For Kentucky bluegrass and smooth brome, the pre- dawn leaf water potentials measured on the ﬁrst day after watering (August 9) and the third day after watering (August 11) did not differ statistically, although the latter tended to be lower (Table 1). The data for the two native species indicate a signiﬁcantly lower leaf water potential on the third day without watering than on the day following the watering event. However, no visual symptom of water stress was noted for either of the native species, suggesting only mild water stress developed for the native species in the control group, perhaps for a short period of time because they were watered twice a week. For all four species, the very low leaf water potential values mea- sured in the drought group on August 15, as well as the accompanied visual stress symptoms of leaf wilting and/or rolling in all four species, indicated the devel- opment of severe water stress, which, however, was alleviated upon rewatering, as seen in the leaf water potential data measured on August 16 (Table 1). After 97 days of growth, the amounts of ﬁnal total biomass (in grams per pot with standard deviation in parentheses) were 63(15), 72(15), 64(15) and 56 (13) for Kentucky bluegrass, smooth brome, western wheatgrass and green needlegrass respectively. The statistical conclusions reached using the nonparametric Mann-Whitney method and the parametric ANOVA were almost always the same. So, in the following, we only present the results from the Mann-Whitney method, except where there are notable differences between the two. The two invasive species had signiﬁcantly higher biomass allocation to roots than the two native species (Table 2). Conversely, the two native species had a Table 2 Biomass allocation to shoots, roots, crowns and rhi- zomes (in percentages) for the pooled data of introduced and native grasses (n = 48). Comparisons between intro- duced and native species are all highly signiﬁcant (P < 00005) for each column according to ANOVA on arc- sine-transformed data.
Group of species Shoots Roots Crowns Rhizomes Introduced 492 45 33 71 7 Native 618 31 46 70 signiﬁcantly higher biomass allocation to shoots than the two exotic species. Meanwhile, the two native species had a signiﬁcantly higher biomass allocation to crowns than the two exotic species. There were varied biomass allocations to rhizomes. The highest invest- ment (in rhizome) was observed in Kentucky blue- grass, followed by western wheatgrass, while smooth brome and green needlegrass had no rhizomes at all. It is interesting to note that the strongly rhizomatous smooth brome did not produce rhizomes in the ﬁrst season's growth, regardless of the water stress level (Figure 1). This zero rhizome investment in smooth brome was also observed in the 2010 experiment (data not shown). The nonparametric and parametric analyses essen- tially gave the same conclusions for all the tests, except in the case of green needlegrass root allocation, where the ANOVA on the arcsine-transformed data yielded a signiﬁcant result (P = 0035). However, the signiﬁcance of this effect was weakened by the con- tradicting conclusion from the nonparametric test regarding the effect of drought on root biomass alloca- tion in green needlegrass. In the two grasses that produced rhizomes (Kentucky bluegrass and western wheatgrass), drought stress reduced the biomass allocation to rhizomes (Figure 1). In Kentucky bluegrass, the rhizome alloca- tion decreased from 54% in the control group to 16% in the drought group. In western wheatgrass, the rhizome allocation decreased from 05% in the control group to <01% in the drought group (in the drought-stressed plants of western wheatgrass, no rhi- zome mass was measured in eight of 12 pots). The biomass allocation to crowns in smooth brome (42%) and western wheatgrass (53%) did not respond to drought stress (P = 014 and P = 047, respectively). However, the biomass allocation to the crowns in both Kentucky bluegrass and green needle- grass was signiﬁcantly inﬂuenced by drought stress. In Kentucky bluegrass, the average value increased to 4% from the control level of 25%, while in green needlegrass, the value decreased to 70% from the high level of 92% of the control group. It was within our original expectation that the only bunch-type grass (green needlegrass) in our experiment had a sig- niﬁcantly greater fraction of biomass allocated to the crowns, compared with the three sod-forming grasses species, for which both crowns and rhizomes serve as food reserves (P < 00005).
Abstract | References | Abstract | Background | Sustainable intensiﬁcation - changing paradigm in forage production | Global grasslands under threat | Land-use changes in the steppe ecosystem of Inner Mongolia Autonomous Region, P.R. China | Land-use change in the tropical savannah (Cerrado) of South America | Sustainable intensiﬁcation of dairy farming - the speciﬁc role of grassland- based forage production | References |