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 effects of fertilizer addition on inflorescence and seed production of the protea, Leucospermum parile, Burgherspost Farm, Western Cape, South Africa

Published source details

Witkowski E.T.F. (1991) Nutrient limitation of inflorescence and seed production in Leucospermum parile (Proteaceae) in the Cape fynbos. Journal of Applied Ecology, 27, 148-158


The Malmesbury pincushion (Malmesburyluisie) Leucospermum parile (Proteaceae) is a monoecious evergreen shrub which regenerates from seed after fire. Concern has been expressed about overexploitation of proteaceous species flowers, which are largely harvested from wild. Due to the nutrient-poor soils of the fynbos habitat in which they occur, it has been suggested that flowering and reproduction of these shrubs are limited by mineral nutrients. Careful addition of fertilizer around selected plants might therefore be used to enhance flowering and seed production, and thus play a role in conservation of some heavily exploited species.

The effects of fertilizer addition of nitrogen, phosphorus and a mixture of all essential nutrients excluding N and P on production, dry mass, nitrogen and phosphorus contents of inflorescences and seeds L.parile were studied for 2 years.

Study site: The study area was the CSIR fynbos biome intensive study site at Pella on the Burgherspost Farm, 62 km north of Cape Town, south-western Cape Province (33º 3l'S, 18º 32'E), 15 km inland from the west coast. The study plants were growing in typical nutrient-poor fynbos.

Experimental design: A factorial fertilizer addition of nitrogen, phosphorus and a mixture of all essential nutrients excluding N and P, were applied to 10 x 5 m plots, arranged in a grid with 5-m wide strips separating each. Levels of addition (based on previous studies) were 5 g N/m² (as NH4N03) and 0.5 g P/m² (as Ca3 (Po4) 2). Fertilizer was applied during 15-17 September 1984. Each treatment was replicated four times except the control which was replicated eight times.

Plant sampling: L.parile plants were c. 3 years old at the start of the study in September 1984. During November-December 1984, the number of inflorescences per study shrub was counted, and eight mature inflorescences, on average, were harvested from 59 plants. In November-December 1985, inflorescences on two shrubs per plot were counted and 10 harvested from one. Further sampling was not possible as vegetation was destroyed by a fire. Seeds were removed from the inflorescences and weighed. Plump (assumed viable) and damaged (assumed predated or diseased) seeds were counted separately. The plump seeds from each nutrient treatment combination were pooled for nutrient analyses. The number of flower buds per shoot was counted on 10 shoots, randomly chosen from one plant per plot in June 1985 and 1986. Canopy volumes and areas were also determined during November-December 1984 and 1985.

Nutrient analyses: All plant material was oven-dried at 80 ºC for 48 h, weighed, and ground for analysis to determine total N and total P.

Flowering: In spring 1984, 122 (42%) of the 293 L.parile shrubs in the 36 plots flowered, while 94% flowered in 1985. The percentage flowering in 1984 was unlikely to have been influenced by the nutrient additions because flowering had commenced before the nutrients were applied.

Infloresence production: Nitrogen addition however, reduced inflorescence number per shrub in the first year, but increased production in the second year. The mean number of inflorescence buds per shoot was not significantly different between treatments in 1985 and 1986, although tending to be higher and more variable with N addition than other treatments. The addition of P alone tended to increase inflorescence production but not significantly so.

Seed production: The percentage of barren inflorescences produced in 1984 in response to addition of N alone was 69% compared with only 23% in control plants. The percentage of barren inflorescences varied between 82% and 92% for all treatments in 1985. The number of plump (viable) seeds recovered per inflorescence varied from 0 to 5 (average 1.5) for control plants. It was greatest for plants amended with only P and lowest for plants amended with only N, but was not significantly so between treatments.

Seed viability and seed mass: Only 0.14 plump seeds per inflorescence were recovered in November-December 1985 compared with 1.19 in 1984. Inflorescences matured earlier in 1985 than in 1984, and had probably released many seeds before harvesting, thus no comparisons can really be made.

Seed dry mass per inflorescence was reduced by N addition in 1984, particularly by N alone. Dry mass per seed ranged from 9 to 77 mg for control plants.

Conclusions: This study shows that N addition in 1984 reduced inflorescence production in L.parile and increased vegetative growth (4.4 and 3.6 g/shoot in N-amended and control plants, respectively) and N storage in shoots (35 and 23 mg N/shoot respectively). In contrast plants in plots with N fertilizer added had significantly increased inflorescence production in 1985 compared with the other treatments. This may have resulted from the allocation of stored N to increased inflorescence production in 1985. At the end of the 1985-86 growing season, N concentrations of shoots of N-amended plants were no longer elevated, indicating that N reserves from the previous year had been exhausted.

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