English

Phylogenetic Trees

High schoolBiologyExhaustive four-taxon Fitch parsimony from six editable binary traits

Unlock this lab with a pass

A pass opens every simulation on the site, including Present mode for class. The teaching notes below stay free for everyone.

See passesI already have a pass

One-time payment, no subscription.

About this experiment

Build a tree hypothesis from shared traits rather than a decorative similarity diagram. Edit six binary characters, compare the exact costs of all three unrooted quartet trees and rotate displayed branches without changing relationships.

Learning goals

  • Distinguish a shared trait matrix from a tree hypothesis.
  • Compare the minimum trait-change counts of alternative topologies.
  • Recognize unresolved ties and equivalent branch rotations.

How to use it

  1. 1Inspect the starting matrix and compare all three topology scores.
  2. 2Toggle one taxon trait and observe which topology becomes most parsimonious.
  3. 3Erase all differences and identify the unresolved three-way tie.
  4. 4Select and flip an internal grouping, then verify that its taxa and score stay the same.

The model

Each topology receives its smallest possible total binary-character change count.
Ties are retained instead of implying unsupported certainty.

Assumptions

  • Four taxa A through D have six complete binary characters. Character order is irrelevant and every state change has equal cost.
  • The model applies Fitch set intersection and union rules to all three possible unrooted quartet topologies and sums the minimum changes across traits.
  • All equally parsimonious trees remain identified. The selected tree can also be a higher-cost alternative for comparison.
  • The drawing uses an arbitrary display root and equal layout depths. It does not infer root direction, time or branch lengths.
  • Parsimony is a simplified inference criterion. Convergence, reversals, unequal rates and uncertain character coding can make historical ancestry differ from its preferred hypothesis.

Try this in class

  1. Q1Why do identical trait rows fail to resolve a unique tree?
  2. Q2Why does rotating an internal group preserve a tree relationship?

Frequently asked questions

Are long branches older or more different?

No. Branch lengths and the display root only arrange this diagram. The model compares unrooted relationships and character costs.

Does the preferred tree prove ancestry?

No. It is the least-change hypothesis under the chosen traits and equal-cost assumptions, which can be challenged by additional evidence.