main difference was reflected in increasing the number of epimorphic stages and distinguishing of the later stages, viz. praematurus, pseudomaturus, and maturus (Murakami, 1958(Murakami, , 1960b(Murakami, , c, 1961a(Murakami, , 1963. It is worth pointing out that Verhoeff's division of the post-embryonic stages in lithobiids can be denoted as "traditional", because of its widespread acceptance until the end of the 1970s (table 1).
In a series of studies, Andersson (1976Andersson ( , 1978aAndersson ( , b, 1979Andersson ( , 1980Andersson ( , 1981aAndersson ( , b, 1982aAndersson ( , b, 1983Andersson ( , 1984aAndersson ( , b, 1990) analyzed post-embryogenesis of Swedish lithobiids (including Lithobius melanops), and generally accepted Verhoeff's division with several accumulated changes. The five "larval" stages were recognized in all studied species. Terminological changes occurred only in the youngest stage, so Verhoff's foetus was renamed larva 0, while the remaining stages kept their names (larva i to larva iv). Considering that Andersson (1979Andersson ( , 1980Andersson ( , 1981aAndersson ( , b, 1982aAndersson ( , b, 1983Andersson ( , 1984a recognized five to nine stages in the epimorphic phase, and observed many problems in distinguishing stages, Verhoeff's terminology for those stages was rejected and stages were named simply as postlarval 1 (PL1), postlarval 2 (PL2), etc. Although most of the authors accepted his terminological suggestions for epimorphic stages (e.g., Albert, 1982Albert, , 1983Barber & Eason, 1986;Daas et al., 1996;Serra & Miquel, 1996;Kos, 1997;Mitić & Tomić, 2008), there were also studies with traditional terms despite the problems in distinguishing certain stages (e.g., Matic & Stentzer, 1978;Lewis, 1981;Eason, 1989). Furthermore, Eason & Ashmole (1992) proposed usage of the terms antepenultimate and penultimate stadiums for young specimens between agenitalis and pseudomaturus stages to solve the problem of morphological variability of the traditional praematurus stage (table 1), but these terms were not accepted by the myriapodologists. Finally, Voigtländer (2000Voigtländer ( , 2007, encouraged by assertions of Pflugfelder (1932) and Dohle (1970) that the earliest post-embryonic stages of centipedes are not true larvae (based on the absence of distinct characters compared with adults and life under the same environment as the adult) proposed a practical change by marking anamorphic post-embryonic stages with Roman numbers (I-V), but epimorphic stages are represented by combination of Roman and Arabic numbers (vi(1), vii (2) etc.) (table 1). However, each of the proposed terminological divisions of post-embryonic stages has its advantages and disadvantages.
In the present study, the new term anamorph is proposed for the naming of anamorphic stages that replace the previously used term larva, without any changes in criteria for determination of them. On the other hand, the traditional terminology has been accepted for nomenclature of epimorphic stages and we propose the usage of this terminology in the future. Moreover, the adoption of traditional names is practical in epimorphic stages, because these terms are more informative, i.e., they indicate the main morphological features in specimens at a certain stage. For instance, the name of pseudomaturus provides more information about characteristics of the stage itself, viz. although reproductively still immature, the specimens have an adult appearance, a certain degree of development post-embryonic development in centipede lithobius melanops of genital appendages and the possibility of sex identification. In contrast, Andersson's name of postlarval 3 stage, for example, has not provided the information on the main morphological characteristics of their developmental stage. Also, Voigtländer's marking scheme using Roman and Arabic numerals for epimorphic stages (e.g., "X(5)"or "xiv(9)") has not provided enough data on whether the specimens' stage is an adult or juvenile. Indeed, delimitation of the stages defined by each moults (Voigtländer, 2007) is the most precise manner to investigate them. However, distinguishing of the epimorphic stages in Lithobiomorpha is possible only in specimens which have been reared in laboratory conditions. Therefore, the Verhoeff terminology represents the most appropriate manner to approximately group the specimens from natural population in suitable stages. It is worth mentioning that merging the two groups of epimorphic adjacent stages in this study was done to reduce the possibility of mistake in geometric morphometric analyses. Namely, in many cases, especially for the male specimens, it is almost impossible to determine objectively the correct stage, based on the degree of development of the genital structures, and/ or identify whether specimens are reproductively capable or not. It is also important to note that specimens are not completely morphologically identical at the beginning and at the end of one stage (e.g., compare figs. 5B, C and D for females, or 6B and C for males) (see also Stojanović et al., 2020b).

Morphological variation during postembryonic development in L. melanops
Morphological variation during post-embryonic development has been previously studied in L. melanops by Andersson (1979Andersson ( , 1980. Besides various meristic morphological characters (number of coxal pores from 12th to 15th legs, the teeth on the forcipular coxosternite, the ocelli, the antennal articles, projections on tergites 9, 11 and 13, spinulation on the last pair of legs, accessory apical claws on the 15th pair of legs, postpedal segments and the pattern on the cephalic capsule), Andersson (1980) also analyzed metric morphological characters (body-length, head-length, ratios of head-length/body-length and head-length/ head-width) using a traditional morphometric approach. Bearing in mind that the focus of the present study was not entirely the same as in Andersson's studies (1979Andersson's studies ( , 1980, i.e., developmental changes after each moult were not investigated and distinguishing of the stages was not the same, we could not compare each finding of the analyzed morphological traits with those in Andersson's studies. Actually, Andersson (1980) pointed out that no more than one coxal pore added for each coxa after each moults, which can be seen opposite with our results. Still, observed morphological variation at the epimorphic stage in our study is the consequences of approximately grouping of the specimens independent from moults. Our results are partially in line with Andersson's (1979Andersson's ( , 1980 findings on teeth on the forcipular coxosternite, the ocelli, and the antennal articles. Indeed, the pattern of the forcipular coxosternite teeth is the same, viz. the pattern 2 + 2 was observed from the A1 stage (L1 by Andersson) to the final stage, whilst no teeth were detected in A0 (L0 by Andersson) (fig. 3). Also, the ocelli developmental pattern observed in the present study is similar to Anderssons's results, i.e. variation was detected from A3 (L 3 by Andersson) stage with linear progression during ontogeny ( fig.  4). Contrasting that, variation of the antennal articles observed in our study is not in line with Andersson's findings (1979Andersson's findings ( , 1980. The observed variation of antennal articles in the first four anamorphic stages is presented in table 2, while Andersson found that variation in this morphological structure was lacking in stojanović et al these stages. Also, intra-and inter-individual variation of the antennal articles was recorded in the present study (table 2). Furthermore, Andersson (1980) investigated sexual dimorphism in these abovementioned morphological characters, but our study represents the first attempt to analyze sexual dimorphism of the morphological characters in any lithobiomorph species using a geometric morphometric approach. Previously, sexual dimorphism and allometry (at an intra-specific level), shape differences (inter-specific level), and size and shape covariation between structures of the forcipular apparatus (prey-catching apparatus) have been investigated only in five geophilomorph species using geometric morphometrics (Baiocco et al., 2017). Broadly speaking, the forcipular apparatus was the most-used morphological structure in different centipede species when a geometric morphometric approach was applied. Thus, the forcipular apparatus was used to reveal evolutionary changes across different centipede orders (Dugon et al., 2012), and to explain intra-specific variation at the inter-population level in Clinopodes flavidus C. L. Koch, 1847 (Zarei & Seifali, 2020). The forcipular coxosternite, as part of the forcipular apparatus, was used to determine inter-population shape differences in the centipedes Clinopodes carinthiacus (Latzel, 1880) (Peretti et al., 2022) and Scolopendra cingulata Latreille, 1829(Simaiakis et al., 2011, and to distinguish species in the genus Scolopendra Linnaeus, 1758 (Siriwut et al., 2015). Additionally, forcipular coxae shape has been investigated as a possible reliable taxonomic character in Scutigeromorpha (Lopez Gutierrez et al., 2011). Until now, only Vujić et al. (2022) investigated forcipular apparatus shape variation in Lithobiomorpha (species L. melanops) using geometric morphometrics, but the focus of this research was different in comparison with our study. Results of the present study indicate that males at Praem have significantly higher values of the forcipule cs than females at the same stage ( fig. 8A). Further, forcipular apparatus cs was significantly different between Age and Mat and between females Praem and Mat ( fig. 9A). We presume that this is the consequence of the faster developmental rate of the forcipular apparatus in males compared with females during the first part of the epimorphic phase, and of a decrease in the forcipular apparatus developmental rate in males than in females during the second part of the epimorphic phase. Females and males may have different life history strategies, i.e., do not equally invest in viability (e.g. feeding and defence) and reproduction at the same developmental stage. Our results suggest that males invest more resources in viability than females at Praem, while the opposite pattern was detected in the Mat group. Supporting this claim requires additional research with an evo-devo approach. Beside the variation in size of the analyzed morphological characters, our results indicate that forcipular apparatus shape significantly differs between Age and Mat ( fig. 10A). Namely, the forcipular coxosternite and the forcipular coxae are broader and the forcipular ungulum is more convex in Mat than at Age, which is probably correlated with the effectiveness of forcipules in catching bigger prey at the Mat stage.
Unlike the forcipular apparatus, variation in cephalic capsule and tergite of the ultimate leg-bearing segment shape using geometric morphometric approach have been poorly investigated so far. There are only two studies in which shape variation of the abovementioned morphological characters were explained in several species of the genus Scolopendra (Simaiakis et al., 2011;Siriwut et al., 2015). The cephalic capsule and the ultimate legs centroid size were significantly different among all analyzed stages in L. melanops, which indicates that developmental rates of the cephalic post-embryonic development in centipede lithobius melanops Downloaded from Brill.com09/10/2023 06:57:39PM via free access capsule and the ultimate legs are constant and different in comparison with the forcipular apparatus developmental rate. Specifically, the anterior part of the cephalic capsule was shorter and protruded less at Mat than in specimens at Age and Praem, while the ultimate legs were more convex in Mat than Age. The ultimate legs are always held in parallel or oblique to the substrate, i.e., prefemur and femur are directed upward, while tibiae and tarsi are positioned in parallel or oblique to the substrate (Kenning et al., 2017). In general, the ultimate legs do not have a significant role in locomotion propulsion (solely touch the ground in a "probing" manner), but may have a significant impact on stabilization of the body while running (Manton, 1977;Kenning et al., 2017). Also, the significance that the ultimate legs have as an anchorage during the capture of prey was demonstrated in several scolopendromorph species, which are able to use their ultimate legs to better fasten themselves and more effectively seize prey with their remaining legs (Kronmüller & Lewis, 2015;Kenning et al., 2017). Also, it is worth mentioning that in some scutigeromorphs, besides the ultimate legs, which are used as an anchor, the walking legs also have an important function during the capture of prey (Haacke, 1885(Haacke, , 1886Kenning et al., 2017). The recent study of Roithmair et al. (2023) suggested a likely role of the ultimate legs in lithobiomorphs during defence, courtship and mating. Bearing in mind these functions of the ultimate legs, we can assume that the more convex shape of these structures in Mat significantly increases adaptive values (both viability and reproduction). Namely, the specimens have a greater chance of escaping from predators due to greater stabilization of the body while running, more effective prey capture due to better fastening on the substrate of the ultimate legs, and higher reproduction success.
Compared with previous studies, our research represents a step forward in methodology used for analyzing morphological variation of lithobiomorphs during post-embryonic development. The proposed terminology could help in overcoming terminological disparities and could be applied to all Lithobiomorpha as a standard. Hence, our paper will provide open opportunities for further research in the field of centipede post-embryonic development.