I. Phylogenetic Classification

A. Definition
    Recall that phenetic classifications are based on overall similarity. In contrast, phylogenetic classifications attempt to reflect the geneology or evolutionary history of a particular group of plants.  

B. Darwin
    The publication of Origin of Species (1859) certainly marks the beginning of the phylogenetic phase of classification. Plants (as well as other species) were now recognized as being dynamic entities that change through time, one species giving rise to successive species. Botanists attempted to reflect these relationships in their classifications.

C. Vertical vs. horizontal relationships
    One way to envision phenetic relationships is that the units that are being classified are grouped "horizontally" - similar units are put into the same groups, etc. However, this horizontal system doesn't include any indication of vertical relationships - or in other words, which group may share a common ancestor, and how recently.  Thus, the goal of phylogenetic classification is to group species to reflect the evolutionary, or vertical, relationships between them.  This is obviously possible because plants have a past history (thanks Mr. Darwin!).  

    There are two major approaches to creating a phylogenetic classification - evolutionary and cladistic.  Both systems attempt to reflect the relatedness of the units in the classification but they approach the problem from different perspectives.   This chapter will feature evolutionary classification.  Click here for information about cladistics.

II. Constructing an evolutionary classification

A. Collect evidence.
    Taxonomists accumulate as much data about the group to be classified as possible. Sources of data include: morphology, anatomy, biochemistry, physiology, ecology, palynology, etc. The more data, the better!  Actually, this stage is the same in for all of the classification systems (phenetic/numerical, cladistic, evolutionary).

B. Weighting the evidence.
    Not all data are considered to be equally valuable for constructing an evolutionary classification. Thus, a taxonomist must make some judgments about which data are most important to best reflect the relationships between the units.

C. Primitive or advanced?
    To determine evolutionary relationships, the taxonomist must determine which features are primitive or ancestral and which are advanced or derived. This is often a matter of opinion. Fossils and comparative studies of various features of extant species provide evidence to support taxonomic opinions.

D.  Some principles for evolutionary classifications (Thorne, 1976)

1. Existing species have descended with change from pre-existing species and are, therefore, the products of evolutionary change. In other words, evolution has occurred, and it has occurred in the past much the same as evolutionary processes are occurring today.
   
2. The prevalence of convergent evolution in habit, function and structure is a predictable consequence of the relatively limited means that angiosperms have for effective reproduction and adaptation in various environmental niches. In other words, convergent and parallel evolution are common.
   
3. The primitive, ancestral condition of any given characteristic can be no more specialized than its condition in a derived, existing species that is primitive for that characteristic. In other words, if a certain taxon is believed to be the most primitive member of a group, then the ancestors of that taxon cannot have features more specialized.
   
4. Evolution may tend toward reduction and simplicity or elaboration and diversity (specialization)
   
5. The rate and direction of evolution may vary in different organs and tissues of plants. Thus, most plants possess a combination of primitive or conserved characters as well as some advanced or specialized characters.
   
6. Most existing angiosperms are highly specialized and greatly modified from their primitive, generalized ancestors. Thus, it is unrealistic to consider living representatives of families as being derived from existing representatives of closely related families.

E.  Construct the classification.
    Once the data are gathered and analyzed, the classification is then constructed. 
 

III.  Evolutionary Classifications (an historical look)

1875. August Wilhelm Eichler (German, 1839-1887) - one of first, used de Jussieaus and others as a starting point. Assumed complex was more advanced.

1892. Adolf Engler (1844-1936) with Karl Prantl (1849-1893; both German) - revised Eichler's work. Die Naturlichen Pflanzenfamilien, 20 volumes. Organized genera of world flora with keys, descriptions. Considered monocots primitive; considered conifers to be ancestors of angiosperms - believed stobilus with unisexual features gave rise to catkin, each scale evolved into a unisexual apetalous flower. Evolutionary trends:

• apetalous → distinct → connate
• actinomorphic → zygomorphic
• unisexual → bisexual
• hypogynous → epigynous

    Some fatal flaws: woody apetalous plants were considered primitive (Amentiferae), monocots and dicots with different ancestors. Several revisions including those by C.F.Diels, H. Melchior, von Wettstein and A. Rendel.

1915. C. Bessey (1845-1915) first U.S. botanist to make contribution. Student of Asa Gray at Harvard. Developed set of about 30 "dicta" or principles of phylogeny. Among other ideas, he:

• Disagreed with Engler
• Considered the Ranales (order with plants like Ranunculaceae and Magnoliaceae) to be primitive; 
• dicots → monocots; 
• hypogynous/epigynous not important in determing evol. trends; 
• strobilus was not unisexual, but bisexual; 
• apetalous flowers advanced, not primitive; and 
• oversimplified – he suggested that modern taxa arose from other modern taxa.

1926. John Hutchinson - English. Director at Kew Gardens. Devised his own system. Based on woody and herbaceous with others radiating from those. Believed in polyphyletic systems rather than monophyletic ones.

Modern systematists.

• Armen Takhtajan (1910- ). Russian.
• A. Cronquist (1919-1992). New York Botanical Garden
• Robert Thorne (1920- ). Rancho Santa Anna BG, Claremont, CA
• Rolf Dahlgren (1932-1987). Danish.

    Cronquist and Takhtajan treat angiosperms as a Division. Major groups start at class level (i.e., dicot vs monocot). Next are a series of subclasses, etc. Major divisions are subclasses.

    Thorne and Dahlgren treat angiosperms as a Class with two major subclasses, monocots and dicots. Then further subdivided into superorders, etc. Major divisions are superorders.

IV. Problems with evolutionary classifications

A. Convergent evolution
    Species with similar selection pressures look alike (similar appearance) and hence, can "trick" a taxonomist.

B. Lack of appropriate fossils.
    Obviously, fossilization is a sporadic, chancy process. And, some events happened so quickly that fossils may not adequately document the changes (i.e., the angiosperms appeared very rapidly in the fossil record).

C. Strict evolutionary classification assumes a monophyletic origin of groups.
    In other words, the ancestors can only come from one group, they cannot be polyphyletic. An example: the members of your immediate family (i.e., grandparents, parents, siblings) make up a group with a single origin (monophyletic). However, you wouldn't include the neighbors in this group since they have a different origin. In other words, a group containing your family and say, me, would be polyphyletic.

    Logical conclusion: all angiosperms arose from a single common ancestor. What was the ancestor?

D.  Intuitive/Reproducibility

VI. Comparison Exercise.
    Check out the Flowering Plant Gateway web site sponosored by Texas A&M Univeristy.  For fun, take a family and compare its treatment in the classifications of Thorne, Dahlgren, Takhtajan and Cronquist.  

VII. Evolutionary Trends in Angiosperms
    The following  trends are, in general, agreed upon by those who practice this method of classification.

• tropical → temperate
• woody → climbing or herbaceous
• vessel-less → vessels
• perennial → biennial or annual
• vascular bundles in ring → scattered
• chlorophyll present → absent
• evergreen → deciduous
• stipules present → absent
• leaves simple → compound
• flowers bisexual → unisexual
• flowers insect-pollinated → flowers wind-pollinated
• flowers solitary → inflorescences
• floral parts spirally arranged → whorled
• perianth undifferentiated → calyx and/or corolla reduced
• flowers with petals → apetalous
• petals free → connate
• actinomorphic → zygomorphic
• hypogynous → epigynous, perigynous
• stamens many → few
• stamens separate → connate
• carpels many → few
• carpels free → connate
• fruits single → aggregate
• fruit a follicle → capsule, berry, drupe, etc.