Individual study: The effects of mowing regime on meadow crane’s-bill Geranium pratense flowering and seed production on a roadside verge near Spaldwick, Cambridgeshire, England
Davis B.N.K. (1973) The effects of mowing on the meadow cranes-bill Geranium pratense L., and on the weevil Zacladus geranii (Payk.). Journal of Applied Ecology, 10, 747-759
In the UK, grassland road verges may act as a refuge for many plants. This study undertaken in southeast England, assessed the short-term effects of a range of cutting regimes on the performance of a moderately common herb, meadow crane’s-bill Geranium pratense.
A length of road verge (c.8 m wide x 170 m long) was selected along a main road (A604) near the village of Spaldwick (National Grid. ref. TL 138724) in west Cambrideshire. In May 1971, the verge was divided into three areas: a 70 m west section were crane’s-bill was moderately common to abundant; a central zone of about 60 m where crane’s-bill was uncommon; and a 40 m east section with Geranium common throughout.
To provide three blocks of six treatments, the west zone was divided into 12 compartments, and the eastern zone into six. Experimental plots (4 x 5 m) were located in each compartment. One set of compartments was left uncut (T1). The five treatments cut once or twice according to programmes that might be adopted by a highways department, were:
T2 - cut once 3 August ;
T3 - cut once in 24 August;
T4 - cut twice on 18 May and 13 July;
T5 - cut twice on 8 June and 3 August;
T6 - cut twice on 18 May and 28 June.
A rotary mower was used with the blades set at 10 cm, the approximate height of the cut of flail mowers (commonly used in verge management). The verge was mown in the middle of October, except those plots which had been cut at the end of August.
Crane’s-bill plant density was assessed on 31 May in 1 x 1 m quadrats divided into nine subplots, scoring presence (0 to 9) in each of 20 square metres per plot. Weekly estimates of peduncles, flowers, unripe and ripe capsules were made.
In addition, a small 'garden' was planted (48 plants in a grid spaced 0.5 m apart). The plants, obtained from the verge in May 1971, were grown at Monks Wood Experimental Station. They received the same treatments (but cutting undertaken using hand shears), allocated in a stratified random design and response was recorded during 1972.
An early cut (mid-May) removed most crane’s-bill leaves and the few earliest peduncles; consequently this had no measurable effect on flowering by July (compared to uncut areas). As late as 8 June, cutting had only a marginal detrimental effect upon flowering (apprent in one block only) as numbers of peduncles and flowers were comparable with those of uncut areas.
The end of June cut (regardless of with or without an earlier cut in May) was the most detrimental treatment, preventing almost all flowering. Cutting on 13 July coincided with peak flowering; a little flowering occured subsequently (end of August and beginning of September) but virtually no capsules ripened. Plots cut for the first time at the beginning of August resulted in little reduction in flowering but seed production was seriously reduced by the removal of about 75% of capsules before they were ripe. The single cut on 24 August had least effect as flowering had finished and the majority of capsules had ripened.
Results from the ‘garden’ experiment were inconclusive. Of the 48 plants, only 17 produced peduncles in 1972.
Conclusions: Mowing in May to early June had little or no effect upon flowering of Geranium pratense, whereas cutting at the end of June virtually eliminated flowering. Progressively later cutting allowed increasing amounts of flowering and ripening of seeds. The author suggests that vegetative reproduction obviates the need for much seed production each year, so mowing in early August would be satisfactory from a floristic conservation point of view.