Wood, Troy E. , Rieseberg, Loren H. .
The Frequency of Polyploid Speciation in Land Plants: A Phylogenetic Approach.
Two methods for estimating the frequency of polyploid species formation have been proposed: 1) the haploid multiples method and 2) the evens-odds method. The first approach is based on the percentage of congeners that possess haploid numbers that are multiples of the lowest number for the genus. This method, which yields an estimate of 30-35% for angiosperms, is biased upwards because it discounts divergence at the polyploid level. The second approach is based on a model that allows inference of a polyploid index - the fraction of changes in chromosome number due to doubling. This fraction is then multiplied by the minimum number of speciation events within a genus that involved a change in chromosome number. According to this approach, 2 to 4% of angiosperm species and 7% of fern species are the result of polyploid speciation. While elegant, this modeling approach is biased downwards because it fails to identify independent origins of the same ploidy level within a genus. We have employed two different phylogenetic methods to obtain unbiased estimates of the frequency of polyploid speciation in land plants. The first method involves the direct mapping of chromosome numbers onto phylogenetic trees, whereas the second tabulates the frequency of chromosome number transitions between sister taxa only. While the second method is less comprehensive than the first, it avoids potential problems caused by missing data. Nonetheless, both approaches provide similar estimates of the frequency of polyploid speciation and indicate that independent ploidal changes within a genus are relatively common.
Log in to add this item to your schedule
1 - Indiana University, Department of Biology, Jordan Hall, 1001 East Third Street, Bloomington, Indiana, 47405, USA
Presentation Type: Oral Paper
Date: Monday, August 15th, 2005
Time: 3:00 PM