Part I:
Review of Concentrated changes test and character optimization

1) Define the following terms

DELTRAN:
ACCTRAN:
MINSTATE:
MAXSTATE:

2) Why would or when would you choose one over another?

Part II:
A more realistic Concentrated changes test
*use this excercise as a means of practicing your Macclade skills (that is with character optimizations, etc. try to choose alternative optimizations that will tend to disfavor your hypotheses (or vice versa), for example)

3) Open file Hammemelidae under Macclades examples ³Morphological Data² folder.

4)There are 56 morphological characters in this matrix of plant characters (with various numbers of character states).  Choose two binary characters that you think might be most labile or environmentally determined. You may do this in any number of ways (e.g., by intuition or by actually examining the homoplasticity in each of the characters).

5) These two characters are your dependent variables.

6) Your independent variable is Habitat.  These taxa either exist in the tropics or in temperate climates.  Your hypothetical ancestor has character = tropics (character state = 0).  And you now need code each character for their habitat.  (what assumptions are you making here??
1.

2.

The codes for each taxon are as follows:

Habitat
0
0
0
0
0
0
0
0
0
0
0
1
0
1
1
0
1
1
1
1
1
0
0
0
0
1
1
 
 

3. Now using your knowledge of Maddisonıs test,  test which of your two variables seems most effected by the independent variable (habitat).
 

4. Imagine now that you have another independent variable, ³Predation.²  This is a binary character (luckily) with states 0= mongoose predation and 1= moth predation.  You would like to see if the different predators have differentially caused morphological changes in you taxa.  Given your intuitive knowledge of plant parts, choose among the binary dependent variables for one you think might be effected by predation and test this hypothesis.
One obvious choice might be toxin production (alkaloids) given that alkaloids are noxious chemicals potentially related to predatory behavior.

Your coding scheme for the Independent variable is:
(I have also provided a complete set of alkaloid data for the taxa in question)

Predation Alkaloids
0 a
1 p
0 a
0 a
1 p
1 a
0 p
0 a
0 a
0 a
0 p
0 a
1 p
0 a
0 p
0 a
1 a
1 a
0 a
1 p
1 p
1 a
0 a
1 p
0 a
0 a
0 a
 

5. what can you say about the influences of predation on Toxin presence/absence?
 

By now you should have noticed the powers and limitations of the concentrated changes test. Remind yourself of the assumptions of this test (and keep in mind that you should think critically when choosing among comparative methods)

Part III:
A beginning to understanding Maccladeıs Molecular analysis capabilities.
 
 

First a demonstration
the point of this excercise is to take you through the many ways one cna use Macclade to analyze molecular sequence data , but while keeping a specific goal in mind.

1) What is a step matrix and why would we want to build one?
2) How does one build a step matrix?
extrinsic, a priori information regarding character state transformation
weight assignment (how to scale?? and what is appropriate and reasonable?
3) At what level in the tree building to tree use to tree interpretation ³hierarchy² can we apply or infer information for a step matrix?