Degree Name

Master of Science (MS)

Semester of Degree Completion

1978

Thesis Director

John E. Ebinger

Abstract

Cirsium altissimum, C. discolor and suspected hybrids used in this study were collected during September 1977 from three locations in east central Illinois. All were combined for subsequent taking of data because the study dealt with incidence of hybridization and morphological discriminators rather than degrees and directions of introgression. Ten possible discriminators were chosen by examination of herbarium specimens of parental plants: I, Involucre width; II, Involucre height; III, Pubescence of upper leaf surface; IV, Angle of departure of main veins from the midrib; V, Number of spines per cm along the margin; VI, Length of indurate portion of spine; VII, Leaf lobing percentage; VIII, Style length; IX, Corolla lobe length; X, Pollen viability. Parental groups were assigned by graphing pollen viability against leaf lobing. F-ratios calculated between groups showed significant differences at the 95% level in the cases of the remaining eight characters. Similarity of each individual to a hypothetical extreme set of character states was computed using a weighted matching coefficient (weight=F-1). Two non-overlapping parental curves and an intermediate curve of sterile plants were produced.

It was concluded that Cirsium altissimum and C. discolor were undergoing hybridization in Illinois. The areas of hybridization appear to be along interfaces created by clearing of trees from land previously conducive to the growth of C. altissimum. A sterility barrier apparently exists that prevents complete swamping in these areas. Field differentiation of the two species along these interfaces of apparent active hybridization can be accomplished by using leaf lobing, involucre width and height, angle of departure of the main veins from the midrib, and spine length. Forms in the field that appear intermediate may or may not be hybrids and group assignment must await the lab where morphological discriminators, weighted with the F-1 technique, can be combined with sterility data to give good group separation.

Download

Share

COinS