Chapter 10 Chapter 5 Classification by Common Ancestry-2

At present, there are many methods, including the study of fossil data, which can be used to judge polarity (Polarity), that is, to judge which of the characteristic sequences is ancestral and which is derivative.But to determine ancestral status with certainty and certainty is at present mostly impossible. The results of the cladogram analysis are recorded in the cladogram.A cladogram consists of a series of binary branches describing successive divisions of a phyletic line.There are two assumptions in the compositional sequence diagram that are completely arbitrary.The first assumes that every existing species is eliminated when a new species appears, and the other assumes that each split is a binary branch.Recognizing that most speciation occurs in isolated, small founder populations, it is clear that such speciation would have no effect on the genetics and morphology of the parent species, which Will remain essentially unchanged for millions of years with new offspring segregating from time to time.

Strictly binary branches are also an unrealistic assumption.After a large taxonomic unit reaches a grade-like stage, it may produce several specialized daughter lines. Although they are sister groups in taxonomy, they can develop in their own ways. , and have nothing in common except that derived from the same parental taxon.Some cladograms constructed recently have recognized this and have changed some binary branches to polytomy (Ashlock 1981).For all the above reasons, Hull (1978) has emphasized that although many cladists claim that their methods are completely objective and non-arbitrary, they are not actually based on fact.It is important to keep this in mind in relation to cladists' critique of the shortcomings of competing taxonomic approaches.

The difficulties encountered by cladistic analysis are exactly the same as those encountered by traditional taxonomists, but this is not the main reason against cladologists.The main reason is the relationship between cladistic analysis and cladistic classification.As far as the cladologist is concerned, once the cladion analysis is over, his task is completed.The reconstructed genealogy, represented by the cladogram, provides classification directly.Cladisms accurately reflect branching patterns and thus provide direct insight into the phylogeny of a taxon.If only information about the order of phylogenetic branch points is to be retrieved from taxonomy, then cladites are the answer.If more is required, the history of a group reflected in the taxonomy is sought.

An approach that does not completely ignore evdutionary divergence and autapomorPh characters (characters unique to one of the sister groups).Cladists disagree with the view that Darwin's genealogy was not itself a taxonomy.Cladists divide taxonomic units not on the basis of similarity but on the principle of full (line) lineage, that is, all descendants of a common ancestor are united into a single taxonomic unit.The result is an incongruous combination of crocodiles and birds, orangutans and humans as joint taxa.Classification is entirely by common derivations, even when unique derivations greatly outnumber common derivations, as in the case of birds from reptiles.

In other words, the cladistic approach ignores that phylogeny consists of two parts: the splitting of evolutionary lines; and the subsequent evolutionary changes of the splitting clades.The latter part is so important for taxonomy because sister groups often have very different evolutionary histories.From two related groups descended from a single most recent common ancestor, one of which is almost indistinguishable from the ancestral group, the other may have entered a new adaptation zone (adaptation zone) and evolved into an entirely new type.Although cladistically named they are "sister groups", taxonomists have traditionally arranged them into different hierarchical levels.Nothing illustrates the difference between cladistics and traditional taxonomies better than Hunnek's view that sister groups must be arranged in the same order no matter how much they differ from each other in divergence after separation .

Cladisms are "phylogenetic classifications" for Hennick, and seek to represent phylogenetic evolution in taxonomy (although the method is ill-suited for this purpose).But his intent was not shared by some of his followers, who not only opposed references to evolution and phylogeny at all but also consciously denied that evolution should be reflected in classifications. (Michener. 1977; Szalay, 1977; Hull, 1979). Let me conclude with a fair assessment of cladistic attempts.The great advantage of cladation analysis is that it is an efficient method for testing the "naturalness" (ie, monophyletic or monophyletic) of taxa that have previously been demarcated by characterization.As the resemblance of species and genera may have many causes, monophyly can only be confirmed by a rigorous analysis of the homology of the characters on which the resemblance is based.

To understand how fundamental the impact of Hunnek's methodology was, one only needs to look up recent taxonomic revisions, especially those pertaining to fish and certain phyla of insects.Even though some scholars, such as Michener, do not believe that cladograms can be directly translated into classifications, they still carefully try to use the principle of common derivation to divide taxa.Cladism analysis is particularly effective when the number of characters is extremely large and the existing taxonomy is extremely imperfect.New cladograms resulting from cladistic analysis have successfully demonstrated that many previously recognized taxa are actually polyphyletic (multi-sourced).However, the translation of detailed analyzes into absolutely corresponding classifications, such as Rosen's (1973) classification of higher bony fishes, has led to a large number of new uses of previously existing taxon names and the creation of many new names, which is even more harmful. Mindful is the introduction of many new levels of class.Some objected to this approach as clearly incompatible with the supposed simplicity of classification, but Bonde (1974: 567) retorted that it was not a "correct argument against Hunnek's theory." It is not against cladistic analysis, but against cladistic classification.

Perhaps the most significant contribution of Hennek's method has been to help clarify the relationship between phylogeny and taxonomy.Simpson, Meyer, and other taxonomists were ambiguous when discussing the relationship between the two.Angiosperm taxonomists, intimidated by the extreme difficulty of reconstructing phylogenetic plant phylogenies, disagreed with zoologists' conclusions that taxa should be consistent with phylogenies and that higher taxa must be monophyletic (Mayr, 1942: 277-28O) conclusions. Davis and Heywood wrote in their textbook Principles of Angiosnerm Taxonomy (1963): "Taxonomists claim that classification should be based on or reflect phylogeny. This aim is unrealistic in sufficient classes...Indeed, we consider the whole notion of phylogenetic classification to be false" (1963: XViii).These authors ignore that fossil data are equally inadequate for most classes of animals, yet their phylogeny must be inferred.It is Hunnek's great credit that he has codified a method that allows such inferences to be made and allows them to be tested repeatedly.The lack of fossil data therefore does not prevent the establishment of a phylogeny.As far as I know, all recognized phylogenies of the order Mammalia are originally based on comparative anatomical studies (via homology), and previously established lines.Phylogenetic development has never been negated by later fossil discoveries.

Should the classification represent the phylogeny (pedigree), should it be based on the system.The seemingly eternal debate over whether development should be consistent with phylogeny or not have anything to do with it is now beginning to be clarified.Clearly, both taxonomy and phylogeny are performed on a hypothetical-deductive basis.This means that a series of propositions must be tested: (1) members of each taxon are nearest relatives to each other (that is, most similar to each other); (2) all members of a taxon are the most recent common ancestor (monophyletic or monophyletic); (3) Linnaeus' taxonomic unit hierarchy is consistent with the deduced phylogeny.

There are many ways to test each of the above propositions, all of which ultimately boil down to the analysis of homologous phenomena.What is most important in the study of cognates is to "distinguish definitions from valid evidence, and get used to deciding whether definitions are appropriate at all" (Simpson, 1975: 17). Only one definition of homology after 1859 is biologically plausible: "A character (character, structure, etc.) is homologous in two or more taxa if it can be traced back to the The same (comparable) characteristics of the putative common ancestor of these taxa."

In order to decide whether the definition is accurate in a special case, many scholars have provided a lot of evaluation criteria.As far as morphological characters are concerned, the criteria listed by Remane (1952) are the most complete.Some of the criteria he cites (for example, with respect to the location of other structures) do not apply to behavioral or biochemical homology; in fact different criteria for evidence of homology may need to be developed for each type of trait (trait).It is therefore indeed unfortunate and inappropriate for Remane to elevate the standard as proof of cognates to that as a definition of cognates. 5.4 Traditional or Evolutionary Classification Approaches Numerical taxonomy and cladistic taxonomy each have a group of followers.Most taxonomists, however, retained traditional taxonomic methodologies, although they also drew advanced methods from these two new schools.This methodology attempts to represent in taxonomy not only the branching of pheletic lines but also their subsequent divergence. Such an attempt may be made by the ranking of various taxa showing whether they have become radically different from their sister groups by invading new habitats or areas of adaptation.The result of this is to transform the cladogram into a phylogram (Mayr, 1969).This school of thought is sometimes called the school of evolutionary taxonomy because it follows Darwin literally.Sometimes it is also called the eclectic taxonomy, because it adopts some methods of numerical classification and borrows the distinction of ancestral-derived characteristics of cladistic classification.The methods and principles of evolutionary taxonomy are described in detail in the textbooks of Simpson (1961) and Meyer (1969), as well as in the articles of Book, Ghiselin, Michener, and Ashlock. The main difference between this approach and the cladistic approach is the appropriate weighting given to unique derivations.Unique traits are derived traits acquired by one sister group that are not possessed by another sister group.Birds were identified as a separate class of vertebrates because of the large number of characteristics (characters) acquired by birds after branching from the clade of ancient lizards, which exceeded the number of characteristics that distinguished ancient lizards from other reptiles many times, Instead of combining birds and crocodiles (the only extant ancient lizards) in the same class or order.Likewise, fleas are assigned a separate order or suborder, although they are clearly from the order Diptera, and lice, although from the order Trichophaera, which in turn are from the nematodes, are also assigned a separate higher taxon.In these cases, as well as in others (eg, a branch that acquires a large number of unique derivations due to a major change in a particular adaptation), purely cladation distorts kinship (Kim and Ludwig, 1978). Thus, the hierarchical arrangement of taxa in evolutionary taxonomy is based on the relative weighting of unique morphologies compared to common morphologies of sister groups. Renxi, Huxley and others have repeatedly emphasized that the progressive evolutionary part of evolution (anagenetic component) often develops into a clear evolutionary grade (grade), or level of evolutionary change, which must be confirmed in the classification. Cladonists objected that this would introduce subjectivity into classification, whereas evolutionary taxonomists advanced two arguments against it.The first argument is that cladism is equally fraught with subjectivity in determining the polarity of evolutionary change, as in mosaic evolution, and in determining evolutionary parallelism.The second argument is that it is not too difficult in most cases to calculate the approximate ratio between the unique and common denominations of two sister groups.Whenever a clade (family tree branch) undergoes a sharp reorganization into a new area of ​​adaptation, this shift may have to be weighted more heavily than proximity to a common ancestor.Unique traits are particularly important because they reflect the occupation of new habitats and new areas of adaptation, which are often more biologically important than common traits. The term and concept of "evolution level" has a long history. Lankester (19O9) believed that protozoa and metazoa were continuous evolutionary stages. Later, after separating sponges as lateral animals, he divided the remaining metazoans into two evolutionary stages: coelenterates and coelomates. Bather (1927) made extensive use of the concept of evolutionary stages and attempted to show how certain genealogies passed through several evolutionary stages in successive geological periods.More recently Huxley (1958) pointed out that the concept of evolutionary level is very useful in elucidating evolutionary development and can serve as a basis for arranging taxonomic levels.Renxi and Simpson have also called attention to the existence of evolutionary levels, because through it many speciation (cladogenesis) can occur without forward evolution. The reason why cladism ignores the existence of evolutionary level is because the concept of evolutionary level allows the confirmation of "paraphyletic taxa".In cladistic terms, a monophyletic group is "paraphyletic" if it is not "holophyletic", that is, does not include all descendants of a common ancestor.The Cladism School believes that the Prarphyletic group is not a real natural group and cannot be classified as an independent advanced taxon. Crocodiles, for example, are traditionally identified as a paraphyletic group because it excludes birds and mammals, two groups separated by having reached a level of evolution that distinguishes them from the rest of the reptiles.Confirmation of paraphyletic groups, while hindering the automatic translation of taxonomy into cladistic patterns, can indicate the degree of divergence in a way that cladograms cannot. 5.5 New taxonomic traits The key to the aforementioned three classification methods, that is, numerical classification, cladistic classification and evolutionary classification, is the analysis and evaluation of taxonomic traits, and traits that lack sufficient information are generally one of the three competing methods. The reasons why the disputes among them could not be resolved for a long time.A taxonomist's most common complaint is that the groups of animals or plants he is studying do not provide sufficient characters to establish their relatedness with certainty.There are two phenomena that have a significant influence on this difficult situation.One is that the phenotypes of some well-known groups of organisms are significantly "standardized", such as the hundreds of species of Rana or the thousands of species of Drosophila, so that only the relationship between them Provides extremely limited morphological clues.Another phenomenon is that any change from this standard pattern usually affects only a single functional system and is associated with a particular adaptation.Shifting to a new food source or adopting a new set of courtship signals may induce perceptible morphological reorganization into many individual traits.But it would be quite wrong to treat these individual traits as separate traits, since phylogenetically they reflect only a single change in function, and Darwin had long warned against placing too much faith in specialization (: p. 414) . A more frequent difficulty for taxonomists is that conclusions drawn from different structures are often contradictory.For example, a study of limbs might show that taxon b is very clearly related to taxon a, while characteristics of the gut suggest that taxon b is the closest relative (margin) of taxon c.Evaluating further features of the extremities or the gut in this case is seldom a complete answer. There are more such situations encountered in each advanced taxon, so taxonomists have paid special attention to seeking new taxonomic traits in the past ten years.While exhaustive morphological analyzes continue to provide new traits, non-morphological traits are increasingly important in establishing taxonomy.These include behavioral traits, life histories, annual cycles, physiology, ecology, parasitism, and any other biological attribute imaginable.Many of these characters are particularly useful in distinguishing species, while some indicate interspecific relationships. Geographical distribution can often provide unexpected clues, as Darwin noted long ago.An unusual Australian genus is more likely to be related to the native Australian family, and its closest relatives are much less likely to be found in South America or Africa.Although the principle of finding close relatives in geographically accessible areas is not applicable to residual species and groups with very strong dispersal ability, it is indeed feasible in many cases, such as Simpson and Thorne respectively. The animal and vegetable kingdoms give many examples.Henick and his followers have pointed out that the combination of cladistic analysis and biogeographical analysis is sometimes particularly convincing. The study of biochemical traits, which was the newest research frontier in taxonomy only a few decades ago, is now one of the most active and productive fields. Immunological research first began shortly after 1900 (Nuttal, 1904), and immunological methods are still used today (Leone, 1964), but many new methods have been added since then.Most prominently it studies the distribution, variation and evolution of molecules.Relatively small molecules, such as alkaloids and saponins in plants, tend to have very strict taxonomic distributions and thus can indicate genetic relationships (Hegnauer, 1962; Hawkes, 1968).In the case of larger molecules their evolution can be studied in a variety of ways, in particular by changes in amino acid sequence.Such changes or differences can often be quantified and used to develop dendrograms representing phenetic distances.The research of some macromolecules, such as hemoglobin, lysozyme, cytochrome C, etc. is time-consuming and requires expensive equipment and instruments, and the wider application also requires automatic analysis and determination.Biochemical methods are most useful when speciation analysis fails or gives only equivocal results.Particularly interesting results have been obtained using electrophoresis for the analysis of enzyme alleles (isoenzymes) (Ayala, 1976).This approach not only reveals a very large number of sister species, but also indicates that the differences between the two species are roughly related to the time along the evolutionary line that led to the separation of the two species.Electrophoretic analysis is most effective as an independent check of morphological analysis results if it is based on a sufficient number of loci (gene loci). The DNA hybridization method leads directly to the genotype (hereditary type); in this method, the compatibility of most of the genomes of the two species to be identified is determined, and the degree of matching directly indicates the degree of kinship.Of course, individual molecular traits are as susceptible to convergence as individual morphological traits.Therefore, the establishment of molecular single-trait classification and the establishment of morphological single-trait classification are equally dangerous. Molecular methods are particularly needed when studying junctions of higher taxa, such as the orders of birds or the phyla of invertebrates.Morphological analyzes have been unsuccessful in this regard due to the inability to find a sufficient number of apparently homologous traits and the often indeterminate polarity of evolutionary trends. The results of morphological and molecular analyzes do not always agree, as the comparison of humans and chimpanzees suggests.Some scholars have therefore suggested that there should be two sets of classifications, one based on morphological characters and the other based on molecular characters.This suggestion seems ill-advised for a number of reasons; not only because different molecular traits are likely to call for different molecular classifications, but also because the suggestion implies several phylogenies, which is patently false.Classification is not the classification of individual traits of organisms but the classification of organisms as a whole.Fusing the discoveries of morphological, behavioral, and various molecular traits into a single, optimized taxonomy will be the task of future synthesis. Philosophers have traditionally expressed some concern about the principle of classification.Indeed, classification (though not biological classification per se) was one of Aristotle's central concerns (see Chapter 4).The substitution of ascending classifications for descending ones in the post-Linnaean period was a major philosophical advance; nineteenth-century philosophers such as Mill, Jevons, and the Thomists still showed great interest in classifications.However, philosophers did not seize the conditions and opportunities created by Darwin's revolution to draw the necessary conclusions on biological classification.Almost unanimously they still cling to essentialism and other outdated notions that evolutionary thought has discarded.For example, they confuse identification and classification, seeing classification as a process of identification involving individual specimens when in fact taxonomy studies populations (species) to which individual organisms are simply assigned (i.e., identified).Even today some philosophers (Hempel, 1965) still believe that "classification is the division of large classes into subclasses" (descending classification), although in fact evolutionary classification is the classification of related taxa into higher taxa. The most serious flaw in the views of most philosophers is that "the classification of animals and plants ... is in principle the same as the classification of inanimate objects" (Gilmour, 1940: 465), and the classification method of the numerical taxonomy school is also based on this assumption .Artificial or arbitrary classifications are justified for objects that are strictly classified according to some quality or characteristic, such as library books.However, there are certain limitations to the classification of objects for which interpretive doctrines already exist (Mayr, 1981).This is true, for example, of the causality of disease and of the classification of organisms on the theory of affinities due to a common ancestor.No meaningful classification is possible if a state of affairs is classified as a product of historical development without due consideration of the historical process from which it originated.Classification of planetary bodies, geological phenomena, human cultural components, or biological diversity, in Gilmour's opinion, would in most cases not adequately reflect the actual interrelationships between events.Thus, since Darwin there has been agreement among evolutionary taxonomists that natural taxa must be monophyletic, that is, they must be descended from some common ancestor.This theoretical basis for all biological classifications is a strong constraint and a complete refutation of the notion that the theory of classification applies equally to living things and non-living things.The younger generation of philosophers such as Beckner, Hull, etc. have recognized this, and are cooperating with those biologists such as Simpson, Meyer, Bock who pay close attention to the relationship between evolution theory and classification to develop the philosophy of biological classification. 5.6 Facilitation of Information Retrieval The conclusions reached by evolutionary taxonomists about phylogenetic relationships are represented by phylograms, which give equal weight to the exact location of branch points and the degree of differentiation (i.e., the number of unique lines) of each phyletic lineage. weighted.Evolutionary taxonomists use this phylogenetic diagram as the basis for their generalizations. But classification has another function: as a review of the information stored in the system.In order for a taxonomy to be the most effective information retrieval system, certain aspects of the taxonomy, the "rank" The terms rank, "Size of taxa", "symmnetry", and "Sequence" refer to these aspects.There is inevitably subjectivity in determining these aspects and they will likely form the focus of long-running debate. In Linnaeus' hierarchical structure, a rank is represented by the order in which a given taxon falls.Determination of grades is the most difficult and arbitrary decision-making process in classification.For cladistic taxonomists, rank is automatically determined by the nearest branch point of the tree, since sister taxa must have the same rank.In contrast, evolutionary taxonomists must decide how many unique lines and their weights to determine the rank difference between two sister groups, and making such a decision when different kinds of traits provide conflicting evidence. Especially difficult.For example, a molecular taxonomist might place orangutans and humans in the same family because of their macromolecular similarities, while Huxley suggested that humans be ranked higher because of the unique characteristics of humans' central nervous systems and their capabilities Become a separate realm (psychozoa, spiritual realm). There are no hard and fast rules for resolving such contradictions. It can only be said that the overall balance of the system should be taken care of, and a hierarchical standard that can make the most effective generalizations should be adopted. The question of the most appropriate size for a stem taxon is even more divided among taxonomists.Some scholars believe that as long as there are a few differences enough to identify new genera, families, and higher taxa.In the jargon of taxonomy, these people are called "Splitters".Most taxonomists prefer large, comprehensive taxa for better representation of kinship and ease of memory.They were called "lumpers".These two factions have been arguing endlessly since Linnai's period. Linnai himself is a dominant faction. He only uses four levels of hierarchical structure (species, genus, order, and door) outside the realm to deal with the natural world. diversity.Even fairly conservative taxonomists have now identified 21 levels (Simpson, 1945).Linnai only confirmed 312 genera for the entire animal kingdom, while modern zoologists have confirmed more than 100,000 genera, of which there are 2045 genera for a single bird.As a general rule it can be said that most taxonomic groups undergo a phase of intensive differentiation when they are actively studied, and that this phase of differentiation is reversed when knowledge of the group is more mature.There is broad agreement that classification as an indexing function of an information retrieval system imposes constraints on the size of taxa and the number of levels in the hierarchy. The numerical school of taxonomy is the only modern school of taxonomy that has endeavored to introduce uniformity and non-subjectivity in the hierarchical arrangement of taxa.Using various methods of measuring distances based on morphological traits (Sokal and Sneath, 1973) or based on genetic distance (Nei, 1975), they proposed absolute differences that characterize species classes for generic separation degree (absolute degrees of difference).When distances are measured on a sufficiently broad basis (such as DNA matches, or isozymes with at least 30 or 40 loci, and preferably more), they can adequately reflect the amount of evolutionary divergence in various species classes (amount of evolutionary divergence).There is evidence that if the degree of morphological differentiation is seriously inconsistent with the degree of molecular differentiation, the criteria for group discrimination based on the degree of molecular divergence should be different in different higher taxa.Morphologically very similar species such as frogs and toads may show clear molecular divergence, whereas clear divergence in morphological and color patterns in bird and mammalian taxa does not reflect any significant molecular divergence.If uniform molecular criteria were adopted, many long-recognized genera of warm-blooded vertebrates would be eliminated as synonyms, and many new genera would appear in the anurans and gastropods to replace morphologically very similar ones. species groups.Given the primary function of classification, it is doubtful whether this is necessary at all. Symmetry issues are posed by evolution, not taxonomists.Ideal symmetry is achieved if all taxa at each meta-level are the same size.For quintarians, the ideal number is five.The idea that all taxa should have approximately the same number of species was first advanced when natural theology still dominated the thinking of naturalists.This question was first asked by A. von Humboldt studied it, followed by von Buch, and another article by an anonymous entomologist discussing this issue in 1835 caught Darwin's attention.The wild inconsistencies in the sizes of taxa seem to be so inconstant as to be unworthy of a Creator's planning.Unfortunately, evolution (and extinction) are so fickle.Some animals have only one species in a whole order, but many genera, especially among insects, have more than a thousand species.It is now clear that the rates of speciation and their survival in different parts of natural systems are highly unequal. Perhaps the most difficult problem in classification is converting a family tree into a linear sequence (order).The question is also simple in principle, as long as there is only one single standard of perfection.As Lamarck said, begin with the most imperfect creature and end with the most perfect.When Cuvier destroyed the natural ladder, he discovered a new ordering criterion in the affiliation of characters.He denied any continuity between the four phyla he identified, yet ranking them according to the degree of development of their central nervous systems clearly dictated the order and thus maintained the basic idea of ​​a perfect standard.Accepting evolutionary thinking has little effect on sequential insights into taxonomy.The meaning expressed by the ladder of nature is simply to propose a further cycle: "more perfect" organisms become "more evolved" or "higher" organisms.In fact, all classifications of animals and plants are based on such an explicit or implicit principle: the more primitive or lower organisms are arranged first, and the higher ones are arranged later.Yet in due course one will ponder over the meaning of the word higher.Why should fish be superior to bees?Why should mammals be superior to birds?Which is higher, the parasite or the free-living type from which it derives? As the study of animal and plant affinities matured, it became clear that neither a well-established standard nor a simple family tree could properly describe the diversity in nature.More correctly, it is best to regard most groups of organisms as highly complex phylogenetic shrubs, with innumerable equal branches: each branch begins with a very simple primitive ancestor, ends with a very complex and specialized The descendants of the end.The fact of the adaptation radiation makes it impossible to establish a really logical theory of taxonomic order.It is impossible to argue that one taxonomic order is superior to another in most natural systems.Hence, there is a growing trend towards the adoption of purely pragmatic criteria in favor of information retrieval (Mayr, 1969).The most important principle is to preserve all generally accepted order, unless it has been positively shown that it pinches unrelated taxa together.Ongoing debate in the taxonomic literature about the optimal order of Angiosperms or Songbirds suggests that these minimal constraints are not sufficient to ensure stability; linear order is, however, a practical requirement.The specimens in the collection are arranged in linear order, as are all revised studies, catalogs and printed texts of reviews. Given that taxonomy is the oldest branch of biology, its current vigor and momentum is indeed remarkable.This is specifically reflected in the establishment of some new taxonomic journals (such as "Taxon" Taxon), "Systematic Zoology" (Systematic Zoology), "Systematic Botany" (Systematic Botany, etc.), published a whole set of important textbooks, held many international seminars and an increasing number of yearbook literature.It involves a wide range of research areas and is not limited to classification methods.Simply describing new species is a never-ending job. The number of newly discovered or at least confirmed types in recent decades is even more astonishing.For example, the new phylum Pogono-phora was not described until 1937, and Gnathostomulida even later (1956).Among the only remaining bony fishes, Latimeria was discovered in 1938, the primitive mollusk Neopilina was discovered in 1956, and the ancient crustacean Cephalocarida was discovered in 1955.A rich intertidal fauna has also been discovered over the past 50 years.The most primitive metazoan Tricera genus was not discovered until the 1970s. Perhaps the most astonishing discovery was the description of Precambrian fossils by Barshoorn, Cloud and Schopf, who pushed back the history of life on Earth from 6.5 years to 3.5 billion years.But sometimes just a closer look at existing fossils can reveal something, as in recent descriptions of the extinct early Cambrian invertebrate phylum Asmata. The progress in the classification of higher taxa of all organisms from bacteria, fungi, protozoa to vertebrates (including primates) also shows the vitality of taxonomic research in recent years.腔肠动物的祖型究竟是水螅还是水母这个长期未能解决的争论近来也已通过多方面的研究得到澄清（绝大多数分类学家认为是水螅）。钵水母纲（Scyphozoa）较之其它类型的腔肠动物具有更多的祖征，而新近确认的立方水母纲（Cubozoa）（Werner，1975）将之与水螅纲很出色地联系了起来。 Thorne，Carlquist，Cronquist，Stebbins以及Takhtajan等人对植物的研究导致被子植物的彻底重新分类。但是亲缘关系不清楚或完全不知道的高级分类单位的数量还很多，还有待今后努力。 早在林奈时代，甚至更早（亚里斯多德）就将生物分成植物界和动物界。真菌和细菌被看作植物。近几十年来对单细胞生物和微生物研究愈加有所进展就越发认识到这种分类的人为性。首先是认识到兰绿藻和细菌与其它一切生物根本不同因而将之归类为原核生物。它们缺少有结构的细胞核和复杂的染色体，而且和其它生物（真核生物）在绝大多数的高分子上也不相同。在细菌中也具有多样性（代谢和其它方面），然而即使分化最大（显然也最原始）的细菌类群甲烷细菌也具有许多其它细菌的共同特征，因而最好的办法是将它们一并归入原核生物界（Monera）。 现在一般也将真菌列入一个单独的界以脱离植物，它们和植物不仅在代谢上不同（不进行光合作用），而且在细胞结构上也不相同（永远是单倍体）。究竟是否要为单细胞动物、植物另立一个原生生物界（Protista）是一个判断力问题。因为原生动物和单细胞藻类的文献和后生动物、后生植物的文献都是分开的，所以单独分出一个界可能有利于信息检索。Margulis（1981）曾经讨论过这些问题。 生物分类之所以能取得稳定的进展原因有很多，其中最重要的是分类方法的改进。 现在已经认识到分类不是一步程序，不能一蹴而就，因此过分简单化的方法很少能取得满意的结果。分类是由一系列步骤组成，（Mayr，1981），在每一步骤中需用不同的程序和最有效的方法。例如数值分类法在首先试验性地划分分类单位时以及根据支序分析差异程度安排分类单位等级时最为有用。支序分类法则在检验由推论得出的分支模式上最为有用。 数值分类法在什么程度内有效并超过人的头脑这个问题并没有解决。形态学性状由于趋同现象，复系（多源，Polyphyly）现象和镶嵌进化而难于解释以致不十分宜于作为数值分析的素材。趋同和复系现象在高分子（可能是DNA）的进化中也出现，但是也有征象表明高分子的某些变化对这些分子的随后进化施加了如此强大的约束力以致于只要有足够大量的高分子同时被评价，则分子相似性要比皂白不分的形态表征分析更为可靠，这一点正是数值分类学派最初提出的。 5．7多样性研究 在本世纪的前半期一般都把“分类学”和“系统学”这两个名词作为同义词看待。 如果要问系统学的任务是什么，分类学家会回答，“描述自然界的多样性（意思是指描述多样性所赖以存在的物种）并将之分类。”然而即使远在列文虎克和施万麦丹（Swammerdam）所处的17世纪就已经很清楚，研究生物多样性并不局限于对物种进行描述与分类，这只是分类学家首先要完成的任务。多样性研究从一开始就包括分析生活史的各个阶段以及两性异形（Sexual dimorphism）。当研究自然界中活的动物时发现不同的动物占有不周的生境，选择不同的食物，具有不同的行为。但是直到本世纪中期随着新系统学和进化综合（或综合进化论，evolutionary synthesis）的兴起才充分认识到多样性研究的重要意义。这时系统学功能的传统定义才明显地显得过于局限，不能反映真实的情况。 这样一来辛普森为了从术语上将分类学和系统学明确地区别开来，他于1961年重新为这两个术语下定义；“分类学”仍然保持传统的定义。而将“系统学”定义为“科学地研究生物的种类和多样性以及两者之间的一切关系。”这样就把系统学理解为多样性的科学，这一新的扩大了的系统学概念后来一直被普遍采用。这个新定义立即引出了这样的问题：系统学应当包含哪些功能？它在现代生物学中应当起什么作用？ 狭义的分类学仍然是整个系统学（科学）的主干和基础。要将现存的动、植物物种列出完整的清单目录并将之安排到分类中去看来是一项永无休止的任务。螨，线虫，蜘蛛，或昆虫中某些被忽视的类群或海洋无脊椎动物的分类学仍然能够卓有成效的花费他们的全部精力去描述新种并将之指定到适当的属中。自然界生物的多样性看来是无限的。 现在每年几乎描述一万个动物新种，如果我们承认未描述种的最低估计数字，那末还要再花两百年时间才能完成单只为现存的物种命名和加以描述的任务。 分类学的一个奇特方面是它的各个分支的独立自主性。根据对生物类群了解的成熟程度，对每一类群采用的方法和概念也各不相同。的确，在现代分类学的某些特殊领域中仍然可以发现从林奈，居维叶一直到新系统学各个不同阶段的概念争论。例如甚至在今天对一些学者来说“分类”这个术语仍然被理解为鉴定体系。在鸟类学中被普遍接受的多型种分类单位在动物分类学的很多其它领域中则是从来没有听说过的。分类学中各个分支的独立性还可以动物分类，植物分类，微生物分类中各有自己的一套命名规则明显地看出。 多样性是生物界的两个重要方面之一，另一个是生命过程。但是研究多样性的重要性并不是一直都被人们认识到。因此，在生物学历史上系统学也有其兴衰起伏。在林奈时代它实际上统治了生物学，在达尔文以后建立系统发育的时期中它又曾一度高涨。但是一部分原因是由于作为前一阶段过分重视的反作用，随之而来的就是忽视（如果不是被压制的话）多样性研究的阶段。只要看一看Max Hartmann的普通生物学和HansDrieseh，TH．Morgan，Jacques Loeb以及其它实验生物学家的着作就不会想到多样性研究也是生物学中一个重要而又繁荣的领域。这种被冷落的现象一部分是咎由自取，因为这个阶段中多样性的研究一般过分着重描述性（环境生态学和大部分分类学）或者片面强调系统发育问题（比较解剖学，Heinroth，whitman的行为学）。当多样性的研究者对更一般性的问题感兴趣时，他们的最终目的往往是重建共同祖先。 目前对于这种情况是在什么时候和怎样改变的还缺乏有效的历史分析。然而很明显的是在20、30、40年代出现了新的事态。很多迹象表明种群系统学是打开这种局面的楔子。它在苏联促使切特维尼可夫（1926）创立了种群遗传学。种群系统学上升为新系统学（壬席，赫胥黎，迈尔），后者又为进化综合作出了决定性贡献。进化思想的传播，特别是种群思想的传播在古生物学，进化形态学、生态学和行为学中都引起了新的概念化。和多样性有关的问题以及根据多样性的比较方法在上述事态的发展上都发挥了重要作用。 对多样性的重新重视深刻地影响了整个生物学领域的概念气氛。例如，几十年来进化被看作是种群中基因频率的变化。这一还原论者的定义将进化生物学局限于现存种的变化，也就是说局限于进化的适应一面。多样性的起源好像并不是进化的一部分而被忽视。这种态度表现在50、60年代的大多数古生物学家的着作中。辛普森以及其它的一些现代古生物学家将他们自己几乎完全囿于进化的纵向方面，甚至在讨论适应幅射时也忽略了多样性的起源问题。直到1972年古生物学家才对多样性的起源予以适当的重视（Eldredge and Gould）。多样性研究在进化形态学中也引发了新的概念。现在已从片面地重视共同祖先（通过同源相似性的研究）转向于注意后裔差异的起源，即注意到了多样性。在行为学中似乎也发生了同样情况，虽然还仅仅是处于发展的初期阶段。 有鉴于对多样性的新态度的影响已经遍及机体生物学的各个领域，因而比较详细地讨论一下系统学的特殊贡献也是值得的。为了驳斥门外汉认为系统学不过是涂上光荣色彩的集邮的普遍印象这也是必需的。现在有一种倾向是把系统学的最重要贡献，归之于邻近科学领域（如种群遗传学，生态学和行为学），虽然这些贡献实际上是由从事实际工作的分类学家作出的，而且也只有通过作为分类学家所取得的经验才有可能作出这样的贡献。将“分类学”或“系统学”的标记（或名称）只限于单纯的记录员的描述工作，而将从更基本的描述工作涌现出的更显着的发现和更广泛的概念贴上另一种不同的标记，这种作法显然是十分错误的。 开始时（17、18世纪）系统学和博物学是同一个研究领域，这一点必须记住。机体生物学的大多数现已确认的分支都是由系统学发展而成。生态学的大部分涉及不同物种的相互作用，诸如竞争，共生，捕食者-猎物关系等等。这些相互作用的本质只有通过详细研究相互作用的物种才能了解。生态学中的MacArthur学派的研究工作几乎全都涉及多样性，生态系统的研究也是如此。由于动物的大多数行为具有种的特异性，又由于大多数行为的进化是由比较不同物种而察知的，因此，可以明显看出生物学的这些分支和多样性研究是多么密切地结合在一起。还有另外一些生物学分支是完全依赖系统学的，这包括生物地理学，细胞遗传学，生物海洋学和地层学。至于系统学对公共卫生，农业及自然保护等这样一些应用科学的必要性就勿容赘述了。 系统学除了作为机体生物学上述分支学科的基础而外也许更为重要的是它在扩大现代生物学的概念方面所作出的贡献。生物学中最重要的一元化学说进化论主要是系统学的贡献。达尔文在贝格尔号航行中遇到的分类学问题以及对藤壶分类专心研究了八年之后写出了决不是偶然的巧合。分类学家还为解决很多个别的进化问题提供了主要线索，包括隔离的作用，物种形成的机制，隔离机制的本质，进化的速度，进化的趋向以及进化的突现问题。在所有这些问题上分类学家（包括古生物分类学家）所作的贡献较之其它类型的生物学家都更重要。 分类学家还积极参与了进化综合（Mayr and Provine，1980）。很多学者成功地将遗传学和进化的主要问题加以集成，如切特维尼可夫、壬席、迈尔、杜布赞斯基、辛普森等人都具有分类学家历史背景。 环境生理学的进展也应归功于系统学。很多动物系统学家，如Gloger，Allen，壬席，对适应性地理变异以及气候规律的建立都作出了重要贡献。地理族（地理宗）中的适应性差异的遗传学基础也正是由一些具有分类学本领的动物学家予以论证的。 多样性研究所作出的最重要贡献也许在于发展了哲学上的新观点。正是由于多样性研究才摧毁了最狡诈的哲学本质论。由于强调每个个体是独特的，和其它个体不同，因而多样性研究者就集中力量研究个体的作用，从而产生了种群思想，这种思想对人类亚群、人类社会和人类的相互作用极端重要。由于指明了每一物种都是独特的，不可代替的，这就告诉我们要尊重进化的每一项单独产物，这就是种群思想的最重要组成部分。 由于强调个体的重要性，发展并运用种群思想，而且尊重自然界的多样性，系统学就为人的概念化提供了一个因次（维），这个概念化因次是被物理科学大部分忽略了的（如果不是被否定了的话），然而对人类社会的幸福和为规划人类的未来却是极其关键的。