THE AGE STRUCTURE OF A HYBRID SWARM IN LIATRIS (COMPOSITAE).

Abstract

Interspecific hybridization has been the subject of continued interest for over a half century. The botanical literature contains hundreds of accounts of species and their natural hybrids. Whereas the morphology, cytology, fertility and chemistry of these entities have received great attention, essentially nothing is known about the life tables of herbaceous perennial hybrids and their parental species. The prime reason for the absence of these data is the difficulty in aging herbs. Moreover, even when it is possible, complete life tables may be impossible to compile because of inability to recognize hybrids before they flower. However, in those cases where plants may be aged but life tables are unattainable, an age census affords information about the age at which reproduction commences, mean adult age, longevity and generation span. Species of the genus Liatris (Compositae) form complex hybrid swarms and can be aged. Thus this genus is favorable for obtaining insight into life history parameters of species and hybrids. The purpose of this report is to compare the age structures of flowering L. aspera, L. cylindracea and L. spicata with their hybrids at Zion, Illinois. Natural hybridization among L. aspera Michx., L. cylindracea Michx., and L. spicata Willd. occurs in a 16 acre tract approximately 1 mile east of Zion, Lake County, Illinois. The population occupies an undulating sand prairie formed on the old beaches of Lake Chicago, the species being distributed in bands along moisture gradients created by a series of dunal ridges and inter-dunal depressions. Liatris aspera inhabits the drier upland sites, L. cylindracea the mid-slopes and L. spicata the moist depressions. Bands of pure species alternate with belts of species confluence, and it is in these belts that hybrids occur (Hadley and Levin, 1967). Each species has a distinctive inflorescence, and hybrids are readily discernable when in flower (Levin, 1968). The hybrids are vigorous and highly fertile, and include F. and advanced generation segregates (Levin, 1969). The perennating organ is a corm, which has conspicuous annual rings. Kerster (1968) has used this character to analyze the age structure of L. aspera at several sites in the Zion prairie. During August 1969, the corms of 100 flowering plants of each species were obtained over a range of sites in the Zion prairie. In addition, all of the plants which were conspicuous hybrids (F. hybrids or intermediate advanced generation segregates) were collected. The collections were limited to flowering specimens since hybrids could not be recognized from vegetative characters in seedlings or older non-flowering specimens. Radial sections of the corms were made in 2 planes, and then studied under a dissecting scope. The annual increment of growth appears as an unpigmented band of parenchyma contrasted with a ring of deeply pigmented cells laid down at the end of the growing season. Most of the corms were readily scored; the number of corms which were aged are as follows: L. spicata, 86; L. cylindracea, 100; L. aspera, 98; L. aspera X L. spicata, 46; L. aspera X L. cylindracea, 18; L. cylindracea X L. spicata, 26. The age structures of L. spicata, L. aspera, and L. cylindracea have several features in common (Fig. 1). The youngest flowering plants are about 7 years old and few individuals are greater than twenty years old. Most of the plants are in their early teens, and as such these age classes greatly influence the mean ages of the species which are as follows: L. spicata 13.08 years; L. cylindracea, 14.10 years; L. aspera, 16.15 years (Table 1). The significance of these differences was tested employing the Student-Newman-Keuls test (Sokal and Rohlf, 1969, p. 242). The difference between L. aspera and L. spicata is significant (P < .01), whereas other interspecific differences are not (Table 2). It is noteworthy that the species with the lowest mean age also has the lowest density per m2 (Hadley and Levin, 1967). In view of the relatively long pre-reproductive period, the age which plants attain is surprisingly low, only a small fraction of each species sample being above twenty years of age. The age structure of reproductive hybrids is depicted in Figure 1. The three hybrid types have similar age profiles, although that of L. cylindracea X L. spicata is flatter and somewhat less skewed than the other two. This feature may be due in part to yearly variations in hybridization, germination, and establishment, and to sampling error. The modal age classes are pre-teen and the mean ages of the hybrids are as follows: L. aspera X L. spicata, 11.02 years; L. spicata X L. cylindracea, 12.76 years; L. aspera X L. cylindracea, 13.05 years (Table 1). None of these differences are statistically significant (Table 2). To aid in the comparison of the age structures of the species and hybrids a composite histogram