Epsilon Archive for Student Projects

Abstract

Erigeron acris L was first described in 1753. Subspecies (ssp.) has since been added resulting in today’s five having distribution in Nordic countries. However the treatment and characteristics of these are not consistent in all floras and some traits are described for a few subspecies and not stated at all for others. Subspecies have perhaps in some cases been added without fully reviewing the taxa that earlier have been described. This study is trying to give some direction into which traits actually separates the subspecies and discusses whether these are enough to define them as separate. 23 variables were studied on 113 herbaria specimens divided among the subspecies, as they are defined today. The specimens were mainly from Lund Botanical Museum. However, specimens from the Museum of Natural History in Stockholm and the Museum of Natural History in Helsinki were also lent and processed.
The data were then analyzed through ANOVA (Analysis of Variance) and PCA (Principal Components Analysis). The Principal Components Analysis was used to illustrate if and how the subspecies overlapped. The ANOVA was used to see if there were any traits significantly separating the subspecies. The first ANOVA was only made on E. a. ssp. acris, E. a. ssp. droebachiensis and E. a. ssp. politus. Four traits could here be found which significantly separated the three ssp. Three characters could not distinguish any from the others and the rest of the characters could distinguish one of the three subspecies, in different combinations.
When looking at the ANOVA made with all five subspecies, four traits could not distinguish any of these subspecies from the others. All other 19 traits had some subspecies partly or completely overlapping and some could only be distinguished from some others. In three of these cases E. a. ssp. acris could be separated from all the other subspecies. The other four subspecies were overlapping with at least one other in all traits.
E. a. ssp. droebachiensis and E. a. ssp. brachycephalus overlapped partly or completely in 22 out of 23 traits. E. a. ssp. droebachiensis and E. a. ssp. decoloratus could as well only be separated by one trait completely. As a consequence, a possible reduction of the number of subspecies is discussed.