Resource allocation in relation to leaf retention time of the wintergreen Rhododendron lapponicum.
Leaf longevity in plants can vary from weeks in the most rapidly growing species to years in evergreen species and even as long as a decade (Ewers and Schmid 1981, Chabot and Hicks 1982). The ecological significance of differences in leaf longevity has been discussed frequently (e.g., Small 1972, Jonasson 1989, Reich et al. 1992, and review by Chabot and Hicks 1982) and many authors have suggested that extended leaf life-span and evergreenness is an adaptation among plants of nutrient deficient areas to conserve nutrients. For instance, Small (1972) reported that evergreen bog plants resorbed a higher percentage of leaf nutrients before leaf abscission than deciduous species growing in the same bog. The high resorption, in combination with the prolonged life-span, yielded a long residence time of nutrients within the evergreen plants. As a consequence, they assimilated more carbon per unit of invested nutrients than the deciduous species. Long leaf longevity and evergreenness could, hence, be an efficient means to reduce nutrient limitation in soils with low nutrient availability (Small 1972, Schlesinger and Chabot 1977, Chapin 1980, Chabot and Hicks 1982, Aerts and van der Peil 1993). This is indirectly supported by observations that evergreenness increases as soil nutrient availability declines (Beadle 1954, Loveless 1961, Monk 1966, Small 1972, Goldberg 1982).
Other studies have, on the contrary, given little support to these interpretations. For examples, Nambiar and Fife (1987) concluded that nutrient resorption from needles of radiata pine (Pinus radiata) was not a mechanism for coping with low soil fertility, Greenway et al. (1992) found no support for the idea that retention of needles in black spruce (Picea mariana) is an adaptation to nutrient stress, and Chapin et al. (1980) could not observe any difference in nutrient use efficiency (carbon gain per unit invested nutrient) between deciduous and evergreen plants.
I showed in a previous study (Jonasson 1989) that removal of old leaves of a number of evergreen and wintergreen species (i.e., plants that retain the leaves over one winter only; Bell and Bliss 1977) did not decrease the allocation of nutrients to new leaves during the subsequent growing period, although old leaves are believed to be a main storage site of nutrients during dormant periods (Reader 1978, Chapin 1980, Shaver 1981). The nutrients allocated to the new growth of defoliated plants must, therefore, have been supplied either from compensatory transport of mobile reserves in stems or roots, or from soil nutrient uptake. In the last case, the observation casts doubt over nutrient conservation as a proximate explanation for long leaf longevity because the hypothesis implies that lost nutrients are hard to compensate for by new uptake (Berendse and Aerts 1987, Berendse and Jonasson 1992, Aerts and van der Peil 1993).
Generally, defoliation did, however, retard growth of the new leaves during their first weeks of development (Jonasson 1989), suggesting that the long retention time principally served to prolong the photosynthetic period and increase the leaves' lifetime gain of carbon (Chabot and Hicks 1982). It can be argued that an extension of leaf lifetime throughout early summer, as that in wintergreen plants, is particularly important, at least at high latitudes, because it allows carbon fixation during the period when light is most abundant (Tieszen 1978, Chapin 1983).
Here I report results of further tests of the role of overwintering leaves as suppliers of carbon and nutrients to new shoots. These tests were elaborated by a suite of three independent, although closely related, studies (numbered 1-3 below) that took place between 1988 and 1990, with focus on one of the previously examined species, the wintergreen arctic-alpine dwarf shrub Rhododendron lapponicum.
(1) In the first study, I analyzed the growth of, and nutrient allocation to, new shoots of R. lapponicum after early spring defoliation of leaves retained over winter. The results are compared with responses to late spring defoliation presented here and with earlier results (Jonasson 1989). Both the early and late spring defoliations were done before the leaves started to senesce and export nutrients, which usually takes place in July, following budbreak of the new leaf buds in late June (Jonasson 1989). Carbon shortage due to the defoliation should cause a greater reduction of shoot growth after early defoliation than after late defoliation. If nutrient limitation due to removal of leaf nutrient reserves has stronger effects on growth than does carbon shortage, early and late defoliation should cause equal reduction of growth. In addition, leaf removal should cause lower nutrient concentrations in defoliated than in undefoliated plants. This should probably be more pronounced in old stems and roots than in shoots as it appears that the canopy retains more stable nutrient concentrations than the woody tissues (Adams et al. 1987, Jones and Dighton 1993). Further, defoliated plants should accumulate fewer nutrients in new shoots and woody tissues than do undefoliated plants.
(2) In a second study, I compared growth and nutrient allocation the year after flowering in specimens with high and low flowering frequency. Flowering usually starts in the end of May or beginning of June, two to three weeks before leaf budbreak (S. Jonasson, personal observation). R. lapponicum develops either reproductive or vegetative shoots; thus strongly flowering plants will have a reduced canopy until leaf budbreak the next year. Carbon limitation due to low leaf biomass and leaf area in strongly flowering plants should reduce shoot production the following year. Nutrient shortage due to low storage capacity of mobile nutrients in the reduced canopy would lead to low stem and root nutrient concentrations and reduced nutrient pool sizes in shoots and woody tissues of strongly flowering plants compared to plants with few flowers the previous year. Because the natural loss of leaf biomass after flowering should have similar effects on leaf area and leaf nutrient pools as leaf loss after defoliation, the expected responses one year after flowering should be similar to those following defoliation.
(3) In a third study, I removed 1-yr-old leaves and the current year's buds on two sets of plants at the time of budbreak. The complete plants, including stems and roots, were harvested 1 mo later and analyzed for treatment effects on growth and nutrient allocation patterns. This study gives another set of data to that of Jonasson (1989) on treatment effects of leaf removal in late spring that I compare with the responses to early spring defoliation (see study 1 above). The main aim of the study was, however, to collect data on shoot, stem, and root biomass and nutrient content at the time of bud-break and at the end of the period of rapid growth 1 mo later, so that growth and nutrient allocations between the component parts of the plants within the treatment groups could be quantified. These data were used to construct nutrient budgets - valid for the 1st mo of rapid shoot growth in spring - for plants with and without access to nutrient stores in 1-yr-old leaves and for plants lacking the strong sink of new shoots.
Field experiments and laboratory methods
The studies took place on stabilized polygon patterned ground on open heath areas at Abisko, North Sweden (68 [degrees] 21[minutes] N and 18 [degrees] 49[minutes] E). Soil characteristics of the sites are described previously (sites AG 1 and AG 2 in Jonasson 1986).
Growth and allocation responses to early spring defoliation were analyzed on an annual basis in plants defoliated in early June 1988 and harvested in June and July 1989 at site AG 2. Within an area of [approximately equal to] 10 x 10 m, 16 of 32 selected, equally sized plants, were randomly chosen and defoliated on 2 June 1988, [approximately equal to]3 wk before the new leaf buds opened, and soon after the ground became free from snow. The remaining 16 plants were marked and used as controls. On 11 June and 28 July the next year, eight randomly selected plants of those previously defoliated, and another eight plants of the undefoliated controls were excavated and separated into last year's leaves (C + 1-leaves), current year's leaves (C-leaves; in July only), old stems, current stems, and roots. The leaf area was measured and the plant parts were dried at 70 [degrees] C and weighed. A 100 mg subsample of the plant material was digested in a mixture of sulphuric and selenous acid (Kedrowski 1983) and analyzed for nitrogen (N) by the salicylate method, for phosphorus (P) by the molybdate method, and for potassium (K), magnesium (Mg) and calcium (Ca) by atomic absorption spectrophotometry.
The treatment effects on growth and nutrient allocation in the cohort formed the year of defoliation (1988), i.e., the C + 1-cohort in 1989, were analyzed on samples harvested in June 1989 before the current year's growth started. Hence, this cohort had been functioning during one entire growing season lasting from late June to late August/early September 1988 and thereafter had been dormant over the winter. The longer term treatment effects on the second leaf cohort developed after defoliation in 1988 were analyzed on the C-leaves harvested in July 1989.
The reproductive effects on growth and nutrition were examined in 1988-1989. Sampling took place at a site close to AG 1 where the flowering frequency was exceptionally high in 1988, so that sufficient plant material of flowering and non-flowering plants of approximately equal size was available. From two groups of 50 plants each, one with flowers on 50-100% of all branches and the other with flowers on 0-25% of all branches in 1988, eight plants were randomly chosen and harvested on 23 July 1989, after which they were separated into components and analyzed as described above.
The late spring defoliation and bud removal was done on 25 June 1990. The experiment took place on AG 2, [approximately equal to]500 m from the area where defoliation was done in 1988. At the time of defoliation, the new leaf buds had just started to open, but individual leaves could not yet be distinguished. Three groups of eight plants each were randomly selected from 24 previously chosen plants of approximately similar sizes. All 1-yr-old leaves were removed from the plants of one of the groups, selected randomly. All current leaf buds were removed from the plants of a second group (hereafter called "de-budded") whereas the third group was left unmanipulated and served as a control.
On 25 July, the whole plants were excavated with the entire root system intact, separated into fractions, measured for leaf area, dried, weighed, and analyzed for nutrients as above.
This experiment was designed to: (1) supply data for construction of a nutrient budget for plants with and without access to nutrient stores in 1-yr-old leaves, and (2) to test whether the responses to removal of old leaves at the time of budbreak were different from those after the early spring defoliation. Note that the treatment effects of late spring defoliation were analyzed 1 mo after defoliation, whereas the plants defoliated in early spring were allowed another mo of growth, plus the subsequent dormancy period, before the effects were analyzed.
Calculations of resource allocation and nutrient budgets
The nutrient budgets of the treatment groups, valid for the 1st mo after budbreak, were constructed by combining data of shoot, stem, and root allocation patterns of biomass and nutrients in the defoliated, de-budded, and control plants. The analyses of allocation within the shoots, calculated on a whole plant basis, include estimations of biomass and nutrient accumulation in current year's shoots of controls and defoliated plants and changes in nutrient pools in 1-yr-old leaves of controls and de-budded plants. For the woody mass, the calculations include quantification of compensatory nutrient transport from stems and roots in defoliated plants and nutrient accumulation in stems of de-budded plants.
Biomass and nutrient increase in current year's shoots. - The biomass increase in current year's shoots ([Delta][B.sub.c]), including new leaves and stems of control and defoliated plants was calculated as:
[Delta][B.sub.c] = [B.sub.July] - [B.sub.June],
where [B.sub.July] represents the observed mean dry mass per plant of the current year's shoots of target (control and defoliated) plants at the harvest on 25 July. [B.sub.June] is an estimate of the average bud dry mass of the target plants on 25 June derived from [B.sub.Bud][center dot]c, where [B.sub.Bud] represents the observed mean dry mass of buds of de-budded plants on 25 June and c equals the ratio of mean C + 1 leaf area of target plants to mean C + 1 area of de-budded plants. Hence, c is a correction factor that takes possible differences of mean bud masses between the target groups and de-budded plants into account. The correction is based on the assumption that the leaf area is proportional to the leaf mass and that the ratio of C + 1 leaf area and C leaf area of the shoots is approximately equal from one year to the next. The validity of the assumption was supported by a correlation coefficient of r = 0.89 (P = 0.003) between the area of C + 1 and C leaves of the eight control plants.
The nutrient increase ([Delta][N.sub.c]) was calculated similarly, by using nutrient contents (N) instead of biomass values (B), i.e.:
[Delta][N.sub.c] = [N.sub.July] - [N.sub.June],
where [N.sub.July] represents observed nutrient pools in harvested current shoots of controls and defoliated plants. [N.sub.June] is an estimate of their nutrient content at budbreak calculated as the product of [B.sub.June] and the observed nutrient concentration in the buds removed on 25 June.
Changes of mass and nutrient pool sizes in C + 1 leaves. - Changes of mass ([Delta][B.sub.c+1]) and nutrient pools ([Delta][N.sub.c+1]) in C + 1 leaves of control and de-budded plants, were calculated as:
[Delta][B.sub.c+1] = ([LMA.sub.June] - [LMA.sub.July])[center dot]LA
[Delta][N.sub.c+1] = ([LNA.sub.June] - [LNA.sub.July])[center dot]LA,
with LMA and LNA representing the observed leaf mass and nutrient content per unit leaf area, respectively, of defoliated plants in June and of the control and de-budded plants in July, and with LA representing the total C + 1 leaf area of the plants.
Observed losses of mass could be due to both respiratory carbon losses, leaching, and resorption. For nutrients, it was assumed that the losses represented resorption because leaching losses generally are small in evergreen plants (e.g., Chapin et al. 1980, Jonasson and Chapin 1985). Hence, the amount of nutrients lost from C + 1 leaves in control plants could potentially contribute to the nutrient increase ([Delta][N.sub.c]) in current year's shoots. The remaining amount, i.e., [Delta][N.sub.c] - [Delta][N.sub.c+1], and the whole increase in current year's shoots of defoliated plants must have been supplied from reserves in stems and roots or from nutrient uptake.
Changes in stem nutrient content. - Possible compensatory transport from stems and roots to current year's buds in defoliated plants was quantified by comparing the observed nutrient pools in the woody plant parts of controls and defoliated plants on 25 July. The changes of nutrient content in stems and roots of de-budded plants that were likely to take place as a consequence of accumulation of resorbed nutrients from C + 1 leaves in the woody tissues after removal of the nutrient sinks in current leaves were quantified in a similar way by comparing the pool sizes in woody parts of de-budded and control plants.
Treatment effects of early defoliation
Responses of shoots developed the year of defoliation [ILLUSTRATION FOR FIGURE 1 AND 2 OMITTED]. - The early defoliation on 2 June in 1988 significantly reduced the growth of the leaf cohort that developed during the year of defoliation i.e., the C + 1 leaves of the year of harvest (Fig. 1a; P = 0.02, t test). On 11 June the year after defoliation, these leaves had 29% less dry mass and smaller N and Ca pools [ILLUSTRATION FOR FIGURE 1a, b, e OMITTED], but nonsignificantly less P, K, and Mg [ILLUSTRATION FOR FIGURE 1c, d, f OMITTED] than the undefoliated controls. The leaf mass and the nutrient content per unit leaf area (data not shown) did not differ between the defoliated plants and the controls, except for P which was significantly higher in defoliated plants, Defoliated plants also had significantly higher P and K concentrations than the controls [ILLUSTRATION FOR FIGURE 2b, c OMITTED]. Hence, defoliation decreased the growth of the new leaf cohort. It decreased, or did not affect, the leaf nutrient pool sizes, and it increased, or did not affect, the nutrient concentrations.
Responses of shoots developed the year after defoliation [ILLUSTRATION FOR FIGURE 1 AND 2 OMITTED]. - The leaf nutrient concentrations [ILLUSTRATION FOR FIGURE 2 OMITTED], dry mass [ILLUSTRATION FOR FIGURE 1a OMITTED], and the nutrient content per unit leaf area (data not shown) of the next leaf cohort, i.e., the second leaf generation after defoliation, did not differ significantly between controls and defoliated plants, with one exception: the concentration of P, but not the P content per unit leaf area, was slightly lower in the defoliated plants (1.6 mg/g compared to 1.8 mg/g in the controls; P = 0.02). The mean dry mass per shoot was nonsignificantly lower (P = 0.32; t test) by 12% in defoliated plants than in controls [ILLUSTRATION FOR FIGURE 1a OMITTED]. Since the nutrient concentrations were comparable, the trend for consistently lower shoot nutrient pool sizes in defoliated plants [ILLUSTRATION FOR FIGURE 1b-f OMITTED] was a function of their lower dry mass per shoot and not a function of differences in leaf nutrient status.
Treatment effects on stem nutrients [ILLUSTRATION FOR FIGURE 3 OMITTED]. - The June concentration of N in old stems [ILLUSTRATION FOR FIGURE 3a OMITTED], i.e., the concentration before growth of the second leaf generation after defoliation began, was significantly higher, and the concentration of Ca [ILLUSTRATION FOR FIGURE 3d OMITTED] was significantly lower in defoliated plants than in controls. The N level declined during the subsequent month so that the concentration approached that in controls in July (data not shown). The July concentration of Ca, on the other hand, increased to significantly higher levels than in controls, but the levels fluctuated widely among individual plants.
The stem nutrient concentrations in controls were remarkably similar between 11 June and 28 July [ILLUSTRATION FOR FIGURE 4 OMITTED]. Also, the July levels of nutrients in defoliated plants and controls were similar, except for the high mean Ca level.
Root nutrient concentrations followed the same general pattern as the stem concentration. Because the root system of some plants could not be excavated entirely, the data are not shown here: it appeared that the concentrations were confounded by the unequal proportion of woody belowground stems and fine roots in the samples.
Leaf mass, leaf and stem nutrient status in post-flowering R. lapponicum
Treatment effects on leaf mass and shoot nutrients [ILLUSTRATION FOR FIGURE 1 and 2 OMITTED]. - The dry mass of the shoot [ILLUSTRATION FOR FIGURE 1a OMITTED] and the leaf dry mass per unit area (data not shown) of the current year's shoots in 1989 were significantly (P = 0.003 and 0.005, respectively; t test) lower in plants with high flowering frequency the previous year (i.e., plants with fewer 1-yr-old leaves) than in plants with low flowering frequency.
On the contrary, leaf nutrient concentrations were similar [ILLUSTRATION FOR FIGURE 2 OMITTED] except that the P level [ILLUSTRATION FOR FIGURE 2b OMITTED] was significantly higher, and Mg [ILLUSTRATION FOR FIGURE 2e OMITTED] significantly lower in plants with high flowering frequency in 1988. The differences were due to a very low variance among the replicates rather than to any marked effect of differences in flowering frequency.
Although the nutrient concentrations were similar in the two groups, the differences in shoot dry mass yielded significantly lower pools of nutrients in the shoots of strongly flowering plants than in plants with low flowering frequency [ILLUSTRATION FOR FIGURE 1b-e OMITTED].
Treatment effects on stem nutrients [ILLUSTRATION FOR FIGURE 3 OMITTED]. - Stems of abundantly flowering plants had slightly, although significantly, lower concentrations of N, and considerably lower concentrations of K the year after flowering than plants with low flowering [ILLUSTRATION FOR FIGURE 3a, c OMITTED].
Treatment effects of late defoliation
Treatment effects on current year's shoots [ILLUSTRATION FOR FIGURE 1 and 2 OMITTED]. - In contrast to the early spring defoliation, the late spring defoliation did not affect the development of the new shoots or the nutrient allocation to them. This is remarkable, because the time between bud-break/defoliation and harvest was only 1 mo, whereas the plants defoliated in early spring, which responded by a decline of leaf growth, had another month at the end of the growing period to compensate for the losses in removed leaves. Further, in plants defoliated at the time of budbreak, the average nutrient concentrations [ILLUSTRATION FOR FIGURE 2 OMITTED] were in fact consistently higher in new leaves of the defoliated plants than in controls 1 mo after defoliation, and significantly so for P and Mg. The leaf mass per unit area (data not shown) declined slightly (P = 0.08; t test), from 106 g/[m.sup.2] in controls to 99 g/[m.sup.2] in defoliated plants. The higher nutrient concentrations in defoliated plants compensated for the lower mass per unit area, so that the content per shoot of all elements was consistently higher than in controls [ILLUSTRATION FOR FIGURE 1b-f OMITTED], although in no case significantly different for individual elements.
Treatment effects on stems and roots [ILLUSTRATION FOR FIGURE 3 OMITTED]. - As with the leaf nutrient concentrations, the nutrient concentrations in old stems and roots [ILLUSTRATION FOR FIGURE 3 OMITTED] were generally lower in controls than in defoliated plants, although significantly so only for N [ILLUSTRATION FOR FIGURE 3a OMITTED]. A part of the lower nutrient concentration in the woody tissues of controls could, potentially, have been due to a dilution effect caused by accumulation of resorbed organic constituents from the old leaves, transport that was prevented in defoliated plants.
The controls also had significantly lower stem and root concentrations of N and P, i.e., the most mobile nutrients, than the de-budded plants (Figs. 3a, b; cross-hatched bars), but the concentration of the less mobile or immobile K, Ca, and Mg did not show any significant treatment effect [ILLUSTRATION FOR FIGURE 3c-e OMITTED]. A large part of the difference in N and P content could be accounted for by accumulation of the pools resorbed from the old leaves of de-budded plants that could not be transported further after the removal of the current year's bud.
Allocation patterns and nutrient budgets
After standardization to equal plant size, control and defoliated plants allocated similar amounts of nutrients to current year's shoots during the 1st mo after bud-break, and de-budded plants, which lacked nutrient sinks in new shoots, accumulated extra nutrients in stems and roots [ILLUSTRATION FOR FIGURE 5 OMITTED]. These are expected results, given the observed small treatment effects of late spring defoliation on shoot, stem, and root nutrient content reported for unstandardized plants in the sections above and in Jonasson (1989).
The pools resorbed from C + 1 leaves between 25 June and 25 July correspond to 91, 76, and 39% of the increases in N, P, and Mg, respectively, that took place in the new shoots of controls during the same period of time, whereas resorbed K and Ca contributed insignificant amounts [ILLUSTRATION FOR FIGURE 5 OMITTED].
Stems and roots of defoliated plants had similar (for P), or even significantly higher (for N), concentrations, the two most mobile nutrients [ILLUSTRATION FOR FIGURE 3 OMITTED], i.e., nutrient pool sizes and nutritional status in the woody parts of defoliated plants did not decline after defoliation.
The late spring defoliation on 25 June 1990 caused an estimated loss of 15% of the biomass and between 21% (N) and 33% (K) of the nutrient pool in the whole plant (Table 1). Of the removed pool, translocation potentially could have mobilized 37% of the mass, 53% of the nitrogen, 36% of the phosphorus, and 15% of the magnesium, whereas K and Ca were not mobile during the 1st mo of leaf development (Table 2).
The translocation from leaves of de-budded plants was in general lower, which is a natural consequence of the reduction of the sink strength for nutrients caused by the treatment. Calcium, in contrast to the other elements, increased by 20% in the old leaves (Table 2). In the absence of sinks in new shoots, the translocated nutrients accumulated in stems and roots [ILLUSTRATION FOR FIGURE 3 and 5 OMITTED].
The pattern of pronounced loss of mass and high resorption of N and P from old leaves, a smaller resorption of Mg, the lack of mobility of Ca and K (which usually is mobile; Berendse and Jonasson 1992), agrees with earlier measurements of leaf nutrient mobility during the entire period of leaf senescence in R. lapponicum (Jonasson 1989). Comparisons with these earlier data indicate that resorption from leaves between 25 June and 25 July, i.e., during the 1st mo of shoot growth, resulted in removal of [approximately equal to]75% of the mobile N pool, 60% of the mobile P, and the entire pool of mobile Mg in old leaves. Also, the decrease of mass of old leaves was close to the earlier measured decline of [approximately equal to]35% in R. lapponicum during the entire phase of leaf senescence (Jonasson 1989) and recent data of 30-36% resorption reported by Karlsson (1994). This mass loss is high in comparison to the most commonly observed loss of [approximately equal to]10% in senescing leaves (Chapin et al. 1990) and indicates that retranslocation of organic compounds from the leaves could contribute substantially to the carbon balance of the current year's shoots in early spring.
TABLE 1. Estimated distribution of mass and nutrients in the current year's leaf buds (C-buds), 1-yr-old leaves (C + 1 leaves), stems, and roots of Rhododendron lapponicum defoliated on June 25, at the onset of the growing season.
C+1 Component C-buds leaves Stems Roots % of total
Dry mass 5.3 15.1 49.8 29.8 N 15.0 21.0 41.1 22.9 P 20.4 24.4 33.6 21.6 K 19.7 33.3 27.0 20.0 Ca 6.6 31.2 37.9 24.3 Mg 8.0 25.1 31.1 35.8
The high proportions of the nutrient accumulation in current year's shoots of controls that could be accounted by transport from senescing C + 1 leaves [ILLUSTRATION FOR FIGURE 5 OMITTED] show that the supply of nutrients from stores in stems and roots, or from nutrient uptake, could be only on the order of 10% of the N and 25% of the P accumulation. How these amounts are partitioned between transport from stores in the woody plant parts and up-take cannot be estimated because there are no data available on stem plus root pool sizes in June that can be compared with the known pools in July. The stable stem nutrient concentrations between 11 June and 28 July in the experiment with early season's defoliation [ILLUSTRATION FOR FIGURE 4 OMITTED] give a "hint," however, that woody tissues were not significant suppliers of nutrients to the developing shoots.
[TABULAR DATA FOR TABLE 2 OMITTED]
The similar amount of nutrients incorporated in defoliated plants and in controls [ILLUSTRATION FOR FIGURE 1b-f and 5 OMITTED] shows that defoliated plants can compensate for lost nutrient sources in old leaves. The increase in nutrient pools that took place in the current year's shoot of the defoliated plants between June and July corresponds to a large part of the plants' total nutrient pool (cf. Table 1). Any extra transport of nutrients from stems and roots to the developing new shoots as a compensation for lost sources of leaf nutrients should, therefore, have resulted in a substantial decline of stem plus root nutrient concentrations in comparison to that in controls. Such a decline was not observed, which gives no support to the hypothesis that defoliated plants can compensate for the lost ability to allocate nutrients from the old leaves to the developing shoots by drawing extra nutrients from stems and roots. The implication is that the defoliated plants received the largest amounts of nutrients from uptake, with amounts corresponding to the sum of uptake plus export from senescing leaves in control plants.
The large import of nutrients to new shoots of defoliated plants and their lack of compensatory transport of nutrients from stems and roots imply that lost nutrients can be compensated for by an increase in nutrient uptake. This does not support the hypothesis that the retention of leaves over the winter in wintergreen plants is an important adaptation to conserve nutrients (Small 1972), or to minimize the annual nutrient uptake under environmental conditions of low soil nutrient availability. The results do not, however, preclude the possibility that nutrient translocation from senescing leaves normally contributes a large fraction of the nutrient pool that is annually incorporated in the new growth, even though the new growth is not dependent on this transport.
Instead of a major adaptation to nutrient deficiency, it appears that the retention of leaves over the winter in R. lapponicum serves to support the new shoots with carbon and guarantee a high growth rate of the current year's shoots. This conclusion is based, first, on the lack of growth decline after late spring defoliation but an observed 29% reduction of the annual growth of new shoots that took place after early spring. defoliation [ILLUSTRATION FOR FIGURE 1a OMITTED], an effect that tended to persist during the 2nd yr after defoliation, although at a lower level [ILLUSTRATION FOR FIGURE 1a OMITTED]. This difference in shoot growth agrees with the predicted response to carbon shortage presented in the Introduction, and occurred in spite of the longer period of time early defoliated plants were given for compensation of losses in removed leaves between the time of budbreak and harvest. Second, plants with a high proportion of reproductive branches one year, and consequently a reduced amount of 1-yr-old leaves the next year, formed much less shoot biomass the year after flowering than did plants with low flowering frequency [ILLUSTRATION FOR FIGURE 1a OMITTED]. Hence, growth the year after abundant flowering was basically similar to the responses to early defoliation but dissimilar to lack of growth responses to late spring defoliation (Fig. 1a; Jonasson 1989). These similar responses to early defoliation and flowering were expected because both result in a reduced potential for carbon fixation as compared to late defoliation and low flowering frequency. It cannot, however, be excluded that flowering also imposed other effects on the subsequent shoot growth than reduced leaf area.
Karlsson (1994) showed that shading 1-yr-old R. lapponicum leaves of non-reproductive plants in early spring (6 June) causes a reduction (although not significant - but see Jonasson [in press]) of [approximately equal to]16-17% in the growth of current year's shoots, i.e., somewhat lower than the 29% reported here after early spring defoliation. Defoliation at the same date resulted in a further, pronounced decline of growth. Although interpreted differently by Karlsson, the extra decline of growth after defoliation suggests that translocation of unusually high carbon reserves in old leaves of this species could add substantial amounts of carbon to support the early season's expansion of the leaf bud, in addition to the carbon supply from early season's photosynthesis.
It could be argued that the reduced growth is an effect of the observed reduced pool size of nutrients allocated to new shoots after flowering and early defoliation [ILLUSTRATION FOR FIGURE 1 OMITTED] and not caused by carbon limitation (cf. Karlsson 1994). This possibility is unlikely, however, because leaf nutrient concentrations increased after defoliation, indicating that the leaf nutrient status did not cause the growth reduction.
Different responses in canopy growth and reproduction after defoliation at different phenological stages have been reported earlier (Marquis 1992). Further, previously reported responses of reduced growth and nutrient accumulation to early spring defoliation of another wintergreen species, Ledum palustre, were similar to those of the early defoliated R. lapponicum reported here (Jonasson 1989). Also, the lack of responses to defoliation at the time of budbreak of two other wintergreen and one evergreen species was similar to those in the present study (Jonasson 1989), Hence, data from several species suggest that the supply of nutrients from senescing leaves to new shoots in, at least, wintergreens, and probably also in some evergreens (Jonasson 1989), can be substituted by nutrient uptake without imposing any pronounced extra cost in terms of shoot growth. This implies that the extended leaf life-span plays a minor role as a nutrient conserving strategy in these life forms. Instead it appears that the extended life-span of the leaves primarily contributes to increase the leaves' lifetime gain of carbon. This does not exclude that other factors than those tested here, including nutritional factors (see e.g., Monk 1966, Chabot and Hicks 1982), also are of adaptive significance for plants with extended leaf longevity.
I thank Professor Terry Callaghan and Dr. Anders Michelsen for their helpful discussions and comments on the manuscript, Bo Norell and Cristina Pomar for valuable help in the field, Vivian Alden for technical assistance, and Karin Larsson for drawing the figures. Abisko Scientific Research Station provided facilities for the field work. The study was financially supported by the Swedish National Science Research Council grant number B-BU 4903-300 and the Danish Natural Science Research Council grant number 11-0421-1.
Adams, M. B., R. G. Campbell, H. L. Allen, and C. B. Davey. 1987. Root and foliar nutrient concentrations in loblolly pine: effects of season, site and fertilization. Forest Science 33:984-996.
Aerts, R., and M. J. van der Peil. 1993. A simple model to explain the dominance of low-productive perennials in nutrient-poor habitats. Oikos 66:144-147.
Beadle, N. C. W. 1954. Soil phosphate and the delimitation of plant communities in eastern Australia. Ecology 35: 370-375.
Bell, K. L., and L. C. Bliss. 1977. Overwinter phenology of plants in a polar semidesert, Arctic 30:118-121.
Berendse, F., and R. Aerts. 1987. Nitrogen-use-efficiency: a biologically meaningful definition? Functional Ecology 1:293-296.
Berendse, F., and S. Jonasson. 1992. Nutrient use and nutrient cycling in northern ecosystems. Pages 337-356 in F. S. Chapin III, R. L. Jefferies, J. F. Reynolds, G. R. Shaver, and J. Svoboda, editors. Arctic ecosystems in a changing climate, an ecophysiological perspective. Academic Press, San Diego, California, USA.
Chabot, B. F., and D. J. Hicks. 1982. The ecology of leaf life spans. Annual Review of Ecology and Systematics 13:229-259.
Chapin, F. S., III. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260.
-----. 1983. Direct and indirect effects of temperature on Arctic plants. Polar Biology 2:47-52.
Chapin, F. S., III, D. A. Johnson, and J. D. McKendrick, 1980. Seasonal movement of nutrients in plants of differing growth form in an Alaskan tundra ecosystem: implications for herbivory. Journal of Ecology 68:189-209.
Chapin, F. S., III, E-D. Schulze, and H. A. Mooney. 1990. The storage and economics of storage in plants. Annual Review of Ecology and Systematics 21:423-447.
Ewers, F. W., and R. Schmid. 1981. Longevity of needle fascicles of Pinus longaeva (Bristelcone Pine) and other North American pines. Oecologia (Berlin) 51:107-115.
Goldberg, D. E. 1982. The distribution of evergreen and deciduous trees relative to soil type: an example from Sierra Madre, Mexico, and a general model. Ecology 63: 942-951.
Greenway, K. J., S. E. Macdonald, and V. J. Lieffers. 1992. Is long-lived foliage in Picea mariana an adaptation to nutrient-poor conditions? Oecologia (Berlin) 91:184-191.
Jonasson, S. 1986. Influence of frost heaving on soil chemistry and on the distribution of plant growth forms. Geografiska Annaler 68A: 185-195.
-----. 1989. Implications of leaf longevity, leaf nutrient re-absorption and translocation for the resource economy of five evergreen plant species. Oikos 56:121-131.
-----. In press. Nutrient limitation in Rhododendron lapponicum, fact or artefact? Oikos.
Jonasson, S., and F. S. Chapin, III. 1985. Significance of sequential leaf development for nutrient balance of the cotton sedge, Eriophorum vaginaturn L. Oecologia (Berlin) 67:511-518.
Jones, H. E., and J. Dighton. 1993. The use of nutrient bioassays to assess the response of Eucalyptus grandis to fertilizer application. 2. A field experiment. Canadian Journal of Forest Research 23:7-13.
Karlsson, P. S. 1994. The significance of internal nutrient cycling in branches for growth and reproduction of Rhododendron lapponicum. Oikos 70: 191-200.
Kedrowski, R. A. 1983. Extraction and analysis of nitrogen, phosphorus and carbon fractions in plant material. Journal of Plant Nutrition 6:989-1011.
Loveless, A. R. 1961. A nutritional interpretation of sclerophylly based on differences in the chemical composition of sclerophyllous and mesophytic leaves. Annals of Botany 25:168-184.
Marquis, R. J. 1992. A bite is a bite is a bite? Constraints on response to folivory in Piper arieianum (Piperaceae). Ecology 73:143-152.
Monk, C. D. 1966. An ecological significance of ever-greenness. Ecology 47:504-505.
Nambiar, E. K. S., and D. N. Fife. 1987. Growth and nutrient retranslocation in needles of Radiata Pine in relation to nitrogen supply. Annals of Botany 60:504-505.
Reader, R. J. 1978. Contribution of overwintering leaves to the growth of three broad-leaved, evergreen shrubs belonging to the Ericaceae family. Canadian Journal of Botany 56: 1248-1261.
Reich, P. B., M. B. Waiters, and D. S. Ellsworth. 1992. Leaf life-span in relation to leaf, plant, and stand characteristics among diverse ecosystems. Ecological Monographs 62: 365-392.
Schlesinger, W. H., and B. E Chabot. 1977. The use of water and minerals by evergreen and deciduous shrubs in Okefenokee Swamp. Botanical Gazette 138:490-497.
Shaver, G. R. 1981. Mineral nutrition and leaf longevity in an evergreen shrub, Ledum palustre ssp. decumbens. Oecologia (Berlin) 49:362-365.
Small, E. 1972. Photosynthetic rates in relation to nitrogen recycling as an adaptation to nutrient deficiency in bog plants. Canadian Journal of Botany 50:2227-2233.
Tieszen, L. L. 1978. Summary. Pages 621-645 in L. L. Tieszen, editor. Vegetation and production ecology of an Alaskan arctic tundra. Springer, New York, New York, USA.
|Printer friendly Cite/link Email Feedback|
|Date:||Mar 1, 1995|
|Previous Article:||Canopy structure and photon flux partitioning among species in a herbaceous plant community.|
|Next Article:||Experimental analysis of intermediate disturbance and initial floristic composition: decoupling cause and effect.|