Bernd F. Hägele-1 and Martine Rowell-Rahier-2
1-Zoologisches Institut der Universität Basel, Rheinsprung 9,
CH4051 Basel,Switzerland
2-Institut de Zoologie, Université de Neuchatel, Rue
Emile-Argand 11, CH-2007 Neuchatel, Switzerland
Genetic variation is the basis for evolutionary change. In order to know whether a population is capable of evolutionary;change when selection is acting, it is of fundamental importance to know if gonetic variation exists for the trait under investigation We were interested whether Adenostyles ailiariae and A. alpina(Asteraceae), two alpine species which both produce pyrrolizidine alkaloids but only A. alpine produces also sesquiterpenes, would harbour genetic variation for their leaf secondary compounds. Since under the carbon/nutrient balance hypothesis nitrogen and carbon based defense compounds are expected to react differently to a situation of low energy supply, we also imposed a light treatment on the clones
We planted clones of genets from different populations into a common garden and subjected them to a high and a low light treatment. After seven weeks, we harvested one leaf from every clone. After extraction of the dried material we determined leaf alkaloids by gas chromatography, sesquiterpenes by HPLC and nitrogen content with a CHN analyser. Leaf nitrogen was strongly increased in heavily shaded plants of both species. In A. alliariae there was no difference in leaf alkaloid content between the two treatments. In A. alpine alkaloid content tended to be higher in the heavily shaded treatment. One sesquiterpene, cacaloi- trimer was tn higher concentrations in the heavily shaded leaves whereas concentrations of the other sesquiterpene, cacalol, were reduced. Since the carbon/nutrient balance hypothesis would ha,ve predicted an increase in nitrogen based defenses and a decrease in carbon based defenses, these predictions were met with the exception of cacalol-trimer.
Broad sense heritabilities revested significant genetic variation in all secondary compounds. However genetic variability for the same compound could vary extremely between populations. We interpret this as an indication that populations experienced different selection pressures in the past.