Shyshkin, 1988. Patterns of ontogenetic evolution

M.A. Shyshkyn. Zakonomernosti evolyutsii ontogenezu // Sovremennaâ paleontologija. Metody, napravlennâ, problemy, praktyčeske prymenenâ: Spravočne pôsibâ: V 2-h tomah. / Pid. red. V.V. Mennera, V.P. Makrydyna. – M.: Nedra, 1988. – Tom 2. s. 169-209.
M.A. Shyshkyn
ZAKONOMERNOŚTI E V O L U C I I ONTOGENEZA
OBŠI ZAMEČANNJA

When discussing the morphological aspects of ontogenetic evolution—which are of the utmost importance to a paleontologist— one must first and foremost bear in mind the interconnection between changes in individual development and the evolution of the organism as a whole. This very problem constitutes, for example, the essence of A.N. Severtcov’s entire work on phyloembryogenesis [926, 927, 928]. Although such an approach to the outcome of development may currently appear to be nothing more than a relic of Häckelian thought [163, 540] or, at best, a permissible methodological convention [1661], in practice it is indispensable if we are interested in the actual evolutionary process. Only in an isolated ontogenetic cycle does the adult phase appear least significant, since it is entirely determined by prior development but itself contributes nothing further. In the evolutionary context, the situation is different. The adult phase is the only one in which, under normal conditions (at least in Metazoa), the function of reproduction lies, which creates genetic diversity, i.e., the material of the evolutionary process. The success of adult organisms in the struggle for existence determines which type of gametes will be used to create the next generation of the population and, accordingly, which phenotypic material will be subject to further selection. In the process of selection, unsuccessful phenotypes are eliminated along with their ontogenies. This means that even if certain adaptations are successful during intermediate stages of development, they do not confer an advantage if, as a result of ontogenesis, they fail to ensure the realization of the required phenotype. The adult stage, as it were, dictates its conditions to the entire ontogenesis. Therefore, E. Haeckel’s often-criticized statement that phylogeny (understood by him as the sum of adult phases) is the cause of ontogeny actually has a profound meaning, though not as precise as the author of the biogenetic law intended. That all phases of ontogenesis change is beyond doubt. The question is different—how are their changes interrelated? Does the adult phase evolve independently, or under the influence of simultaneous morphological changes in early development, or through both pathways at once? Finally, is it possible in phylogeny for later changes to have a retroactive influence on the course of earlier phases—that is, one that is not possible within the framework of a single ontogenesis? In answering the first question, a striking contradiction emerges in contemporary views on the role of ontogenesis in evolution, which essentially combine two incompatible concepts. On the one hand, it is acknowledged that normal development is a regulated process, steadily directed toward achieving the final result and striving to eliminate all deviations that arise along this path [1192, 1196, 2439, 2441]. It follows that these variations should not directly influence the form of the adult organism. The evolution of the latter must be associated with formative deviations, which pertain to the end of development, where the possibilities for regulation diminish. At the same time, the prevailing view is that changes in early or intermediate phases may serve as the cause of undiscovered deviations (deviation) in the entire course of ontogenesis, radically altering the form of the adult in the offspring. Usually, this contradiction is not even noticed. Both concepts have long existed and have different origins in classical morphology, but only the first of them, as will be shown, can be reconciled with the requirements of evolutionary theory. The foundation for it was laid by two empirical generalizations by K.M. Ber [1291]. One of them concerns the reduction of embryonic variability in later stages (indicating, according to K.M. Ber, the presence of higher-level control regulating development). The other constitutes the well-known Bera’s law, or the law of embryonic resemblance. Within the framework of evolutionary theory, as Charles Darwin [309] had already pointed out, this phenomenon implies that natural selection tends to alter organisms mainly in later life. These views were supported by developmental mechanics, which experimentally demonstrated the capacity of morphogenetic processes for self-regulation, as a result of which the adult organism turns out to be more stable than the means of its realization (in ontogenesis or during regeneration). This generalization, known as “Rü’s law” [921, 2184], or the principle of equifinality [1524], It confirmed the high conservatism of the early phases of development, showing that the evolution of the adult organism must proceed primarily through changes in late ontogenesis [2185]. But the main source of such views was E. Haeckel’s fundamental biogenetic law [1687], which is a translation into evolutionary terms of the old idea of the parallelism between individual development and the “ladder of existence.” The fulfillment of this law (i.e., a brief repetition of phylogeny in ontogeny) cannot be considered otherwise than on the basis of the final changes in development. Since E. Haeckel was not so much interested in evolution in ontogenesis as in a method for phylogenetic reconstructions, he does not provide any special explanations for the reasons behind such a course of changes, except for the much-discussed “laws” of heredity and adaptation. Thus, the biogenetic law constitutes a complete theory of the evolution of ontogenesis, which consists of two propositions: a) the adult phase evolves through the addition of new terminal phases to ontogenesis, which ensures the recapitulation in ontogenesis of the forms of adult ancestors (palingeny); b) intermediate phases undergo their own adaptive evolution, which distorts the ontogenetic record of the adult phase’s transformations (cenogenesis). The biogenetic law is usually associated with the mechanistic-Lamarckian views of E. Haeckel [359, 565, 1196], while asserting that he was opposed to the discovery of Mendelian factors and the mutational process [1661, 1892]. But in reality, the question of the nature of heredity does not play a decisive role in the evaluation of the law [1822]. In fact, regardless of whether the emergence of evolutionary novelties is linked to late somatic changes transmitted in some way through germ cells (the Lamarckian concept) or through direct changes in those cells themselves (the concept of hereditary factors), in any case a new cycle of development must begin with the transformation of the zygote, i.e., distinguish itself from the parental form from the outset. E. Haeckel clearly understood this condition and in no way saw it as an obstacle to the biogenetic law, pointing out that changes, which arise at a specific moment of development and are transmitted via parental gametes, begin in the offspring at the very same stage (“the law of simultaneous inheritance” [1687]). In this, he directly follows Charles Darwin [310], who formulated the principle of “heredity at the corresponding age” and emphasizing that the appearance of a hereditary deviation in gametes and its visible manifestation in the individual development of offspring are two different things. Therefore, it is not surprising that “the recognition of determinants or genes as carriers of heredity did not prevent many researchers from acknowledging the biogenetic law [1380, 1692, 2472] or, at the very least, the connection between evolutionary changes and later stages [503, 709]. The opposing view of the evolution of ontogenesis, allowing for changes in adult organisms through deviations in the early stages of development, originates with E. Geoffroy Saint-Hilaire [2198] and F. Müller [719]. In their work and that of contemporary authors, this view most often shares a common limitation—the comparison of normal ontogeneses (or the morphogenesis of homologous organs) and the identification of the observed final ontogenetic difference (Bera’s embryonic divergence) with the actual evolutionary process. For example, if the morphogeneses of homologous organs A and B in two different forms coincide up to phase x and then diverge, it is claimed that organ B arose from A after a divergence at phase x [2141]. We will demonstrate the methodological inconsistency of this argument. Views on the possibility or necessity of such a path of evolution never formed, unlike Häckel’s, a comprehensive theory. Nevertheless, by the beginning of the 20th century, they had become predominant. One of the reasons for this was the accumulation of facts from comparative embryology that, according to the researchers, did not agree with the biogenetic law [At the same time, it was overlooked that the law’s incomplete fulfillment was already anticipated in its premises regarding cenogeny and the abbreviation of repeated phylogeny. Thus, requirements were imposed on the law that do not follow from its content]. The recognition of evolution through earlier deviations in development eliminated the thesis regarding the primary necessity of recapitulation, though at the cost of its essence becoming an enigma [565, 1869]. Another point of criticism of Häckel’s views was linked to the growing attention of embryologists, that a true explanation of the course of individual development must be sought in the investigation of its immediate causal factors, rather than in the construction of abstract historical principles that replace empirical knowledge. This approach led to the emergence of experimental (causal) embryology, which revealed the distinct diversity and specificity of the structure of embryonic cells in different organisms. The latter clearly demonstrated that the zygote is as much a product of evolution as the adult organism, and that the initial stage of development of a higher organism is not a repetition of its unicellular ancestor. “A chicken egg corresponds no more to the initial link in the phylogenetic chain than the chicken itself” [1733]. All of this leads to the conclusion that evolution proceeds through a change in the entire ontogenetic cycle from bottom to top in each successive generation, rather than through the addition of final stages [1448, 1763, 2259] [Centuries earlier, the same arguments were used by proponents of the idea of preformed development in their critique of the theory of parallelism. It was pointed out that the development of an organism cannot pass through the phases of lower classes, since it differs from them in its purpose already in the embryonic cell [540]]. Ontogeny, therefore, constitutes phylogeny rather than repeating it [1623]. At the same time, some authors had in mind only gradual transformations, while others—the possibility of abrupt deviations, abruptly changing the adult form. Within the framework of these views, however, any theoretical possibility of parallels between ontogeny and phylogeny, inconsistent with the obvious facts of their existence, disappeared. This objection was circumvented by the argument that such parallels have purely morphogenetic, rather than historical, causes, i.e., that the ancestral adult phase may persist in the ontogeny of the descendant only because it constitutes an indispensable basis for the realization of subsequent phases [1623]. Another explanation lies in the fact that most such examples, which invoke the biogenetic law, in reality do not refer to the repetition of the form of adult lower forms in the ontogeny of higher ones, but simply to the preservation of common phases of development in both, i.e., to the manifestation of embryonic resemblance, as K.M. Ber [1623, 1767, 1970, and others] has already pointed out. This argument, repeatedly cited later as well, for example [1317, 1319], is usually regarded as a key and controversial supporting model of Haeckel’s evolution. But, strangely enough, they fail to notice that in the reconstruction of ontogenesis, the retention of the embryonic form from the initial phase becomes just as implausible as the appearance of recapitulation! In general, the position of later researchers regarding the biogenetic law appears more moderate and allows for the possibility of various pathways of ontogenesis. Accordingly, various classifications of these pathways, or modes, emerge [565, 926, 928, 1317, 1319, 1604, 2141, 2124], in which the Heckelian mode of evolution through late appendages occupies one place or another—from relatively significant, for example in A.N. Severtcov, to insignificant in G. de Bira [1317]. Why ontogenesis varies in different cases, these views do not explain (unless one considers the common assertion that earlier changes are the path to the formation of large systematic groups). A.N. Severtcova [927] directly states that answering this question is not at all part of her task, which consists solely in determining how evolution can proceed. She acknowledges that, in her view, phyloembryogenesis does not depend on the adoption of any particular evolutionary theory. The very diversity of the identified modes often leads their authors to conclude that all logically possible paths of individual development are in fact realized in evolution [2141]. This is tantamount to acknowledging that there are no common regularities in the transformed ontogenesis. Such a disappointing conclusion to a century of research on the problem following E. Haeckel is, in general, not surprising given the circumstances, since it was based on a comparison of only normal ontogenies, without any connection to the analysis of the mechanisms that realize the material of evolution—ontogenetic variability. The next piece of evidence in this regard is S. Gulda’s work [1661] on the relationship between ontogeny and phylogeny, which is the most extensively studied on this topic over the past decades. The methodology here is the same—identifying evolutionary modes and assessing how strongly they reflect the course of phylogeny. An attempt to combine them with two types of adaptive strategies (r- and K-selection) does not change the traditional view of the problem. The evolution of ontogeny remains the sum of isolated processes, each of which proceeds under its own specific conditions. The question of their common origins over time remains unaddressed. ONTOGENESIS AND STABILIZING SELECTION The morphological evolution of ontogenesis cannot be understood based solely on its adaptive outcomes, i.e., by simply combining “ready-made” types of normal development. At its core lies the process of evolution of formative mechanisms, which transforms the typical adult organization (the adaptive norm [1184]), and along with it the standard path of its realization in ontogenesis. Every elementary change in the adult norm is associated with the selection of one of its deviations, i.e., with the preservation of certain individual developmental variants that realize a given advantageous deviation. This selection of variants of individual cycles, based on the principle of the unambiguity of their outcome, must necessarily reshape the typical course of development as a whole (see “Evolution as an Epigenetic Process”), and the main task of evolutionary theory is to elucidate the laws of this morphogenetic restructuring. The morphological evolution of ontogenesis is merely its external expression. The basis for solving this problem is found in the theory of stabilizing selection, or the epigenetic concept of evolution (see “Evolution as an Epigenetic Process”), according to which the emergence of elementary adaptive changes is expressed in the increased stability of the ontogenetic realization of initial variation. In light of these views, the simplicity of typical development, as revealed by experimental embryology, is viewed as the result of selection for maximum protection of normal morphogenesis. The mechanisms of this protection (self-regulation) are most clearly evident in “regulatory” ontogenies (spinal, echinoderm, etc.), where development in the early phases proceeds through the induction of interactions between germ layers, which determines the direction of their further differentiation. A stable course of development is supported by a broad expansion of the thresholds of normal response of the buds across the most diverse parameters: the quantity and activity of metabolites, the duration of competence periods (the ability to interact normally), and so on. Therefore, minor fluctuations in these indicators, caused by genetic or external disturbances, are buffered within sufficiently wide limits and do not affect the further course of development. Strong influences, irreversibly disrupting development, lead either to death or to definitive anomalies that reduce viability. Such stability is characteristic of "mosaic" ontogenies (mollusks, arthropods, etc.), where organogenesis occurs in small groups of cells, and regulation apparently occurs within them. The blastomeres here are, in principle, just as multipotent as in “regulatory” ontogenies (e.g., the blastomeres of a mosquito’s eye, which can develop into an antenna or a limb), but normal determination is so reliably established that at first glance it appears to be the only possible outcome. There is no clear boundary between the two types of development in other respects as well: “mosaic” phases occur in “regulatory” ontogenies and vice versa. In general, thanks to self-regulation, normal ontogenesis is a channeled sequence of events (credo, or lit. “necessary path” [2441]), which seeks to overcome all obstacles on the path to the realization of a standard organization. The problem under consideration—the relationship between the evolution of the adult organism and its ontogenesis—must first be framed differently – how is the formation of new adult characteristics related to the restructuring of their formative mechanisms? According to the epigenetic concept [1178, 1180, 1181, 1192, 1196], in general terms, this relationship boils down to the following. An elementary evolutionary change arises through selection as a modification of the previous phenotypic norm (see Figs. 26, 27, see “Evolution as an epigenetic process”), realized only under those conditions against which it has an adaptive advantage. This is primitive (dependent) form-formation, in which the environment acts directly or indirectly (through functions) as a determining factor of development.