Material and Methods Plant MaterialIn 1995 cuttings of 16 L. alba populations from all over Guatemala (Table 1) were transplanted in an experimental field at ICTA (Instituto de Ciencia y Tecnologia Agricola), Cuyuta, Guatemala. The term 'collection source' in Table 1 specifies the origin of plant material, which was collected from house gardens ('cultivated') or from the wild. Cuttings were planted at distances of 75 × 50 cm in a randomized block design with three replications, 16 plants per replication.The decision to collect cuttings instead of seeds from wild populations derived from the extremely low germination rates found for L. alba of only 1-2%. Herbarium specimens of each population were deposited at the Universidad del Valle, Guatemala City, Guatemala. 90 days after transplantation, leaves and flowers were harvested and dried in a greenhouse at an average temperature of 32°C for 4 days. The temperature in the greenhouse was regulated by a ventilator. Morphological CharactersThree out of several morphological characters were especially interesting in the context of essential oil composition: leaf shape, width of the leaf, and length of the flower stem. The length of the flower stem, measured at the 4th node from the top, is defined as the distance between the axilla and the node of the involucre.

Three selfing and 13 crossing experiments between tetraploid individuals of Achillea ceretanica, Achillea collina, Achillea distans ssp.‘styriaca’ Saukel (ined.), and Achillea pratensis (Achillea millefolium complex, Compositae), five F1 crosses, three backcrosses and one further selfing experiment were carried out in order to study the inheritance of longipinenone (1) and its hydroxyl derivative (2). From these crossings, 1294 plants were studied by qualitative thin layer chromatography. Progenies from parent and F1 plants without longipinenones (0‐type, ll) uniformly contained none of these two sesquiterpenes. All other crossing experiments showed typical segregation patterns of 0‐type, L‐type (longipinenone (1) without hydroxylongipinenone, L.hh) and H‐type (hydroxy‐longipinenone (2) and occasionally longipinenone, L.H.) in the ratio of 1 : 1 and 1 : 3. According to these results both derivatives are under dominant genetic control regulated by genes L and H, whereby hydroxylation takes place after synthesis of longipinenone (1).