首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 15 毫秒
1.
Resown S23 perennial ryegrass pastures showed considerably greater response to the application of fertilizer N (0·1075 kg N/ha per year = 0–960 Ib N/ac) at 305 m O.D. than the native Festuca ovina/Agrostis tenuis and Molinia caerulea dominant communities on identical brown earth and peaty gley soils. Ryegrass DM production during 1967–70 increased with N application rates up to 538 kg N/ha per year (480 Ib N/ac) on the acid-brown earth, while on the extremely N-deficient gley soil yield responses were recorded up to 1075 kg N/ha (960 Ib N/ac). Percentage N recovery by ryegrass, bowever, although improved by grazing and re circulation, was less than under lowland conditions and the response to N during the growing season was also lower, exceeding 20 kg DM/kg N from applications in May, June and July only. The recovery of N by Festuca/Agrostis and Molinia in 1968–70 ranged from 3% and 2% at 938 kg N/ha to 14% and 9% at 117 kg N/ba per year, respectively. The respective maximum average DM yields recorded were 29 t/ha and 2.2 t/ha from the native communities and 70 t/ha and 62 t/ha from the corresponding resown pastures (2610, 1950, 6250 and 5520 Ib/ac). The results are discussed in relation to the strategy of land improvement in upland areas.  相似文献   

2.
The effect of three spring management treatments on the vertical distribution of dry‐matter (DM) yield and morphology of four cultivars of perennial ryegrass (Fennema, Corbet, Foxtrot, Melle) in mid‐season was investigated. The management treatments commenced with cuts on 15 February (Early), 1 March (Medium) and 29 March (Late), each followed by a 28‐day re‐growth period until the next cut and then further 21‐day re‐growth periods after each subsequent cut. This created four mid‐season measurement periods across the management treatments at cut 3 (5 April–17 May), cut 5 (17 May–28 June), cut 6 (7 June–19 July) and cut 7 (28 June–9 August). Tiller and sheath height and their ratio, and leaf lamina length, were measured prior to the four mid‐season cuts (cuts 3, 5, 6 and 7) when measurements of DM yield and proportions of leaf, stem and dead material in three herbage horizons (Lower, 0–8 cm; Middle, 8–15 cm; Upper, >15 cm) were made. There were significant responses in mid‐season to the management treatments involved complex interactions between management treatment and cutting date, which modified seasonal patterns in DM yield and leaf:stem ratio. There were significantly greater tiller heights, tiller:sheath ratios and leaf lamina lengths but lower sheath heights from the Early to Late management treatments. The greatest responses in morphological characteristics occurred in the Middle horizon compared with either the Lower (predominately stem and pseudo‐stem), or the Upper (predominately leafy) horizons. Distribution of DM yield between Middle and Lower horizons but not overall DM yields was significantly affected by management treatment. Morphological differences between cultivars were mostly in the Middle horizon and ranking of the cultivars was similar across the management treatments. The different responses of cultivars Fennema and Melle showed that genotype had a significant effect regardless of management. The leafiest mid‐season swards were achieved by delaying initial spring defoliation in the cultivar which had the lowest stem production.  相似文献   

3.
A trial containing cultivars of Vicia faba var. equina, minor and major was harvested as a whole-crop in 1971. The aim was to estahlish the DM yield obtainahle from a selection of spring-hean cultivars and evaluate this in relation to previously reported yields from winter heans. DM yields ranged from 48 to 86 hkg/ha over the three cutting dates. Significant differences were found hetween cv. in digesdhiUty, the range being 54 to 71%, and one hroad-hean cv. was shown to have a high digestihility at all cuts. Crude-protein contents of 13 to 19% were recorded. Percentage DM rose from 16 to 18% at the first cut to 23 to 45% at the third, hroad-hean cv. rising more slowly than horse- and dck-hean cv. Spring-bean cv. had a lower DM yield than winter cv. Horse hean was superior to tick bean in DM yield, percentage DM, and D-value. Broad-bean cv. proved difficult to dry and therefore may be more suitable for ensilage than dehydradon.  相似文献   

4.
Established swards of two diploid and two tetraploid red clover varieties sown pure received 0, 75, 150, 225 or 300 kg ha?1 N fertilizer and were cut three times in June, August and October 1971. The total yields of herbage DM for red clover varieties ranged from 8.01 to 11.32 t ha?1; swards sown with tetraploids Hungaropoly and Hera Pajbjerg were superior by 25% in DM yield and 23% in CP yield. The red clover contribution to these total yields of DM ranged from 6.05 to 10.69 t ha?1; tetraploid clovers outyielded diploids by 42% in DM yield and 39% in CP yield. The mean effect of N level on yield and on compositional attributes was slight. Total yields of herbage DM, averaged over all varieties, ranged from 9.50 to 10.22 t ha?1 and of total herbage CP from 1.76 to 1.91 t ha?1. The influence of N level on the red clover contribution was negligible. DM yields ranged from 8.54 to 8.72 t ha?1 and CP yields from 1.60 to 1.64 t ha?1. Superiority of tetraploid clovers over diploids was again confirmed. Red clover swards sown pure can give high yields without the application of fertilizer N.  相似文献   

5.
Small-plot trials with Lupinus albus cv. Kievskij Mutant (sweet) and L. angustifolius cv. Kubesa (bitter) and Uniwhite (sweet) were sown in April 1977. Sequential whole-crop samples of the above ground dry matter (DM) were taken to assess the interaction of DM yield and forage quality during crop growth. Both species had similar DM yields, exceeding 11 t DM ha1 in August. L. albus displayed superior whole-crop in vitro digestibility from late July, due to the large contribution to total DM yield from this time from its succulent, highly digestible pod shells. L. angustifolius showed higher N contents in June and early July, but N concentrations fell below those of L. albus from early August. The steady decline in digestibility of forage from L. angustifolius suggests this species should be harvested rather earlier than L. albus. Harvested in mid-August, L. albus should yield 11 t DM ha-1, ata D-value of 64 and 23·5 g N per kg DM, whereas L. angustifolius harvested 10–14 d earlier should give the same DM yield, but at a D-value of 58 and 22·0 g N per kg DM. Lupins can thus be regarded as having good potential forage yields, but providing material of only moderate quality.  相似文献   

6.
Residual response to an initial application of nitrogen was measured in 3 years and was found to be quite large, e.g. a response of 8 Ib DM per extra Ib N at the first harvest, when applying 125 Ib N/ac (140 kg/ha) rather than‘75 Ib (84 kg/ha) and allowing 6 weeks' growth, was increased to 21 Ib DM when yield from residual cuts was taken into account. Response was large in relation to the amount of N estimated to have been left behind after the first harvest. Marginal recovery of N, already quite high at the first harvest, was increased to a little over 90% when yield from residual cuts was taken into account Recovery, however, was lower than this when 10 weeks' growth was allowed before the first harvest, and it seems that, in this case, some N may have been lost from the system. A supplementary experiment suggested that the residual effect of N on DM yield could be obscured to some extent by fresh applications of N.  相似文献   

7.
Two field experiments were conducted at Teagasc, Moorepark, Ireland, to determine the effect of sowing date and nitrogen application on the dry‐matter (DM) yield and crude protein (CP) content of forage rape and stubble turnips. The first experiment consisted of three sowing dates (1 August, 15 August and 31 August) with four rates of fertilizer N (0, 40, 80 and 120 kg N ha?1) on forage rape DM yields. The second experiment consisted of three sowing dates (1 August, 15 August and 31 August) with four rates of fertilizer N (0, 40, 80 and 120 kg N ha?1) over two soil sites (fertile or nitrogen depleted) on forage rape and stubble turnip DM yields. A delay in sowing from 1 to 31 August characterized a 74·5% decrease in forage rape DM yield, while stubble turnip DM yield decreased by 55·5%. Forage rape DM yields increased positively up to 120 kg N ha?1 at the first two sowing dates over both sites. In contrast, stubble turnips showed less response beyond 40 kg N ha?1 on site 1 in the first two sowing dates, while DM yield increased positively up to 120 kg N ha?1 on the less fertile site. The results indicate that the optimal sowing time for forage rape and a stubble turnip in Ireland was early August.  相似文献   

8.
POTENTIAL PASTURE PRODUCTION IN THE UPLANDS OF WALES   总被引：1，自引：0，他引：1  
Study of the growth of heavily fertilized swards of S23 perennial ryegrass (Lolium perenne) at a hill and a lowland centre in mid-Wales during 1968–69 demonstrated that lower productivity in the uplands could only be partially attributed to a shorter growing season and poorer summer growth conditions. Comparison of yields with potential transpiration estimates indicated that winter damage had an important bearing on production during the following year. In 1969, the adverse combination of a mild autumn and a severe winter led to a spring tiller density which was less than 25% of normal at the hill centre. With an annual fertilizer input of over 500 kg N/ha (480 lb N/ac) total DM production in 1968 and 1969 was 10.6 and 4.4 t/ha at the upland site compared with 14.8 and 10.1 t/ha in the lowlands (9450, 3920, 13,190 and 9000 lb/ac).  相似文献   

9.
Studies have shown that yield responses of grass leys to phosphorus (P) fertilization vary considerably and initial soil test P (STP) does not always predict such variation. To identify the sources of variation in dry‐matter (DM) yield responses to P, we used a meta‐analysis to review quantitatively 38 Finnish field experiments on various soil types and growth conditions and to develop yield‐response models. The major sources of variation in the yield responses to annually applied P were soil type, soil acidity (for organic soils), initial STP and control yields fertilized with nitrogen (N) and potassium (K). According to Mitscherlich‐type exponential models, the maximum yield increases due to P fertilization in the low‐STP class were 18% over the control on mineral soils and 28% on organic soils. However, only about 40% of the variation in yield could be attributed to annual fertilizer P applications. Control yields explained about 25% of variation in yield responses to P fertilization, and P addition became less efficient with increasing control yields, which were associated with rising N rates. While applied P leads to infrequent or small yield responses on mineral soils, other practices to increase grass yields should be prioritized, including optimal N fertilization. On moderately acidic organic soils that show clearly lower P responses than do similar but less acidic soils, liming should be the first step to increase yields. Moreover, the models derived in this work indicate that revising P fertilizer recommendations in Finland would avoid unnecessary application at non‐responsive STP levels.  相似文献   

10.
The response of established perennial ryegrass swards to irrigation was compared over 2 years. The swards received 250 (N1) or 500 kg N/ha (N2) and were cut 4 times each year. Water was applied in 5 irrigation treatments according to the potential soil water deficit: the soil was partially, or fully returned to field capacity after cutting; returned to field capacity after cutting and then whenever a 25 mm deficit arose; irrigated with 25 mm water whenever a 25 mm deficit arose independently of cutting; or irrigated at minimum intervals of 14 days, provided that the deficit was then at least 50 mm. There was little effect of irrigation on yield from the first cut in late May. Frequent irrigation gave the greatest increase in total annual yield (3.1 t/ha), hut partial irrigation after cutting was most efficient in use of water and equipment, provided that drought conditions were avoided. High concentrations of N in the herbage were associated with low DM yields, hut more N (percentage N DM yield) was recovered from the irrigated than from the unirrigated swards. More N was recovered than was applied in the N1 treatment; less was recovered than was applied in N2. In the N1 treatment the nitrate concentration was <500 ppm hut it was 1000–4000 ppm in N2. There was no consistent effect of irrigation on nitrate or K concentration.  相似文献   

11.
The effect of sowing date, within the period from late May to mid-August, on the DM yield of five leafy forage brassica crops sown to a stand was examined over the period 1976–78. Kale was the crop most sensitive to variation in sowing date and yields declined progressively with later sowings. There was no yield benefit from sowing rapes or stubble turnips earlier than mid- to late-June. Raphanobrassica produced yields as good as rape when sown in June. In 1978, DM yields of rape, kale and Raphanobrassica increased from September to early November, largely as a result of stem growth; kale was slower to accumulate DM than rape in late summer but continued growing longer into the autumn. The DM concentrations of all crops increased from September to December. Rape had a higher DM concentration than stubble turnip. Raphanobrassica was 10–25 g kg-1 lower in DM concentration than rape or kale. Raphanobrassica produced 23–59% flowering plants from late-May to early-June sowing. Harvest date affected DM yield and concentration, leaf: stem ratio and flowering percentage. The effects of sowing dale on DM concentration were small.  相似文献   

12.
A series of eight fertilizer N rates, ranging from 0 to 377 kg N/ha per year, at increments of 53.8 kg N, was applied to a S170 tall fescue/SlOO white clover sward. The annual rates were split into 8, 6 or 4 equal dressings for harvests at intervals of 3, 4 and 6 weeks, respectively, over a 24-week growth period from March to September, in each of two years. Residual harvests were taken 6 weeks later in October. Curves relating annual herbage yields to N rate were fitted to tbe data. Total herbage DM responses to 108 kg N/ha were small because of the typical white clover yield N rate interaction, but were almost linear from 108 to 323 kg N/ha at all harvesting frequencies. Tbe decline in rate of response at tbe bigher N application levels was less marked with frequent than with infrequent defoliation. Tall fescue DM responses were substantially linear from 0 to 323 kg N/ha. For the first and second years, relative DM yields of total herbage for harvesting frequencies of 3, 4 and 6 weeks were 100:108:120 and 100:111:131, respectively. Total herbage CP response curves were similar to those for DM, but continued to rise more steeply to the highest N rate tested, 377 kg N/ha. Tall fescue CP responses were linear from 0 to 377 kg N/ha. Harvesting frequency did not markedly influence CP yields. Total herbage DM yield was 14% less in the second year compared witb the first. The decline was due to a reduction in white clover, attributed mainly to the effect of repeated N applications, also the subsidiary effects of companion grass and weather.  相似文献   

13.
Yield response curves were developed for potatoes (Solanum tuberosum cv. Russet Burbank) with sprinkler-applied N fertilizer on a high-frequency schedule. The potatoes were grown on a Quincy sand to loamy sand (mixed, mesic, Xeric Torripsamments) typical of sandy-textured soils of the center-pivot irrigated areas of the Columbia River Basin of Oregon and Washington. These soils require daily irrigation at peak evapotranspiration because of low water-holding capacity. Yields were measured for total quantities of N, ranging from 100 to 665 kg N/ha, applied in small frequent increments as a urea-NH4NO3 solution. Maximum tuber yields ranged from 67 to 85 Mg/ha in 6 experiments over a period of 6 years. Economically optimum yields were obtained at fertilizer rates between 300 and 400 kg N/ha depending on cost-price ratio of N fertilizer and potatoes. Fertilization in this range should optimize tuber yield and quality with fertilizer N efficiency and economic return on fertilizer investment.  相似文献   

14.
Rates of N applied to sandy loam soil with wheat as the preceding years crop was studied as to the influence on tuber yields of potatoes grown for early summer harvest in Southwestern Indiana. Most of the yield response was to the first 75 lbs/acre (84 kg/ha) increment of N which increased the yield of US #1 potatoes 61 cwt/acre in 1967 and 52 cwt/acre in 1968. The total solids were not significantly decreased by the first increment of N applied which was also the N range of rapid yield increase. N fertilization increased the N content of the foliage from 3.0 to 4.0% in 1967 and 3.97 to 5.17 in 1968 and the tubers from 1.36 to 2.96 in 1968. The first 75 lbs N increment was efficiently converted to crude protein, 48% incorporation, but the succeeding increments were poorly converted.  相似文献   

15.
A field plot experiment was carried out on an established grassland sward from 1983–88 inclusive to examine the effects of time of application, chemical form of nitrogen (N) and cattle slurry dry matter (DM) content on yield and efficiency of N use. Four forms of fertilizer N (a semi-organic fertilizer, a combined 2.1:1 (w/w) semi-organic/calcium ammonium nitrate (CAN) fertilizer, CAN and urea, each supplying 300 kg N ha?1 year?1, were applied with or without unseparated or separated cattle slurry at 93 and 73 g kg?1 DM respectively, both supplying approximately 150 kg N ha?1 year?1. All fertilizers and slurries were applied in three equal dressings (February/March, May/June and July/August). The efficiency of use of fertilizer and slurry N was evaluated by measuring DM yield, N uptake and apparent recovery of N in herbage at all harvests during each growing season. Fertilizer N application significantly increased (P<0.001) the mean yields of herbage at each harvest in all years. The form of fertilizer N had no significant effect (P>0.05) on first harvest and total herbage yields, nor on N uptakes by herbage at the first harvest in any year. The performance of urea and of CAN was more variable at the second and third harvests relative to that of the semi-organic or combined 2.2:1 (w/w) semi-organic/CAN sources which had similar efficiencies of N use. Lower DM production was associated with reduced uptake of N. Values for mean overall apparent recovery of N ranged from 57.9 ± 2.67% for the semi-organic fertilizer to 50.2±3.05% for CAN. Unseparated cattle slurry and separated cattle slurry produced similar herbage yields and N responses that were lower and more variable than with fertilizer N. The overall mean apparent recovery of N from unseparated cattle slurry was 25.5 ± 5.03% compared to 5.0 ± 4.74% for separated cattle slurry. Efficiency of N use was highest with spring applications and least with mid-season applications. Recoveries ranged from ?29% for separated cattle slurry applied in June 1984 to 56% for unseparated and separated cattle slurry applied in February 1988 and June 1987 respectively. No interactions were recorded between cattle slurry and fertilizer N in terms of DM production or N uptake by herbage. The results of this study support the use of a fertilizer N source, selected on a least-cost basis, in combination with slurry to promote spring herbage production. For subsequent production, N should be supplied in fertilizer form only. The use of urea is risky under low rainfall conditions. Mechanical separation did not improve the efficiency of use of slurry N.  相似文献   

16.
Six annual legumes were evaluated as components of cereal-grass-legume intercrops in two experiments at two sites differing in elevation by 789 m. Barley (Hordeum vulgare L.) and Westerwolds rye-grass (Lolium multiflorum Lam.) were seeded on all intercrop plots. Dry-matter (DM) yield, crude protein (CP) and organic matter digestibility (OMD) were measured. DM yield and N content were used to estimate legume N fixation. Experiment 1 was conducted at both sites. At the lower site, Persian clover (Trifolium resupinatum L.) and annual alfalfa (Medicago sativa L.) accounted for 70% of the DM yield in harvest 1 (July), increased CP and OMD, but did not affect intercrop yield. They increased harvest 2 (August/September) intercrop yield by 263% and CP concentration by 65 g kg?1 DM. They increased harvest 3 (October) yield by 275% and CP concentration by 78 g kg?1 DM. Inclusion of striate lespedeza (Lespedeza striata) did not affect intercrop yield or quality. Annual legumes failed to establish at the higher elevation site and therefore had no effect on DM yield or forage quality. In Experiment 2, in which the performance of Westerwolds ryegrass was also compared with that of Italian ryegrass, and conducted at the lower site only, Persian clover and berseem clover (T. alexandrinum L.) increased CP of all three of the year's harvests. These two species contributed 29% of the DM yield in the first harvest (July) but did not affect total intercrop yield. They increased harvest 2 (August) yield by 313%. Persian clover increased harvest 3 (October) yield by 318% and berseem clover increased harvest 3 yield by 405%. Barrel medic (Medicago truncatula) and snail medic (M. scutellata) contributed 29% of harvest 1 yield, and increased both DM yield and CP content. Medics did not regrow. Aubade Westerwolds ryegrass contributed a greater percentage of the DM yield than did Maris Ledger Italian ryegrass at harvests 1 and 2. Ryegrass type did not affect total DM yield but did affect forage quality; intercrops containing the Italian ryegrass had higher CP at harvest 2 and higher OMD at harvest 3 than those containing the Westerwolds ryegrass. Over both experiments, at the lower elevation site, stands with Persian clover, berseem clover or alfalfa produced 80% of the yield of barley-ryegrass receiving 250 kg N ha?1, and 165% of the yield of unfertilized barley-ryegrass. Berseem and Persian clover fixed about the same amount of N over the growing season; 188 kg N ha?1 in Experiment 1 and 134 kg N ha?1 in Experiment 2.  相似文献   

17.
Four levels of nitrogen, 188, 285, 358 and 392 kgJ.ha (150, 225, 285 and 350 units/ac) supplied as anhydrous ammonia were injected on each of four occasions, 18 Dec, 5 Feb., 12 Mar. and 16 Apr. DM and CP yields were estimated at each of three harvests taken on 4 June, 19 July and 28 Sept Total DM yields showed only a 10% difference between the highest yielding injection date, 12 Mar., which yielded 12270 kg/ha (10946 Ib/ac) and the lowest, 16 Apr., which yielded 11169 kgJ.ha (9963 Ib/ac). Early injection of ammonia promoted growth during the late spring and later injections resulted in increased yields at the two later harvests. The responses per kg fertilizer N were 11.9 kg DM (10.8 lb/unit) and 3.6 kg CP (3–3 Ib/unit). It was concluded that ammonia should be applied to grassland before April and, preferably, during March, for maximum yields of DM and CP.  相似文献   

18.
Multi-species legume-grass swards play a key role in organic farming systems as a source of forage and of symbiotically fixed nitrogen (N). Forage and N balances, based on site-specific data on dry matter (DM) yields of herbage, are useful planning tools in describing forage and N supply at the field and farm level. In practice, measurements of DM yield are often imprecise or missing, and thus a weather- and site-specific assessment of DM yield is required. An empirical model to predict DM yield from legume-grass swards was developed based on weather and soil data commonly available at field and regional scales. The main underlying hypothesis was that water use, calculated from cumulative water balances, can be used as a predictor of DM yield. The model was calibrated with data from a multi-year field experiment in Müncheberg, north-east Germany and was tested with data from other countries of Europe. In the calibration data set, highly significant linear relationships were found between water use and DM yield for DM yield of single harvests and for annual DM yield. The only additional variable significantly improving the prediction of DM yield was cut number. For the validation data set the DM yield for single cuts and annual yields was predicted with a similar accuracy as found with other models requiring the use of more information. The models described offer a straightforward weather- and site-specific means of predicting DM yield with a satisfactory level of precision, especially for annual DM yields, and thus can help to reduce planning failures concerning forage and N supply in organic farming systems in Europe.  相似文献   

19.
The changes in dry matter (DM) yield, botanical composition and nutritive value of herbage to ruminants of two wet grasslands, Arrhenatherum elatius grassland (Experiment 1) and a Molinia caerulea fen meadow (Experiment 2), in which a range of cutting and fertilizer treatments were imposed in 1999, were assessed after 4–7 years of treatment imposition. Both experiments had a split‐plot design with four replicates. In Experiment 1 the three main‐plot cutting treatments were two cuts with a delayed first cut, three cuts and four cuts during the growing season of each year. In Experiment 2 the cutting treatments were two cuts with a traditional harvest time, two cuts with a delayed first cut and three cuts. The four sub‐plot fertilizer treatments were an unfertilized control, application of a phosphorus and potassium (PK) fertilizer, application of a nitrogen (N) and PK fertilizer to the first cut only (N1PK) and application of PK plus N applied to each of two, three or four cuts (Nc PK). Application of fertilizer influenced yield and botanical composition of herbage more than the cutting treatments while the opposite occurred for nutritive value of the herbage. Application of fertilizer increased the proportion of tall grasses in Experiment 1 and forbs in Experiment 2. The proportion of Equisetum palustre, present only in Experiment 1, was reduced from 0·33 to less than 0·01 by increased cutting frequency together with the NPK fertilizer treatments. In Experiment 1 diversity of vascular plants was negatively affected only by the four‐cuts treatment while on both wet grasslands other cutting and fertilizer application treatments had no effect. Changes in DM yield of herbage caused by the cutting and fertilizer application treatments were similar for both vegetation types with DM yield increased significantly by fertilizer application but only slightly or not reduced by increasing the cutting frequency. Nutritive value of herbage was positively correlated with cutting frequency and was most influenced at the first cut.  相似文献   

20.
An experiment involving 24 grassland sites was carried out to measure the influence of a large number of variables on upland yields between 900 and 1450 ft above sea level in western Aberdeenshire. The grass was harvested in mid-July and at the end of October 1966. Few factors were significant in the first harvest yields. In the second harvest yields, aspect and soil parent material were both significant at the 1% level, and available soil phosphate at the 5% level. Total yields (DM) for the growing season were correlated with parent material (2% level), with available soil P (1% level) and with the P and K content of the leaf (1% and 5% levels). Yields were also correlated with sward age (5% level) and sward condition (1% level). A number of variables were highly correlated with sward age and condition, including the leaf content (second harvest) of K, P and crude protein (0.1%, 1% and 1% levels).
The highest yield was 4700 kg/ha/year and the average yield 2585 kg/ha/year, but the results showed that the average could be substantially increased by good management, especially by P fertilizer. The results confirmed that there was no marked fall-off in yields with increasing altitude (r=0.066, 0.194 and 0.112) and there appeared scope for reclamation on suitable sites at higher elevations.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号