Plant analysis aims to either detect deficiency at the time of sampling (di
agnosis) or predict its occurrence at a later stage of growth (prognosis).
Its use is based on the presumption that the plant nutrient status will eit
her be constant with plant age or follow a predictable pattern of change ov
er time after sampling. However, a period of deficiency during plant growth
followed by the recovery of nutrient uptake to satisfactory rates may caus
e an irreversible impairment of growth which plant analysis fails to diagno
se or predict. Several cases are considered, each involving a temporary def
iciency of, or interruption to nutrient supply. Such cases generally involv
e but are not restricted to micronutrient deficiency. For example, B defici
ency impairs early seedling growth when seeds low in B are planted, even on
B fertilised soils. Low B concentration in seeds diagnoses the subsequent
impairment of seed germination or seedling establishment: however, leaf ana
lysis after emergence does not. Similarly, Zn deficiency impairs early grow
th of transplanted oilseed rape (Brassica napus L.) seedlings and eventuall
y depresses seed yield. However, leaf analysis during crop growth fails to
diagnose a Zn deficiency. Finally, temporary B deficiency induced by low va
pour pressure deficit or low soil water especially during reproductive deve
lopment may depress yield markedly but remain difficult to diagnose by plan
t analysis. Strategies for diagnosing and predicting such temporary deficie
ncies are discussed including the measurement of environmental parameters s
uch as pan evaporation or rainfall and their inclusion in multi-variate reg
ression models of plant response to nutrients.