黄土高原子午岭油松林的种子雨和土壤种子库动态

对黄土高原区子午岭不同林龄(18a、29a、40a、54a)油松(Pinus tabulaeformis carr.)人工林及天然林(约75a)的种子雨和土壤种子库进行了研究.结果表明,该区油松种子雨一般从每年9月初开始,一直到11月底结束,种子雨降落历程与林龄大小有关,种子雨发生时间和降落高峰期有所不同.不同林龄的油松种子雨强度不同,种子雨总量大小顺序为:40a人工林((489 9±8.64)粒· m-2)＞29a人工林((346.8±7.45)粒· m-2)＞54a人工林((327.1±8.13)粒· m-2)＞天然林((146.9±5.25)粒· m-2)＞18a人工林((78.1±2.72)粒· m-2).种子雨总量随林龄的增加而增加,约40a时达到高峰,种子雨活力也以40a时最高.不同林龄油松林土壤种子库存在显著差异,其中18a人工林种子库最小,40a人工林种子库最大.从种子雨降落到次年4月,5种林分土壤种子库总量下降了42.34%～53.59%,空粒种子增加了26.72%～48.69%;从4月到8月份种子腐烂率由10.28%～13 62%增加到57.25%～63.28%.动物的搬运、取食和种子腐烂死亡是种子库损耗的主要因素.土壤种子库中的油松种子主要集中在枯枝落叶层,其次为0～2cm层,2～10cm层种子最少.到8月中旬,土壤中98.26%的油松种子都已丧失活性.不同林分下油松幼苗的密度差异较大,40a人工林下幼苗最多,其余依次为29a人工林、54a人工林和天然林,18a人工林下的实生苗极少,幼苗死亡率极高.在一定龄级范围内,人工林结实能力和更新潜力随林龄增加而增加,40a时更新潜力最大.虽然有大量种子下落,但由于种子大量损耗和幼苗死亡,通过环境筛作用而最终可以成熟的个体数量十分有限.     

文峪河上游华北落叶松林的种子雨、种子库与幼苗更新

华北落叶松林下更新不良,为探究其制约因素,开展了山西省文峪河上游5个华北落叶松林分的种子雨、土壤种子库和幼苗更新的研究。结果表明:(1)华北落叶松种子主要集中于9—10月散落。2011年为华北落叶松种子丰年:种子产量高,种子雨密度达(961.93±377.40)粒/m2;种子质量高,完整种子占(89.31±16.13)%。2012年为种子平年,种子产量低,种子雨密度为(252.73±115.12)粒/m2。华北落叶松种子雨主要源于毗邻树木,华北落叶松纯林和落叶松云杉林的种子雨密度显著高于其他3个针阔混交林。(2)土壤种子库主要由上年种子雨组成,2012年4月的土壤种子库密度为(695.18±297.23)粒/m2,完整种子占(59.73±9.56)%。种子自然萌发前,约(78.98±24.76)粒/m2具发芽力,基本可满足更新需要。但种子活力保持期少于2 a,只能形成短期持久土壤种子库。(3)华北落叶松更新不良,种子年后仍难以实现幼苗建成,当年生幼苗的出现频度平均为1.6%,且林下难以存活。幼苗发生与种子储量关联性不强,种源条件不是制约华北落叶松更新的主要因素。     

茂兰喀斯特森林林隙种子雨、种子库空间变异

以贵州茂兰国家级自然保护区喀斯特森林为研究对象,研究了林隙植被的种子雨、种子库的数量特征及动态变化规律。结果表明:观察期林隙种子雨量达117.4±32.6粒/m2,其中未成熟种子56.3±10.3粒/m2,成熟被害种子15.7±4.7粒/m2,成熟有效种子45.4±8.2粒/m2,林隙更新的种子来源比较丰富。林隙中种子雨的丰富度具有时空异质性特征,将对林隙植物的更新格局产生深刻影响。枯落物层中的种子密度、物种数及Shannon-wiener多样性指数从林隙中心至非林隙林地递减,而土层中的种子变化则相反。林隙中心、近中心、林隙边缘的种子密度分别为2415±639粒/m2、2218±421粒/m2和1815±311粒/m2,林隙植物有很好的更新潜力。林隙与非林隙中枯落物层种子的相似性系数最大,5~10cm土层的次之,0~5cm土层的相似性系数最小。林隙与非林隙中均表现为枯落物层的种子库与现存植被的相似性较大,并随土层的加深而减小。研究结果表明茂兰喀斯特森林林隙植被的更新主要来源于土壤种子库,土壤种子库对林隙填充与发育的贡献较大。     

科尔沁沙地刺榆林种子散布的空间格局

通过对林缘和孤立单株不同方向远离种子源的顺序取样与林下随机取样调查,分析了刺榆林下和林缘的种子雨空间数量特征和孤立单株及林缘的种子散布格局,以揭示科尔沁沙地刺榆林的空间扩展潜力.结果表明:3个样地中,刺榆林下的平均种子雨密度最大,为(13732.5±3106.2)粒·m-2,孤立单株东南样带最多,为(5449.4±1429.3)粒·m-2,并显著地高于其他方向,西北最少,为(650.2±631.6)粒·m-2,并显著地少于其他方向;林缘东和南样带显著地高于西和北样带.孤立单株和林缘不同方向种子密度的变异普遍较大,其变异系数为25.7%～106.3%.刺榆具有风播植物远离种子源的种子散布特性.8个样带在单位面积和累积面积上远离种子源的种子散布格局均具有多样性,包括直线、幂、指数、二次抛物线和对数函数.在科尔沁沙地,刺榆林向南、东南和东等多顺风频率方向的空间扩展潜力将大干北、西北和西等多逆风频率方向.     

呼伦贝尔沙地天然樟子松种子扩散机制研究

通过对沙地樟子松母树种子落种时空格局的观测,发现沙地樟子松种子大部分在高峰期集中飞散在母树周边,与母树距离的增加种子数量减少,在距离母树4m左右处,落种密度达到最大,且母树北侧和东侧的落种密度要高于南侧和西侧.此外文章还探讨了落种密度和距离与母树本身的关系,如树高和冠幅等,推算出了树高和扩散距离之间的量化关系,计算得出在已知风速的情况下沙地樟子松种子的最大扩散距离公式.最后将沙地樟子松种子更新过程分为母树、种子和幼苗阶段3个阶段和母树的结实与生产、种实扩散、种子萌发和幼树生长4个时期,从环境状况和母树自身特征角度分别探讨了沙地樟子松更新的影响因素,并在此基础上建立了沙地樟子松种子更新的时空模型.     

准噶尔盆地东南缘梭梭种子雨特征

为了研究梭梭种子散布规律,通过布设种子雨收集器结合室内实验分析,对准噶尔盆地东南缘梭梭种子雨特征进行了研究。研究显示:(1)梭梭种子雨的累积密度达到平均189粒/m2,其中有活力种子占约80%;(2)种子散布的高峰集中在11月初到11月15日时间段,其落种量占整个种子雨的65%。其后种子雨密度随时间逐渐减小;(3)整个种子雨过程,不同时期散落的种子雨质量存在差异,表现为不同时期散落种子的萌发率呈现出先增大后减小的趋势;(4)变异函数分析表明,梭梭种子雨在8.12 m的有效变程内,种子雨具有明显的空间格局,其由空间自相关和随机因素引起的空间异质性各占50.0%。准噶尔盆地东南缘梭梭种子雨密度大且质量较高,同时其时空分布异质性较高,这些特征均将影响梭梭种群的分布格局和种群更新。     

毛红椿天然林种子雨、种子库与天然更新

2008-2011年,调查江西九连山国家级自然保护区毛红椿天然林的种子雨、种子库及林下幼苗数量.结果表明:在毛红椿天然林,种子雨散布时间为10月下旬至12月下旬.2010年不同样地的种子雨强度为虾公塘气象观测站(320.3±23.5粒·m-2)＞虾公塘保护站(284.7±24.2粒·m-2)＞大丘田保护站(251.6±24.7粒·m-2),分别以222.0、34.3和22.6粒·m-2完好种子供土壤萌发更新;毛红椿种子储量取决于结实量、鸟类取食和种子活力等因素,鸟类取食是其种子储量大幅下降的首要因素;由于种子不耐储藏以及大量腐烂,种子有效贮藏期不足1个月.12月天然林种子库平均萌发数≤2株·m-2,次年1月土壤种子库种子量最少,为6.7～11.8粒·m-2,平均仅萌发0.4～0.6株·m-2,与林下实生幼苗分布极少相吻合.毛红椿种子雨储备、种子库种子活力保存及幼苗建成等因素影响其天然更新.     

秦岭中段不同坡向锐齿栎种子雨、土壤种子库与幼苗更新

以秦岭中段不同坡向(阳坡、半阳坡、半阴坡及阴坡)锐齿栎天然林为对象,对锐齿栎种子雨、种子库的数量、质量动态以及幼苗发育过程进行研究.结果表明: 锐齿栎种子雨一般从8月中旬开始,9月中下旬到10月上旬达到高峰期,11月中上旬结束;不同坡向的种子雨降落历程、发生时间和组成不同;半阳坡种子雨强度最大,为134.13粒·m^-2,其余大小依次为阳坡、半阴坡、阴坡;阳坡种子雨降落时间最早、持续时间最长,高峰期持续时间也最长,阴坡种子雨降落时间最晚、持续时间最短,高峰期持续时间也最短;种子活力及成熟种子占其种子雨的比例大小顺序为半阳坡>阳坡>半阴坡>阴坡.从种子雨降落结束到翌年8月,各坡向土壤种子库总储量大小顺序均为半阳坡>阳坡>半阴坡>阴坡;各坡向的土壤种子库中成熟、未成熟、被啃食种子数量和种子活力均随时间变化呈递减趋势,而霉烂种子数量增加;各坡向土壤种子库中的种子均主要集中在枯落物层,其次为0~2 cm土层,而在2~5 cm土层中只有极少量种子存在.4种坡向中锐齿栎幼苗的密度差异显著,半阳坡幼苗最多,其余依次是阳坡、半阴坡和阴坡.半阳坡更适合锐齿栎种子的萌发和幼苗生长.     

神农架巴山冷杉种子雨的时空格局

巴山冷杉(Abies fargesii)是我国特有树种, 其分布区以秦巴山地为中心。在湖北神农架自然保护区选择不同的巴山冷杉群落设置样地, 通过布设种子收集器结合室内实验分析, 对巴山冷杉的种子雨进行研究。各样地巴山冷杉种子雨开始和持续时间大致相同, 始于10月初, 持续超过2个月, 但是不同的巴山冷杉种群表现出不同的种子雨时空格局。巴山冷杉－箭竹(Fargesia nitida) 群落种子雨的平均强度为167.93±111.14粒/m2, 有活力种子比例占22.31%, 落种高峰期集中于10月27日到11月2日间, 种子雨呈现聚集分布。巴山冷杉－茵芋(Skimmia reevesiana)群落的种子雨强度在3个样地中最小, 只有16.41±14.41粒/m2, 有活力种子仅占3.05%。对巴山冷杉－陕甘花楸(Sorbus koehneana)群落, 沿等高线方向和垂直等高线方向布置收集器带, 收集器自群落内中心母树延伸至林冠范围之外(距中心母树的距离>18 m)。收集结果表明种子雨落种高峰为10月15–21日, 种子雨主要集中在林冠范围内, 林冠内种子雨数量占到了收集总数的87.95%。方差分析结果显示各方向的种子雨数量不存在显著差异, 按照远离中心母树的距离进行分段比较, 0–6 m区段内的收集数量和其他各区段间(6–12 m, 12–18 m, >18 m)存在显著差异, 扩散的种子数量与离开中心母树的距离间的关系近似正态分布。研究结果说明巴山冷杉种子雨的强度大但种子质量不高, 而且种子雨的扩散距离有限。这些特征将影响到巴山冷杉种群的分布格局和种群更新。     

濒危植物南方红豆杉种子雨和土壤种子库特征

对福建南平大坪村样地和福州旗山样地濒危植物南方红豆杉种子雨动态进行了2a的观测,并在种子雨结束后不久的3月份和种子雨开始降落之前的9月份进行土壤种子库取样。南方红豆杉种子雨持续3个多月,高峰期从11月中旬开始到12月下旬结束,约40d,种子雨绝大部分降落在树冠范围内,种子雨中成熟种子占85%以上,2007年和2008年种子雨强度差异不显著(P0.05)。南方红豆杉土壤种子库主要集中在树冠范围内,在种子雨刚结束的3月份,土壤种子库主要分布在枯枝落叶层,且密度显著大于其它两层(P0.05),9月份第2次取样,枯枝落叶层种子库密度明显减少,当年种子雨仅有3%可补充到土壤种子库中,南方红豆杉种子雨中绝大多数种子受到动物取食、人为因素和环境因素的影响而损失掉无法进入土壤种子库。     

川西米亚罗亚高山云杉林种子雨和土壤种子库研究

 对川西米亚罗亚高山20、30、60年人工云杉(Picea asperata)林及天然林的种子雨和土壤种子库进行了研究。结果表明：该区云杉种子雨一般从每年的10月初开始下落,一直到翌年的1月底结束,但不同林龄的云杉种子雨强度不同,60年人工林种子雨量最大,然后依次为30年人工林、天然林、20年人工林,其大小分别为(1 088.16±52.34)粒•m－2、(973.45±63.12)粒•m－2、(579.99±28.93)粒•m－2、(66.73±5.71)粒•m－2。云杉降雨历程与其林龄大小也有一定的关系,表现在种子雨发生时间和降雨高峰期的不同。30、60年人工云杉林及天然林下土壤种子库存在显著差异,两次取样（2003年3月26日和2003年8月21日）土壤种子库大小分别为(507.2±40.32)粒•m－2和(267.6±25.14)粒•m－2、(1 065.6±88.06)粒•m－2和(872.8±77.12)粒•m－2、(472.8±20.82)粒•m－2和(185.5±22.48)粒•m－2。土壤种子库中云杉种子主要都集中在枯枝落叶层,占66%左右,0～2 cm层占24%左右,2～5 cm层占10%左右。到8月21日,土壤中所有云杉种子都失去活力,腐烂死亡和动物取食是种子库损耗的主要因素。云杉种子库属于Thompson和Grime定义的第Ⅱ类。该区云杉幼苗一般在6月初开始出现,6月中旬左右达到出苗高峰,3种云杉林下幼苗密度差异很大,60年人工林下幼苗最多,其次为30年人工林和天然林。由于生境适合度低,死亡的幼苗数量较多,特别是自然死亡数。综合种子雨、种子库和转换的幼苗数量几个方面来看,在一定龄级范围内,人工林结实能力和更新潜力随林龄增加而增加,且人工林在更新潜力方面比天然林（约150年）要好。虽然该区云杉林下有大量种子下落,但由于种子大量损耗和幼苗死亡,种子萌发为幼苗再通过环境筛作用而最终补充到云杉种群的个体数量非常有限。     

Composition and seasonal dynamics of seed rain in broad-leaved Korean pine (Pinus koraiensis) mixed forest in Changbai Mountain, China

To explore the composition and spatio-temporal dynamics of seed rain in broad-leaved Korean pine (Pinus koraiensis) mixed forest, 150 seed traps were set up in a 25 hm2 plot in Changbai Mountain. Seeds, fruits, anthotaxy and others in seed traps were collected, identified and divided into 4 types. From 2005 to 2006, we collected 47 different types. Total number of seeds and fruits was 121291, including 23147 mature seeds and fruits (19.1% of the total). Tilia amurensis and Fraxinus mandshurica, with the most seeds and fruits, accounted for 90% of the total. The analysis on seasonal dynamics of seed rain showed that there were the largest number of seeds and fruits between July and October, which were composed of immature seeds and fruits. In mid-October, mature seeds and fruits reached their peak, but immature seeds and fruits still accounted for high proportion. There were 91 traps that contained 100–200 mature seeds and fruits, and one trap without any mature seed or fruit. The largest number of species found in a trap was 7, and usually 3 or 4 species were found in most of the traps. There were obvious relationships between spatial patterns of mature seeds and fruits and their parent trees, indicating that their mature seeds and fruits were not dispersed far from their parent trees.     

松嫩草原榆树疏林对不同干扰的响应

榆树疏林是中国东北松嫩草原沙地植被演替的顶极群落,其对维持物种多样性,保持水土,涵养水源等都发挥着重要的功能。过去20a间,榆树疏林在人为干扰(农耕和放牧)下,其群落结构和物种组成发生了巨大变化。在1983年和2004年,对松嫩草原不同人为干扰下榆树疏林群落结构和物种组成等进行调查,结果显示:农耕和放牧一方面使松嫩草原榆树疏林群落多物种消失、生物量和多样性显著下降、植被结构简单化;另一方面,也是导致建群种——家榆低矮化,灌木化的主要因素。     

小兴安岭阔叶红松(Pinus koraiensis)林种子雨的时空动态

研究种子散布的时空动态对于揭示群落更新机制和植被斑块分布格局具有重要意义。于2005～2007年在凉水自然保护区的9hm^2阔叶红松林永久样地的中心地带（150m×150m）,设置287～319个种子接收器（面积为0.5m^2,网口位于离地面1m处）,定期收集并鉴定其中的种子。结果表明：（1）乔木树种的种子雨强度在不同的年份间存在差异,由2005年到2007年逐年递减（分别为（864.2±1084.3）粒·m^-2、（300.9±349.4）粒·m^-2和（144.8±195.5）粒·m^-2）。11个树种均表现出了年际间的差异,一些树种如水曲柳、糠椴、白桦、红皮云杉、冷杉在某一年份结实量很小或几乎不结实,而红松受人为干扰较大。（2）种子散布存在明显的季节变化,从5月份到11月份均收集到了各树种的种子,且种子雨在10月达到高峰,11月中旬基本结束,但不同的树种的时间变化形式不同。高峰期,种子雨以完整种子为主;而在这之前的种子雨主要以未成熟种子为主。（3）种子雨的构成在各年份间保持稳定。（4）变异函数分析表明,在不同的年份种子雨空间异质性不同,而空间异质性与种子雨强度呈正相关。     

地上竞争与地下竞争对科尔沁沙地榆树幼苗生长的影响

榆树疏林草原对科尔沁沙地植被恢复和景观保护有着重要意义.本文采用双因素两水平控制试验,从幼苗生物量、地下/地上生物量、茎高、根茎比、叶片数等方面,研究了草-树地上、地下竞争对科尔沁沙地榆树幼苗生长的影响.结果表明:对于1年生榆树幼苗,单株平均生物量表现为无竞争＞地上竞争＞全竞争＞地下竞争;地下/地上生物量表现为地下竞争＞全竞争＞无竞争＞地上竞争;幼苗高度表现为地上竞争＞无竞争＞全竞争＞地下竞争;根茎比表现为地下竞争＞全竞争＞无竞争＞地上竞争;叶片数表现为地上竞争＞无竞争＞地下竞争＞全竞争.地下竞争对1年生榆树幼苗生长影响显著,而地上竞争对榆树幼苗生长无显著影响.地上竞争与地下竞争对2年生榆树幼苗生长的影响均不显著.科尔沁沙地草本植物对榆树幼苗生长的影响主要通过地下竞争的方式实现,但地下竞争并没有改变榆树幼苗的资源分配方式.随榆树幼苗龄级的增长,草本植物竞争作用的影响逐渐减弱.     

SEED DYNAMICS OF ABIES BALSAMEA AND ACER SACCHARUM IN A DECIDUOUS FOREST OF NORTHEASTERN NORTH AMERICA

We studied the seed dynamics of Abies balsamea and Acer saccharum in a sugar maple-yellow birch forest of Québec, Canada. Seed rain was censused every week from spring to autumn in 1988 and 1989. The soil seed bank was sampled four times during the 1988 growing season. Abies seed rain varied significantly between 1988 and 1989. A total of 92.5 ± 7.0 seeds nr^-2 (mean ± 1 SE) were produced in 1988; 1989 seed production was null. Seed viability in 1988 averaged 31.5%. Dispersal during the winter period accounted for ca. 22% of the annual seed crop. The spatial dispersion pattern of the seed rain was contagious for the autumn period, but was not significantly different from random for the winter period. There was only a weak relationship in the spatial distribution of the seed rain between the autumn and winter periods. Abies did not maintain a persistent seed bank on the site; viable seeds were present on the soil only after seed dispersal had started. Postdispersal seed mortality was relatively high, ca. 70%. Acer seed rain started soon after the flowering period, but consisted at that time only of aborted, underdeveloped samaras. Viable seeds were disseminated from September through early winter. Dispersal during the autumn accounted for ca. 98% of the annual seed crop. There were significant differences between seed rain abundance of 1988 and that of 1989 (118.1 ± 9.5 samaras m^-2 and 158.9 ± 16.3 samaras m^-2 in 1988 and in 1989, respectively). Seed viability also varied significantly between the 2 years, i.e., 19% in 1988 and 5% in 1989. For both 1988 and 1989, samaras were contagiously dispersed over the site. There was a significant positive relationship between the spatial distribution of the seed rain in 1988 and that in 1989. Seed rain abundance was high mainly in the proximity of mature Acer trees. As for Abies, Acer did not maintain a persistent seed bank in the soil; viable seeds were present only in the autumn seed bank sampling, after seed fall had started. Postdispersal seed mortality was relatively low, i.e., 20%. Abies and Acer are quite similar in their regenerative traits, and these contrast sharply with those of Betula alleghaniensis, a regular member of this forest community. Differences in regenerative traits may contribute to the coexistence of these tree species considering the gap regime of the system studied.     

Deposition of exotic bird-dispersed seeds into three habitats of a fragmented landscape in the northeastern United States

Dispersal by frugivorous birds facilitates invasion by many exotic plants. We measured the seed rain of ornithochorous native and exotic plants at three habitats of a fragmented landscape of the northeastern United States for 1 year. We studied maple-beech forests, old fields, and abandoned conifer plantations. Across all sites we collected 2,196 ornithochorous seeds, including seeds from six exotic species and 10 native species. The majority (90%) of collected seeds were from exotic species. Seed dispersal was broadly similar among habitats, though seed rain of exotic species was higher in old fields than forested habitats. Seed rain was not strongly influenced by artificial perches for most species. However, seeds of exotic species were more commonly found in traps under an artificial perch in old fields. Seed rains for the exotic Elaeagnus umbellata, Rhamnus cathartica, and Rosa multiflora were positively associated with local density of mature plants. Seed rain of R. cathartica was positively associated with abundance of seedlings but not saplings, suggesting that post-dispersal mortality was important. Seed dispersal of the exotic Lonicera spp. was high in all habitats, accounting for 66% of all seeds collected. With the exception of Lonicera spp., seed rain of common exotic invaders was affected by the abundance of seed sources, and these species might be effectively controlled by elimination of local fruiting plants. Fruits of Lonicera morrowii, which has extensively invaded our area, are apparently a common component in the diet of frugivores.     

Spatial Distribution and Site-Specific Spraying of Main Sucking Pests of Elm Trees

Elm trees are important landscape trees and sucking insects weaken the elm trees and produce large amounts of honeydew. The main objectives of this study were to identify main honeydew-producing pests of elm trees and do site-specific spraying against these pests. To map the spatial distribution of the sucking pests in the large scale, the study area was divided into 40?×?40 m grids and one tree was chosen randomly from each grid (a total of 55 trees). These trees were sampled twice a year in 2011 and 2012. Each sample was a 30-cm branch terminal. Eight samples were taken from each tree in four cardinal directions and two canopy levels. The number of sucking insects and leaves of each sample were counted and recorded. Spatial analysis of the data was carried out using geostatistics. Kriging was used for producing prediction maps. Insecticide application was restricted to the regions with populations higher than threshold. To identify within-tree distribution of the honeydew-producing pests, six and four elm trees were chosen in 2011 and 2012 respectively, and sampled weekly. These trees were sampled as described previously. European elm scale (EES), Gossyparia spuria (Modeer) and two species of aphids were the dominant honeydew-producing pests. The results revealed that the effects of direction, canopy level and their interactions on insect populations were not statistically significant (P?<?0.05). Site-specific spraying decreased the amount of insecticides used by ca. 20%, while satisfactory control of the sucking pests and honeydew excretion was obtained. Considering the environmental and economic benefits of site-specific spraying, it is worth doing more complementary works in this area.     

浑善达克沙地四种生境中榆树天然更新幼苗发育的比较

对浑善达克沙地腹地风蚀低地、覆沙草地、平坦流动沙地和流动沙丘阴坡 4种类型生境中植物群落和榆树幼苗生长的比较研究结果表明 :(1)尽管单个样方内植物种数生境间没有差异 ,但群落总盖度覆沙草地的最低 ,流动沙丘阴坡上的居中 ,风蚀低地和流沙平地的最高。榆树幼苗的分盖度风蚀低地的最高 ,在流沙平地的次之 ,在覆沙草地的最低 ,流动沙丘阴坡的介于流沙平地和覆沙草地的中间 ,但与二者没有显著性差异 ;(2 )风蚀低洼地的幼苗生长的最好 ,平均高度、主根长度、叶片数和叶面积均显著大于生长在覆沙草地和流沙平地的幼苗的。生长在流动沙丘阴坡上幼苗的植株高度、叶片数目和叶面积介于它们之间 ,主根长度与风蚀低地的没有显著差异 ;(3)风蚀低洼地榆树幼苗根、茎、叶各部分生物量都明显地高于生长在其它 3种生境中幼苗的 ,流动沙丘阴坡上榆树幼苗的生物量还明显高于生长在覆沙草地和流沙平地上的 ,生长在覆沙草地和流沙平地的榆树幼苗的生物量基本上没有差异。该结果说明 ,榆树幼苗在风蚀低地生长最好 ,其次是在流动沙丘阴坡 ,这两种生境可能是浑善达克沙地榆树更新的主要地方。     

Seed rain into upland plant communities in Hong Kong, China

Hong Kong is an extreme example of tropical landscape degradation, with no substantial remnants of the original forest cover and a highly impoverished disperser fauna. Seed availability is a potential limiting factor in vegetation recovery in such landscapes. To assess the quantity and quality of the seed rain of woody taxa, seed traps were placed in the major upland vegetation types: fire-maintained grassland, shrubland, and secondary forest. Within the grassland site, traps were placed under isolated trees, isolated male and female shrubs of Eurya chinensis, and in the open. Seeds were collected every 2 weeks for 2 years. The seed rain was highest under female shrubs in grassland (6455 seeds m⁻² year⁻¹), where it was almost entirely confined to their fruiting period. Next highest were isolated trees (890 seeds), followed by male isolated shrubs (611 seeds), shrubland (558 seeds), forest (129 seeds) and open grassland (47 seeds). The number of seed taxa was highest in shrubland (59), followed by isolated trees (42), forest (42), female isolated shrubs (28), male isolated shrubs (15), and open grassland (9). The seed rain differed in species composition between the forest, shrubland, and grassland sites, while the differences within the grassland site were largely in terms of quantity. Birds (particularly bulbuls, Pycnonotus spp.) are known or inferred to be the major dispersal agents for 85% of the seed taxa trapped, 99% of the total number of seeds trapped, and 99.8% of the seeds trapped in the grassland site. Few taxa and of the total seeds were dispersed by wind and no seed taxa were definitely dispersed by fruit bats. The results suggest that even in the most degraded landscape the seed rain is adequate for the development of woody vegetation cover, but that human intervention will be needed for the restoration of plant diversity.