Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Individual study: The effect of timing of umbel and complete stem removal on regeneration of giant hogweed Heracleum mantegazzianum, in the Slavkovský les Protected Landscape Area, Bohemia, Czech Republic

Published source details

Pysek P., Krinke L., Jarosik V., Perglova I., Pergl J. & Moracova L. (2007) Timing and extent of tissue removal affect reproduction characteristics of an invasive species. Biological Invasions, 9, 335-351


Giant hogweed Heracleum mantegazzianum (a native of southwest Asia), has become invasive in many European countries. Response of giant hogweed to experimental removal of plant parts was studied in the Czech Republic. The study assessed how plants respond, in terms of quantity and quality of produced fruit, to the removal of different amounts of flowering umbels and/or vegetative (i.e. leaves/stem) organs (for a summary see:; and if regeneration ability depends on the time of treatment (summarised here).

Study area: The study was undertaken in the Slavkovský les Protected Landscape Area in the west of the Czech Republic.

Treatments: Based on the results of an experiment conducted in 2002 (see:, only treatments removing all flowers (which led to substantial reduction of fruit set) were applied. In June 2003, 70 plants were randomly selected.

On 7 June 2003 (Time 1, terminal umbels at the bud stage), 10 plants were allocated as controls, 10 were cut at the stem base 5 cm above ground; and 10 plants were cut just above the leaf rosette. Treatments were similarly applied on 20 June 2003 (Time 2, beginning of flowering of the terminal umbel), and on 2 July 2003 (Time 3, beginning of fruit development on the terminal umbel, other umbels still flowering); on these dates, another 10 plants were subject to each of the two cutting treatments. Terminal umbels cut on 2 July were left on the ground to see if the fruits ripened.

Plant responses: In August and September fruits were harvested. The following were recorded for each plant: average fruit weight (calculated from the weight of 25 fruits); total fruit weight; total number of fruits; germination time of each seed; and germination percentage.

To assess if treatment effect was related to plant vigour, stem basal diameter and plant height were recorded for each plant on 7 June.

Of the 60 treated plants, 53 (88%) regenerated, producing on average 1,425 fruits. The percentage of plants which produced fruit and number of fruits produced was: 90% and 3,310 at Time 1; 95% and 461 at Time 2; and 80% and 446 at Time 3. More vigerous (larger) plants regenerated better and produced more fruit than smaller ones.

Total fruit weight of controls was c.4.8 g; total fruit weight on regenerating plants was affected by treatment and application time. At Time 1, total fruit weight of plants with rosette retained was c.3.4g, and with the rosette removed c.2.4g. At Time 2, plants with the rosette retained (c. 1.8g) did not significantly differ from that of plants with the rosette removed at Time 1. Whole stem removal at Times 2 (c.1.0g) and 3 (c.0.8g) resulted in greatest decreases; at Time 3, even for plants with the rosette retained, fruit weight was low (c.0.9g).

In general, the effect of treatments on the number of fruits was greater when applied later in the summer. However, at a particular application time, the two cutting treatments were never significantly different in total fruit weight, but there was a tendency for plants with rosettes to produce heavier fruits. Average germination of seeds of cut plants (63%) was only a little less than controls (78%).

At Time 3, fruits developed on the terminal umbels of 17 (85%). Average fruit number (1,840) and weight (5.7 mg) were much lower than in controls (9,863 fruits and 13.1 mg, respectively). Importantly in terms of management, 24% of seeds from cut umbels left on the ground germinated, thus these umbels should be destroyed.

Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: