Compute the following metrics:

MeanGermPercent

Mean/average germination percentage per unit time (\(\overline{GP}\)) (Czabator 1962) .

MeanGermNumber

Number of seeds germinated per unit time (\(\overline{N}\)) (Khamassi et al. 2013) .

MeanGermPercent(
  germinated.seeds,
  germ.counts,
  total.seeds,
  intervals,
  partial = TRUE
)

MeanGermNumber(germ.counts, intervals, partial = TRUE)

Arguments

germinated.seeds

Number of germinated seeds

germ.counts

Germination counts at each time interval. Can be partial or cumulative as specified in the argument partial.

total.seeds

Total number of seeds.

intervals

The time intervals.

partial

logical. If TRUE, germ.counts is considered as partial and if FALSE, it is considered as cumulative. Default is TRUE.

Value

The value of mean germination percentage as %

\(\mathrm{time^{-1}}\) or mean number of seeds per time interval as

\(\mathrm{count \, time^{-1}}\).

Details

Mean germination percentage per unit time (\(\overline{GP}\)) is computed as follows (Czabator 1962) .

\[\overline{GP} = \frac{GP}{T_{k}}\]

Where, \(GP\) is the final germination percentage, \(T_{k}\) is the time at the \(k\)th time interval, and \(k\) is the total number of time intervals required for final germination.

Mean number of seeds germinated per unit time (\(\overline{N}\)) is computed as follows (Khamassi et al. 2013) .

\[\overline{N} = \frac{N_{g}}{T_{k}}\]

Where, \(N_{g}\) is the number of germinated seeds at the end of the germination test, \(T_{k}\) is the time at the \(k\)th time interval, and \(k\) is the total number of time intervals required for final germination.

References

Czabator FJ (1962). “Germination value: An index combining speed and completeness of pine seed germination.” Forest Science, 8(4), 386--396.

Khamassi K, Harbaoui K, Jaime ATdS, Jeddi FB (2013). “Optimal germination temperature assessed by indices and models in field bean (Vicia faba L. var. minor).” Agriculturae Conspectus Scientificus, 78(2), 131--136.

See also

Examples

x <- c(0, 0, 0, 0, 4, 17, 10, 7, 1, 0, 1, 0, 0, 0)
y <- c(0, 0, 0, 0, 4, 21, 31, 38, 39, 39, 40, 40, 40, 40)
int <- 1:length(x)

# From partial germination counts
#----------------------------------------------------------------------------
MeanGermPercent(germ.counts = x, total.seeds = 50, intervals = int)
#> [1] 5.714286
MeanGermNumber(germ.counts = x, intervals = int)
#> [1] 2.857143

# From cumulative germination counts
#----------------------------------------------------------------------------
MeanGermPercent(germ.counts = y, total.seeds = 50, intervals = int, partial = FALSE)
#> [1] 5.714286
MeanGermNumber(germ.counts = y, intervals = int, partial = FALSE)
#> [1] 2.857143

# From number of germinated seeds
#----------------------------------------------------------------------------
MeanGermPercent(germinated.seeds = 40, total.seeds = 50, intervals = int)
#> [1] 5.714286