It seems clear that a complete phylogenetic tree (best as itcan be reconstructed) would be important and useful. But whyshouldn't it coexist with the current taxonomic system, which,complete with paraphyletic groups, offers a different sort ofinformation? It seems that both systems would be of use tobiologists.
1. (Jennifer Yeh): Your proposal to have separate phylogeneticand Linnean taxonomies has also been proposed by an Italiansystematist, Alessandro Minelli. While I am not opposed tohaving separate, coexisting taxonomies, I feel that traditionaltaxonomies are not among those we want to keep around. In myview, traditional taxonomies should be replaced byphylogenetic taxonomies, which might then coexist withtaxonomies of relevant classes. Traditional taxonomies arehybrids between phylogenetic taxonomies and taxonomies ofrelevant classes. Although some authors consider this astrength of the traditional taxonomies, I consider it aweakness. Consider the traditional taxon Reptilia, whichconsists of ectothermic amniotes. As far as I can tell, thereare two "taxa" of interest here, and Reptilia is not one of them. One of them is amniotes--a group consisting not only of whatare traditionally called reptiles, but also of birds andmammals. This taxon is significant because it is a clade--aphylogenetic entity consisting of an ancestor and all of itsdescendants. The other significant "taxon" is ectotherms, theclass of organisms that rely largely on external heat sourcesto maintain their body temperatures. This group includes notonly some amniotes (traditional reptiles), but also most non-amniotes (amphibians, fishes, etc.). There are othertheoretically relevant classes: herbivores, carnivores,homeotherms, heterotherms, etc. Traditional taxonomies arehighly arbitrary in recognizing a taxon Reptilia for theectothermic amniotes, but not recognizing a taxon for theuricotelic amniotes or one for the oviparous amniotes. Intheory, we could name all of these hybrid groups as taxa, but Idon't know what good it would do us. In any case, this could notbe done in the context of the the traditional taxonomic system,which does not permit partly overlapping taxa.
Obviously there are problems with reconciling Linnaeantaxonomy with cladistic relationships. In your opinion, is itfeasible to do so, or should we begin thinking about a newtaxonomy, with or without binomial nomenclature? Have anyattempts been made to come up with a new system, and if so,what happened?
2. (Barrak Pressler): Although it is possible to reconcileLinnean taxonomy with cladistic relationships, there is a costin terms of instability. The solution to this problem, as yousuggested, is a new system of taxonomy. It's interesting thatyou should ask whether attempts have been made to do this,because Jacques Gauthier and I have written several papers inwhich we have at least started to develop new a taxonomicsystem based on phylogenetic principles. It would be difficultfor me to summarize these papers here, so instead I'll give youthe references: 1) de Queiroz, K. and J. Gauthier. 1990. Phylogeny as a central principle in taxonomy: Phylogeneticdefinitions of taxon names. Syst. Zool. 39(4):307-322. 2) deQueiroz, K., & J. Gauthier. 1992. Phylogenetic taxonomy. Annu.Rev. Ecol. Syst. 23:449-480. 3) de Queiroz, K. & J. Gauthier. 1994. Toward a phylogenetic system of biologicalnomenclature. Trends Ecol. & Evol. 9(1):27-31. These ideas are brand new, so it's difficult to knowwhether they will catch on, though some people have alreadyadopted them in empirical studies.
I found your in Systematics and the Darwin Revolution to bevery thorough and convencing. However, given that we haveproblems with the Linnaean taxonomy what would be the bestway to make this change universal? Wouldn't it also be hardwhen working with information from the two conflictingsystems?
3. (Megan Chen): Thanks for the positive comments. Yes, thetransition from a Linnean system of taxonomy to aphylogenetic one will cause some difficulties. During thetransition period, it will not always be easy to know which oneis being used by a particular author. The only way that thechange will become universal is if everyone accepts it, and Isuspect that this will take at least one human scientificgeneration. The development of a new set of taxonomic rulesis critical to making the transition as fast and smooth aspossible, and this is something I've been working on withJacques Gauthier for the last several years (see comment 2).
?. Topic: The Darwinian Revolution Since writing your 1988article on the delay of the Darwinian Revolution in biologicaltaxonomy, has there been any progress towards the adoption ofa truly phylogenetic systematics?
4. (Christina Campbell): 1) You say that Brooks and McLennanadvocate a method for determining common ancestry in whichsimilarities and differences between groups are used to definethese similarities and differences. I would describe theprocess of phylogeny reconstruction as using similarities anddifferences between groups to infer their recency of commonancestry. This does not seem circular. Perhaps you arereferring to the practice of later using the phylogenetic tree tointerpret the similarities as homologous or non-homologous. In this case, there may be problem of bias in that thesimilarities are assumed to be homologous unless the treesuggests otherwise (similar to a null hypothesis in statistics),but I would not say that the argument is circular in the sensethat the premise presupposes the conclusion. 2) Your nextquestion raises several issues about species concepts. Oneconcerns the relationship between degree of similarity andinferences about the number of species. As an advocate of anevolutionary lineage concept of species, I agree with you in notplacing too much emphasis on simple degree of similarity. Some things can be rather different and still be part of thesame evolutionary lineage, while others can be very similarand be parts of different evolutionary lineages. Therefore, thesimple fact that different populations of chimps are moredifferent in some set of characters than are differentpopulations of humans, does not seem sufficient evidence tosay that they represent separate species. And yes, this couldbe the result of human mobility. On the other hand, one wouldwant to know how the degree of genetic difference wasassessed and which genetic systems were examined. If geneticdivergence was measured between individual organisms (ratherthan between populations) for non-recombining (e.g.,mitochondrial) genes, then human mobility would not seem tobe able to account for the lower levels of differentiation inhumans relative to chimps. 3) There are two problems here. The first is where to draw the line in theory. This question canbe answered and indeed, it has been answered several times. The problem is that different people give different answers. Idon't know if biologists will ever be able to agree on where todraw the line in theory, since they've been disagreeing about itfor many years. The second issue is whetherwe can make accurate inferences about lineages from thedegrees of difference, given that we have agreed about whereto draw the line in theory. Biologists have developed variousanalytical methods that allow us to make fairly accurateinferences about whether populations are part of the sameunit, defined in a number of different ways, but no inferencecan be made with absolute confidence.
If I understand correctly, phylogenetic systematics relie oncommon ancestry of evolution. But many argue that differentphenetics, such as in birds and reptiles, can't be ignored nomatter how recentiy they share a common ancestry. Also, whathappens if new phylogenetic ideas come forward, dophylogenetic systemics have to be altered everytime the newidea is put forward? I would like your comments. Thank you.
5. (Ki-Wan Kim): You are correct in saying that someapproaches to macrotaxonomy are based entirely on recency ofcommon ancestry, while others are based at least partly ondegree of similarity. Whether the latter can or should beignored is a matter of debate. Since you have read my paperthat was assigned for this discussion, you probably knowwhere I stand on this issue. Changing ideas about relationshipsmay cause taxonomic changes whether the taxonomy isphylogenetic, traditional, phenetic, or something else. Thenature of the changes is somewhat different, as JacquesGauthier and I have discussed in the papers for which I gavereferences above (comment 2). Some taxonomies are moresensitive to changes than others, but the degree of sensitivityhas more to do with the precision and explicitness of thetaxonomy than whether it is phylogenetic.
In your 1988 paper on systematics and the Darwinianrevolution, you argue that both classification andsystematization have central and complementary roles. Doesthe relative importance, in your opinion, of each role changewhen discussion changes from animals to plants to protozoansor even with different groups of organisms. ie. Isclassification relatively more important when dealing withanimals than with plants? ...protozoans? etc. Thank you!
6. (Chris Brown): I haven't thought about this issue much, but Idon't think that the relative importance of classification andsystematization depends on the group (plants versus animalsversus protozoans). In fact, I can't even say that one isrelatively more important than the other in any particulargroup. My view is that they are both essential, even in thecontext of very specific questions. Sometimes one will appearmore important, but this usually results from the fact that theother is so obvious that we don't think about it. For example,in performing phylogenetic systematization we might easilytake for granted that the systems of interest are members ofthe classes called species and clades. Similarly, when weclassify organisms with regard to a particular biologicalphenomenon, we often take for granted that the entities beingclassified are organismal systems.
Topic: The Darwininan Revolutionin biological taxonomy.
Re: 1988. Systematics and theDarwinian Revolution. Philosophy of Science 55:238-259At the close of your article, you predict the replacement ofLinnaean taxonomy. What is your view of the replacemenmt?Is an evolutionary classification possible? I believe that itmay work for existing entities, but how can we includepaleospecies with living descendants? Aren't all the speciesthat radiated from a common ancestor sub-taxa of thatancestor? Every living thing may have to be classified as asub-species or subgenus of the original species. It wouldappear that classifications can never fully reflectevolutionary history. Other than eliminating organizations ofspecies into classes and assigning taxonomic names only tonodes on a cladogram, we may be forced to accept paraphyletic taxa in a classificatory scheme.
7. (Cristoph von Pohl): I do indeed believe that thereplacement of the Linnean system with an evolutionary one ispossible. As for how I view this replacement, I must onceagain refer you to the papers for which I gave references above(comment 2). The problem that you raised concerningpaleospecies is only a problem in the context of the Linneansystem, in which the species category is viewed one level in ahierarchy of taxonomic categories. In the phylogeneticapproach that Jacques Gauthier and I have proposed, thespecies category is not viewed in this way. Instead, it isviewed as one of only two important taxonomic categories--species and clades. Furthermore, we advocate eliminating theentire hierarchy of Linnean categories, and consequently,species do not have to be assigned to taxa representing any ofthose categories. Thus, a species that was the ancestor ofsome large clade would be assigned only to thatlarge clade and not to any of its subclades. For example, ifspecies x was the ancestor of all birds, we would say simplythat it is a bird, and we would not worry about what order,family, and genus it belongs to. It is almost certainly true thattaxonomies can never fully reflect evolutionary history, butthis does not mean that they cannot reflect part of that historyaccurately--in other words, we are not forced to acceptparaphyletic higher taxa.
Have there been any recentfindings which you feel support your theory particularly well?
8. (Jacques Finlay): Much of what I wrote in comment 1applies to your question. Rather than repeating it, I will try toextend the line of reasoning to ecology. I would say thattraditional taxonomies are not as useful as they may appear,and that ecologists might benefit from consideringalternatives. These alternative are not necessarilyphylogenetic taxa; sometimes they are taxonomies of trueclasses. For example, it strikes me that some of the groupsstudied by ecologists are somewhat artificial. I've seennumerous ecological studies that limit their focus along thelines of traditional taxonomies. For example, someone mightstudy competition among nectar feeding birds. I wonder whythe study is restricted to birds when all organisms that feedon nectar are competing for this resource. On the other hand,sometimes evolutionary ecologists will find phylogenetic taxavery useful. For example, in studying the process of adaptiveradiation, it would seem artificial restrict an analysis to thespecies in a paraphyletic genus. The adaptive radiation isformed by all species descended from a particular ancestor,not only those assigned to a particular genus. Evolutionaryecologists are not unaware of the benefits of taking aphylogenetic approach to their studies. Jonathan Losos, aformer TA/GSI in your course, has done some interesting worktaking a tree-based approach to ecological questions. Ingeneral, the tree-based approach is a very active area incurrent evolutionary ecology (see Brooks and McLennan, 1991,Phylogeny, Ecology, and Behavior; Harvey and Pagel, 1991, TheComparative Method in Evolutionary Biology).
Topic:Homology In your statement
in Systematics and the Darwinian Revolution, you alluded tothe fact that homologous structures provide evidence for close ancestry. Given that, what is your basis for determining whatsorts of homology this includes? My question stems fromreading B.K. Hall's Homology: The Hierarchical Basis ofComparative Biology (1994). In his introduction, in sectionVII, referring to George Lauder, he states that "non structuralcharacters (behavior, function) stand as homologues in theirown right." Can behavior independent of structure be a factorin determining how organisms are hierarchically evaluated? And how can behavior by itself be used at all in the study ofevolutionary homologies?
9. (Michael Feinberg): This question is not in my area ofexpertise, but I'll try to answer it. I don't see any reason whybehavioral or functional characters cannot be used insystematics, and in fact, they have been. One of the classicexamples is Konrad Lonrez's study of the duck family. You askwhether "behavior independent of structure" can be used todetermine how organisms are related. Certainly, behaviors arenot independent of the structures with which they areperformed, but this doesn't negate their usefulness assystematic characters. The problem is potential redundancy,which applies to structural as well as to behavioralcharacters. Incidentally, my brother, a former student in theclass you are now taking, has argued that behavioral charactersare not significantly worse for reconstructing phylogeneticrelationships than are traditional morphological characters (deQueiroz and Whimberger, 1993, Evolution 47:46-60). I'm notsure what you mean when you ask about the use of "behavior byitself" in the study of evolutionaryhomologies. I would think that the use of structures bydifferent organisms to perform similar behaviors might qualifyas evidence for homology of the structures, but that use ofstructures in different organisms to perform differentbehaviors wouldn't be evidence either for or against homologyof the structures. Behaviors themselves might be homologous,but this cannot be inferred from the behaviors alone. That is,other information than the behavior itself (such as it'sdistribution among taxa) is needed to make inferences aboutinheritance of the behavior from a common ancestor.
10. (Francis B. Gonzales): 1) I think I said something aboutthis in the paper you read (the first paragraph that begins on p.254). It's true that we can never know evolutionary historywith absolute certainty, but I don't think that this is as big aproblem as it is often made out to be. I might even go so far asto say that most of what we think of as scientific progress hasbeen built on theories the truth of which is not known withabsolute certainty. Furthermore, basing taxonomies onevolutionary history is not all that different from basing themon "known characters", even from a practical standpoint. Thatis to say, both are subject to revision. Therefore, it seemspreferable to have taxonomic revisions reflect changes in ourideas about how things are related in terms of theirevolutionary history rather than reflecting some subjectivedecision about which characters are more important. 2) Myuse of the terms "phylogenetic systematics" and "evolutionarysystematics" in the paper you read has caused someunfortunate confusion. I used both terms to designate anapproach to taxonomy that recognized evolutionary systems(species and clades) as taxa. This terminology was derived>from my emphasis on the distinction between classificationand systematization. Perhaps this was not a wise decision,because the term "evolutionary taxonomy" has been applied to avery different approach to taxonomy that was prevalent duringthe middle part of this century and is still embraced by somebiologists. I consider this approach to be incompletelyevolutionary in that the principle of evolution plays only therole of an after-the-fact explanation for the existence of taxarather than a central tenet from which the concepts of taxa arederived. For further discussion see my 1992 review of thetextbook by Mayr and Ashlock (Systematic Biology 41:264-266).
10b. (Kevin Scully): Thanks very much for the flatteringcomments about my writing style. I try to write clearly, butfor me this requires a lot of work (numerous revisions). In thecase of the paper that you read, I also tried deliberately tomake my writing accessible to a more general audience,specifically, the kind of people I thought would be reading thejournal Philosophy of Science, where the paper was published. I estimated that these people would have a general knowledgeof biology, but not a detailed knowledge of systematics. As aguide, I tried to write the paper so that my father would beable to understand it, but he later told me that he didn't. I'mpleased that somebody appreciated myefforts. There are different reasons why the writings ofscientists are not always accessible to people other thanexperts in their own field. Some scientists just don't writewell. Although writing is an important part of science, wereceive relatively little formal training. Some scientists arein a hurry to get their papers published (a pressure imposed onthem by the discipline as a whole) and don't put in the effort,even though they have the ability. This approach can backfire.There is a clear cost to writing unintelligible papers--nobodyreads them. Finally, some scientists write well, but they don'tintend for their papers to be read by people outside of theirspecialized field. I suspect that many of my other papers aredifficult for most people, even systematic biologists, to read.Even though I always try to write clearly, many of my paperswere written for a very specialized audience. Certain topicsare sufficiently esoteric that non-specialists probablywouldn't want to read about them no matter how well thepapers were written. If you have a chance, look up some papersby Robert J. O'Hara (1988, Syst. Zool. 37:142-155; 1992, Biol. &Philos. 7:135-160; 1993, Syst. Biol. 42: 231-246). I find hiswriting very easy and enjoyable to read.
11. (April Loui): When I first came up with the idea ofreplacing the Linnean system of taxonomy, I thought that thechances of this actually happening were very slim. Therewould be a tremendous amount of tradition to overcome. Butafter learning more about the history of systematic biology,I've become more optimistic. The reason is that I see whatappears to be a series of related changes in the history ofsystematics that set the stage for replacement of the Linneansystem of taxonomy with a phylogenetic one. In my opinion,the Darwinian revolution in taxonomy didn't happen all at oncebut instead has been progressing in stages over the last 130+years. The replacement of the Linnean system is the nextlogical stage in the progression. I think that the new systemwill be more functional than the current one in many respects(see the third paper cited in my comment 2), though it willtake some getting used to. I don't think that it will requireresearchers to base all their observations on DNA sequences. Alot of systematics can be done without DNA sequences, whichis just another kind of data in need of interpretation.
12. (no name given): Deciding where species begin and end(particularly the former) is one of the primary components ofthe "species problem." Some people want to draw the line atthe initial separation of lineages, others at the firstacquisition of a diagnostic character, others at the point wherethe gene lineages coalesce, and others at the point where theseparation is irreversible. I'm not sure why you think thatblurry lines leads to the biological species concept. As I seeit, speciation is a continuous (blurry) process, but more or lesssharp lines can be drawn at several different places. The BSCcorresponds with one of those places.
13. (Julie Froehlig): Although I categorized traditionaltaxonomy as the result of classification, I don't think that theclasses recognized in it are the most useful ones (see comment1). Most of the classes that are important in biology result inrelatively simple classifications (e.g., producers versusconsumers; endoterms versus ectotherms), which alreadycoexist with the traditional taxonomy that serves as a generalreference system. I think that the traditional taxonomy willultimately be replaced by a phylogentic one, but there will stillbe a system of rules governing taxonomic practice (seecomment 2). Whether or not this becomes "internationallyaccepted" remains to be seen. In my view, classification andsystematization are both important in biological taxonomy. But the important classes are not kingdoms, phyla, classes,orders, etc.; they are populations and clades. And the speciescategory is not one level in a hierarchy of artificial classes; itis a particular kind of population. Finally, the individual taxathat are members of the categories called species and cladesare not artificial classes but systems that owe their"individuality" to the processes of interbreeding and commondescent. I might add that in a later paper (de Queiroz, K. 1992. Phylogenetic definitions and taxonomic philosophy, Biol. &Philos. 7:295-313), I revised my position on the interpretationof monophyletic taxa as classes. I concluded that phylogenetictaxa could be viewed as classes in the sense of groups whosemembers belonged to those groups because they possessed aparticular property or properties. The property shared by themembers of a clade, however, is not an organismal trait butrather descent from a particular ancestor.
14. (Jim Pearson): I agree with your suggestion that the ideaof evolution had to overcome beliefs that would seem torepresent a more significant barrier to acceptance than thosecurrently hindering acceptance of a phylogenetic approach totaxonomy. Perhaps "resistance" was a poor choice of a term onmy part, since something can fail to change both because ofactive opposition ("resistance") and because of passive factors("inertia"). However, I do not think that the lack of change insystematics can be attributed to active opposition broughtabout by the realization that change will lead to moreconfusion and controversy. Active opposition is a short-termphenomenon that only comes into play when people becomeaware of the potential for change. In the case of systematics,many of the controversial issues didn't become controversialuntil long after 1859, and once people became aware ofpossibility of change, changes happened in spite ofconsiderable active resistance. If active opposition cannotaccount for the slow progress of the Darwinian Revolution insystematics, then the cause must then be sought in a terms ofone or more passive factors. I proposed that it was largely theresult of taking certain traditions for granted (e.g., the ideathat taxa are classes of organisms sharing certain traits). Finally, I want to point out that taxonomy is not the only areaof biology in which the Darwinian Revolution was delayed. Therecent proliferation of tree-basedapproaches to studying ecology, behavior, physiology, etc.(see comment 8) seems to represent the next stage of theDarwinian Revolution in those disciplines.
15. (Scott Kozinchik): Yes, I do advocate changing thetaxonomic system, though not completely. The underlying basiswould be changed, as would many of the recognized taxa, butmany of the same names and taxa would remain. Although Iagree that many studies can be successfully conducted withthe phylogenetic trees resulting from systematic analysis,that is, regardless of the taxonomies derived from them, it isdebatable whether the traditional taxonomy has been usedsuccessfully for a long time. One might say instead that wehave been able to make progress using the traditional taxonomyin spite of its shortcomings, though it has often lead us astray(see Schwenk, K. 1994. Biol. J. Linn. Soc. 52:69-82 for anexample). We may be able to "get by" with the traditionaltaxonomy, but a phylogenetic one will work better.
16. (David Kilimnik): The fact that we don't always know allthe details of phylogeny is not really a problem, because alltaxonomies (including traditional ones) are provisional. Andjust because we don't know all the clades does not mean thatwe have to resort to paraphyletic grades. In almost every case,we know something phylogenetic relationships. considersnakes. Although the relationships of snakes are poorlyunderstood at one level, we can assign them to a number ofmonophyletic taxa (Amniota, Lepidosauria, Squamata, to nameonly a few). Concerning similarity, I didn't mean to imply thatphylogenetic inferences are based on something other thansimilarities and differences. The point is similarities must beinterpreted in the appropriate context. When to goal is toreconstruct phylogenetic relationships, one must consider howthe similarities evolved in the context of a branching tree. When we do this we discover that some organisms can besimilar in retained ancestral characters even when they do notshare particularly recent common ancestors, and otherorganisms can differ substantially from one another (if onepossesses many derived characters) even though they share arelatively recent common ancestor.
17. (Karen Chen): My opinion is that the current taxonomicsystem does not even accomplish its own stated goals, and thata phylogenetic system would accomplish those goals better(see de Queiroz, K. & J. Gauthier. 1994. Toward a phylogeneticsystem of biological nomenclature. Trends Ecol. & Evol.9(1):27-31). In other words, any disadvantages of adopting aphylogenetic approach are greatly outweighed by theadvantages. To the extent that this is true, it seemsreasonable to push for replacement of the current system. Maintaining the old system along with the new will onlygenerate confusion, which goes against the basic purpose oftaxonomy--communication. I also think that the traditionaltaxonomy is anthropocentric (see p. 252 of the paper you read)and that progress in other scientific disciplines hasoften been associated with shifts away fromanthropocentric perspectives.
18. (Toni Ann Vaughn): The current taxonomic system doesallow us to name things, but my colleague Jacques Gauthier andI argue that the names are ambiguous and unstable under thatsystem. The problem is that the names are implicitlyassociated with evolutionary entities, but the rules ofnomenclature are based on nonevolutionary concepts. This iswhat causes all those annoying name changes that hindereffective communication. A phylogentic approach to biologicalnomenclature would also allow us to name things, and it wouldnot cause many of the arbitrary name changes that occur underthe current system (see references given in comment 2).
19. (no name given): Given that we accept the principle ofevolution, any grouping based on a shared character can beinterpreted as evolutionary in the sense that the characterpresumably owes its existence and distribution to the processof evolution. But this grants the concept of evolution only thesuperficial role of an after-the-fact explanation for thegrouping. This is evident from the fact that groupings based onshared characters (including those based on the way organismsmake a living) existed before the concept of evolution wasaccepted and are still recognized by anti-evolutionists. For ataxon concept to be more than superficially evolutionary itmust be based on evolutionary principles (rather than beingretrospectively justified in terms of evolutionary principles).Contrast the concept of a taxon as a group of organisms thatmake their living in a certain way with the concept of a taxonas an entity made up of a certain ancestor and all of itsdescendants. The first concept makes no reference to anyevolutionary phenomenon, while the second makes explicitreference to the evolutionary phenomenon of ancestry anddescent. The first could be applied even if evolution did notoccur, while the second would be meaningless if evolution didnot occur. I'm puzzled by your example of the emerald tree boaand the green tree python. Nobody I know recognizes a taxon forthese two distantly related species. The fact that one COULDrecognize such a taxon on the basis of shared traits despitetheir distant evolutionary relationship gives clear evidence ofthe non-evolutionary nature of such an approach.
20. (Scott Turner): I believe that several of the importantstages in the Darwinian Revolution in taxonomy resulted fromabandoning the perspective that you describe, in which taxa areviewed as arbitrary groups of organisms. First, during theModern Evolutionary Synthesis, biologists accepted populationor population-lineage concepts of species, which implied thatspecies taxa are not arbitrary groups of organisms. Next,during the cladistic movement, biologists accepted a conceptthat equated higher taxa with clades (systems of speciesrelated by common descent), which implied that higher taxa arenot arbitrary groups of species. Both of these advancesresulted from granting the principle ofevolution a more important role in taxonomy--specifically, therole of a central tenet underlying the concepts of species andhigher taxa (clades). Part of the reason that taxon namescurrently seem so arbitrary is that we still have not grantedthe principle of evolution the role of a central tenet innomenclature (see references given in comment 2). Therefore,names like Hominidae, Pongidae, and Homininae change theirmeaning not only as the result of changes in ideas aboutevolutionary relationships, but also as the result of arbitrarydecisions related to the assignment of taxa to the Linneancategories of family and subfamily. The problem with yourstatement that a universally accepted Linnean taxonomy wouldbe as useful as a phylogenetic taxonomy is that there is nouniversally accepted taxonomy (Linnean or phylogenetic), and abasic tenet of the current codes of nomenclature is to opposethe intellectual tyranny that would be needed to achieve one. As evolutionary biologists, what we want is a system thatallows disagreements about relationships but promotesstability with regard to the evolutionary meanings of taxonnames when there are no such disagreements. The Linneansystem does not accomplish this goal, but a system based onphylogenetic principles would.
21. (Mehrdad Afrahi): You suggested that the problem withessentialistic concepts of taxa is that their advocates avoidenumerating the characters by which they recognize particularspecies and simply say that their members share an essence.However, some people consider a species conceptessentialistic if the name of the species is thought to bedefined in terms of necessary and sufficient characters. Onecould hardly advocate such a concept and avoid enumerating thecharacters. In my view, the problem with essentialisticspecies concepts is that they attribute the reality of speciesto an underlying essence. This begs the question of speciesreality in that it rests on the reality of essences, which hasnot been established to the satisfaction of most scientists. Unfortunately, the term "essentialistic" is confusing becauseseveral different ideas are commonly associated with theterm.
22. (Duncan Parks): I'm not sure I understood what you meantby "practical utility", but I'm going to infer from the rest ofyour question that it has to do with Linnean categoriesconveying something about diversity. In my opinion, theLinnean categories misinform us as much (or more) than theyinform us along those lines. If we want to compare clades ofequal diversity or equal disparity or equal age or equalanything else, we would be much better off measuring thevariable of interest directly rather than using assignment tothe same Linnean category as a (poor) substitute. As youyourself say, the estimate is rough.
23. (Joshua Gatts): In order to differentiate among closelyrelated species we need to "measure" their similarities anddifferences. But under the BSC, the degree of differencecannot, by itself, give us the species. We need to makeinferences about reproductive isolation from what we knowabout the similarities and differences. Thus, many differences in rapidly evolving nonfunctionalcharacters might not be considered indicative of separatespecies, but a single difference in a character that is criticalto mating might be. The same is true for phylogenyreconstruction, that is, the similarities and differences bythemselves don't tell us which species are most closelyrelated. We need to make inferences about common ancestry>from what we know about similarities and differences. Thus,many similarities in retained ancestral characters might notbe considered indicative of recent common ancestry, but a fewsimilarities in derived characters might be. Concerning yourlast point about taxonomic changes, one would think that itwould be desirable to have our taxonomies change to reflectchanges in our ideas about evolutionary relationships.
24. Name:Sandra Maldague Topic:Cladistics How is cladisticsdifferent from systematics? Isn't cladistics the systematization of individuals according to common ancestor/number of traits in common?
If cladistics is the same as systematics, and you favorsystematics as a means of taxonomy, then why do you state"Similarity turns out to be a red herring" (p.246, Systematicsand the Darwinian Revolution, 1988). Can it be argued thatproving the presence of a common ancestor for several speciesis having to prove similarities between those species, whether phenetic or genetic?
24. (Sandra Maldague): Your observation is perceptive that myconcept of systematization on the basis of common ancestryrelationships is more or less equivalent to certain cladisticpractices. Several cladists were not so perceptive andbasically accused me of advocating a traditional approach totaxonomy simply because I used the terms "evolutionary andphylogenetic systematics" instead of "cladistics". The reason Isaid that similarity is a red herring is not that I'm againstusing similarities to infer common ancestry but that theinference is more complicated than the simple idea that mostsimilar equals most closely related (see comments 16 and 23).
25. (John Sherman): Should classes and systems beincorporated in to the same taxonomy? Yes and no. Although Inow think that most of what I called systematization in thepaper you read could also be interpreted as classification, Istill think it is important to distinguish between classes thatcorrespond with the parts of systems and those that do not (deQueiroz, K. 1992. Phylogenetic definitions and taxonomicphilosophy, Biol. & Philos. 7:295-313). Mixing these two kindsof classes in a taxonomy would only lead to confusion. Hybridclasses (e.g, traditional paraphyletic taxa) suffer from similarproblems. Please see comments 1 and 13 for further discussionrelevant to your question.
Paul Fine (evo-cr@violet.berkeley.edu.) Topic: Non-divergent evolution
25b. (Paul Fine): I don't believe I said that the concept ofadaptive zone was not useful but rather that justifications forrecognizing paraphyletic grade taxa based on the concept ofadaptive zone were not as evolutionary as they might seem. Isometimes like to classify species concepts as theories versusdeductions. In the former case, we accept that we can alreadyrecognize species, and our species concept is a theory intendedto explain the existence of those species. In the latter case,we accept some process as important, deduce from it theexistence of a particular kind of entity, and then call thoseentities species. Van Valen's ecological species concept fallsinto the former category. He assumes that he already knowswhat are the species of oaks, and then he proposes naturalselection as the explanation of why their component organismsare similar. Although I think thatattempting to explain organismal similarity is worthwhile, Ialso think that accepting the reality of previously recognizedspecies can be counterproductive. Some of the mostsignificant breakthroughs in systematics have come whenpeople deduced the existence of entities from know processesand called into question the reality of previously recognizedtaxa. This has been the case for both species and higher taxa(see comment 20).
26. (no name given): I implied in section 5.4 of thepaper you read, there are two kinds of biologically significantentities that we commonly call taxa--species and clades. Mostof what I talked about in the paper had to do with clades orhigher taxa, not with species. If we view species as entitiesthat result from a different process than common descent,such as interbreeding, then they are not really part of thehierarchy of clades. This position has an importantconsequence, namely, that species resulting from reticulation(hybridization, symbiosis) are not problematical (becausemonophyly is not a necessary property of species).
27. (Carmen Bailey): In a truly phylogenetic approach totaxonomy, all the Linnean categories would be eliminated.Nevertheless, we might co-opt the species category for anentirely different category of entities (rather than the lowestlevel in the hierarchy of Linnean categories--see comment 26).Concerning the names of individual species, there are severaloptions. We could use binomials in which the first name is notviewed as the name of a genus; we could use nonuniqueuninomials corresponding with the second name of thebinomial; or we could use unique uninomials formed by fusingthe binomial names. These are a few of the possibilities. Jacques Gauthier and I discuss advantages and disadvantages ofdifferent options for naming species in our 1992 paperPhylogenetic Taxonomy (Annu. Rev. Ecol. Syst. 23:449-480).
28. (Pamela Wong): Species will not be defunct, but they willbe more clearly distinguished from clades (see comment 26). I've been fairly careful to avoid advocating any particularspecies concept in my recent papers, but in my earlier papers Iwas less careful. You are probably right in saying that in thepaper you read for discussion I tended toward the biologicalspecies concept. I now think that many species concepts arebasically lineage concepts that draw the line in differentplaces (see comment 12). Although I advocate a generallineage concept of species, I don't have a strong preference forany of the specific versions of it. For further information, youmight take a look at my paper with Michael Donoghue (deQueiroz, K., & M. J. Donoghue. 1988. Phylogenetic systematicsand the species problem. Cladistics 4(4):317-338) as well asone by Robert O'Hara (1993. Systematic generalization,historical fate, and the species problem. Syst. Biol. 42:231-246).
29. (Sheila Conway): Unfortunately, yours was the last]question, and it got a bit garbled in transmission. Ithink I got enough of it to figure out what you were asking. Your concerns about the potential difficulties of replacing onetaxonomic system with another are very reasonable, but thechanges would not be as drastic as you might think. The mainthings that would be different are first, that we wouldn'trecognize any paraphyletic taxa and second, that we woulddecide nomenclatural issues differently (the second issomething that many biologists don't pay much attention toeven under the current system). Most existing taxon nameswould still be used, though some of them would be used in aslightly to moderately different sense. Jacques Gauthier and Ianticipated that people would worry about this issue, so weincluded a phylogenetic taxonomy of the vertebrates in one ofour papers to illustrate that what we were proposing wasn'tgoing to be so radically different, at least from theperspective of understanding a taxonomy (see de Queiroz, K., &J. Gauthier. 1992. Phylogenetic taxonomy. Annu. Rev. Ecol.Syst. 23:449-480).