Research Training Program
Highlights from 2007

VIRTUAL POSTER SESSION
2007


A Numerical Taxonomic Study of Muhlenbergia montana and M. filiculmis (Poaceae: Chloridoideae)

Emma Harrower
Research Training Program, 2007



Introduction


Muhlenbergia montana (Nutt.) Hitchc.

Muhlenbergia Schreb. (Poaceae: Chloridoideae) is a large genus of grasses found in the Western Hemisphere. Muhlenbergia montana (Nutt.) Hitchc. and Muhlenbergia filiculmis Vasey are very similar in their morphology, ecology, and distribution.

A discriminant analysis was performed to evaluate the best characters used to differentiate between the two species. Principal components analysis was performed to evaluate the morhophological variation in the data and to determine whether there are two morphologically distinct groups that correspond to the current taxonomic treatment of the species. A geographic distribution map was made to show the location of all plants sampled.


Materials and Methods

Morphological observations were based on unmounted field collections and herbarium specimens from the United States National Herbarium. Specimens representing the range of morphological and geographical variation were chosen for study. Eighteen morphological characters were measured on a total of 114 specimens (62 M. montana; 52 M. filiculmis). Discriminant analysis, principal components analysis, and summary statistics were performed in PC-ORD and Systat 11. Distribution maps were created in ArcView 9.2.


Results

Phytogeography and Ecology

Muhlenbergia montana occurs in the southwestern United States from California, Colorado, and Wyoming to Utah, Arizona, New Mexico, and throughout western Mexico to Guatemala. Muhlenbergia filiculmis has a smaller distribution and is found only in Colorado, Wyoming, Arizona, and New Mexico. Muhlenbergia filiculmis grows between (2134-) 2465-2965 (-3240) meters and M. montana occurs between (1420-) 1922-2632 (-3390) m. Both species are found on rocky slopes in open grasslands and dry meadows.

Multivariate Analyses

Ninety-eight percent of all individuals were properly classified in the discriminant analysis (DA) with only two specimens of M. montana being misclassified. A scatterplot of the scores on DA factor 1 and DA factor 2 shows two distinct groups, corresponding to M. montana and M. filiculmis. Variables with the highest loadings on DA axis 1 were panicle length (0.927), leaf blade length (0.882), leaf blade width (0.875), and sheath length (0.868) and on DA axis 2 were lower glume awn length (-0.635), upper glume awn length (-0.544), lower glume length (-0.484), and plant height (0.338).

A scatter plot of the scores on principal components (PC) 1 and 2, and 1 and 3 depicted two distinct groups. A total of 74.9% of the variation in the data set was accounted for by the first three PC axes.

Morphological Characters

Means and standard deviations for each character are given. Muhlenbergia montana tends to be larger in most characteristics. Muhlenbergia filiculmis can be distinguished from M. montana by having shorter panicles (3.0-) 3.8–7.4 (-10.0) cm, smaller leaves (1.1-) 1.8–8.4 (-18.2) cm, and narrower leaves (0.2-) 0.5-1.1 (-1.5) mm. Lemma awn length was the sixth best discriminant factor but it may be a more useful identification character than others due to its distinctiveness. Poorer characters for use in identification are pedicel length, lemma and palea length, and glume awn length because these characteristics do not differ significantly between species.

KEY SEPARATING M. FILICULMIS AND M. MONTANA

1. Panicles (5.6-) 11.9-25.9 (-44.0) cm long; lemma awns (3.0-) 7.0-16.4 (-24.0) mm long; leaf blades (5.1-) 11.9-32.9 (-51.8) cm long; leaf blades (0.7-) 1.4-2.6 (-4.5) mm wide …...................................... M. montana

1. Panicles (3.0-) 3.8-7.4 (-10.0) cm long; lemma awns 0.5-3.5 (-8.7) mm long; leaf blade (1.1-) 1.8-8.4 (-18.2) cm long; leaf blades (0.2-) 0.5-1.1 (-1.5) mm wide ............................................................ M. filiculmis

 

Discussion

Morphological characteristics can be used to easily distinguish between M. montana and M. filiculmis. The discriminant analysis was able to clearly separate both species and classify 98% of the individuals successfully. The PCA visually displayed the morphological variability and showed that the two species differ most significantly in their overall size. An an amplified fragment length polymorphism (AFLP) study is in preparation and will be used to assess the genetic variability among individuals and between populations of these two species.

Acknowledgements

This project was funded by an endowment from the Smithsonian Women’s Committee. Dan Cole provided technical assistance with the GIS map.




Smithsonian Institution
National Museum of Natural History

Research Training Program

The information presented here, as part of the Research Training Program Virtual Poster Session, represents preliminary data as the result of ten-weeks of investigation in-residence at the National Museum of Natural History. This is not an official publication nor are the finding presented here necessarily conclusive or definitive.

As preliminary information, these results and/or findings should not be cited as part of conclusive work. Please contact the author if you would like further information about this research as well as the resulting scientific publication and/or presentation.