细枝柃性别表达特性及资源分配

已有的资料将柃木属(Eurya)描述为严格的雌雄异株植物, 性别变异现象极为少见。目前仅在柃木(E. japonica)和钝叶柃(E. obtusifolia)等少数种类中报道过两性花的存在。近几年笔者发现细枝柃(E. loquaiana)存在性别变异现象, 性别变异株上具有不同性别类型的花。该文从单花和植株水平分析了细枝柃的性别表达特性, 并对不同类型花的花部构件生物量分配进行比较分析。结果表明, 细枝柃具有6种类型的花, 从单花水平上看, 细枝柃性别有雌性、雄性及两性3种类型; 细枝柃性别在植株水平上体现较为复杂, 有雌株, 雄株, 雌花和两性花同株, 雄花和两性花同株, 雌雄异花同株及雌花、雄花、两性花同株6种类型; 在细枝柃花部构件生物量分配中, 雄花(包括雄株花和变异株雄花)花部构件生物量分配中雄蕊生物量的分配低于雌花(包括雌株花和变异株雌花)中雌蕊生物量的分配; 两性花中, 雄蕊生物量分配低于雌蕊, 这是其优化资源分配的手段, 进而获取最大适合度收益。     

长白山山杨种群的性比格局及其空间分布

以 5.2 hm²固定监测样地内山杨雌、雄植株的长期定位观测数据为基础,分析了长白山次生杨桦林内山杨(Populus davidiana)种群的性比格局、空间分布及其与环境因子间的关系。结果表明:山杨雌树和雄树的胸径大小差异不显著,始花大小相近,雄树的生长速率大于雌树,成熟个体的性比不显著偏离1 ∶ 1。山杨同性植株趋于相邻,雌雄植株的空间分离不显著;O-ring单变量点格局分析显示山杨在空间上呈异质性泊松分布。山杨同性别大树与小树在＜18m尺度上呈空间正相关;不同性别之间空间独立,说明性别之间的资源竞争不激烈。山杨雌树和雄树的空间分布与林分密度极显著正相关,林分密度对山杨雌雄植株空间分布变异的解释量较小,空间变量对雌树分布的解释量较大。     

色季拉山急尖长苞冷杉种群不同龄级立木的空间分布格局

植物种群空间分布格局是种群个体在水平空间的分布、配置状况，是种群对环境适应的生存策略及适应机制的反映。以色季拉山急尖长苞冷杉（Abies georgei var.smithii）原始林为研究对象，在色季拉山东坡设置100 m×100 m固定样地，采用相邻网格法将样地划分为25个20 m×20 m的调查单元，对样地内所有基径≥0.1 cm的急尖长苞冷杉进行调查，记录林木的坐标、胸径、基径、冠幅等信息。依据基径和胸径两个指标将林木划分为12个径级和6个龄级（幼苗、幼树、小树、中树、大树、老树），对种群的径级特征及空间分布格局进行分析。结果显示：（1）急尖长苞冷杉种群在不同径级上的分布成倒"J"型。（2）种群整体、幼苗、幼树随空间尺度的增大依次呈现集群分布、随机分布和均匀分布，小树、中树、大树、老树在所研究尺度内均为随机分布。（3）幼树与幼苗在小尺度上接近或达到负关联，在中、大尺度无显著关联；小树、中树、大树、老树在空间分布上与幼苗、幼树的关联性符合Janzen-Connell假说，但小树、中树、大树、老树之间无显著关联。上述结果表明：色季拉山急尖长苞冷杉种群为稳定增长型种群，小径木以集群方式相互庇护以提高自身适应环境的能力，大径木在小、中尺度抑制林下更新，在大尺度则促进更新。该结果有助于揭示色季拉山急尖长苞冷杉种群的现状与动态规律，可为深入研究该种群维持机制提供理论依据。     

贵州闽楠天然种群特征、空间分布格局及空间关联分析

为更好地保护闽楠天然种群，探究闽楠天然种群特征、空间分布格局及其空间关联性，对针对性地提出保护措施具有重要意义。以贵州台江县登鲁村闽楠天然种群为研究对象，通过野外设置60 m×60 m固定样地，采用种群径级结构代替年龄结构、点格局分析方法分析闽楠种群年龄结构、不同生长阶段空间分布格局及其关联性。结果表明：（1）闽楠种群年龄结构呈倒"J"型分布，幼龄个体数量多，随着年龄增加，能够成功进入后期的个体数量急剧下降，死亡率高。存活曲线趋于Deevey-Ⅱ型，属于稳定型种群。种群在第5龄级（7.5 cm≤DBH<12.5 cm）的生命期望（e_x）最高。（2）闽楠种群分布格局由随机分布转向随机分布。在幼苗、幼树、中树阶段空间分布格局趋于集群分布，大树阶段趋于集群分布。（3）不同生长阶段之间空间关联性表现有所差异。小尺度上，大树与幼树呈负关联，大尺度上，两者呈正关联；小尺度上，幼树与幼苗、大树与幼树都呈正关联，大树和中树无关联，在大尺度上，大树和中树、幼树与幼苗、中树与幼苗以及中树和幼树都呈负关联且随着空间尺度的增加负关联性逐渐增强。总体而言，该闽楠天然种群较稳定，但对环境干扰敏感，幼龄个体因呈聚集分布受密度制约而竞争剧烈，成年个体则因对环境的竞争而呈随机分布。为维持种群的长期稳定，需要分阶段适时进行人工调控。     

雌雄异株树种黄连木种群性比及空间分布

研究了100m×140m固定样地内黄连木种群的性比格局和空间分布。结果表明,调查样地中dbh≥4cm的黄连木植株共有2116株,其中包括526株雌树,1200株雄树,性别未确定植株390株。黄连木性比(雄/雌=2.28)显著偏雄(P0.001)。雌树和雄树平均胸径分别为7.34和7.81cm,雄树胸径显著大于雌树胸径(P0.05)。黄连木幼树、雌树和雄树均呈显著聚集性分布。黄连木幼树与雌树、幼树与雄树均在较小尺度上表现为相互吸引;雌树与雄树则在空间上相互排斥,即雌树与雄树存在空间分离现象;黄连木不同大小以及不同性别植株之间主要表现为相互排斥。     

浑善达克沙地黄柳种群性比及空间分布

雌雄异株植物的性比和空间分布格局对其繁殖策略、种群进化及发展起着至关重要的作用。黄柳是浑善达克沙地重要的固沙先锋植物,但针对其种群性比方面的研究却鲜有报道。采用点格局分析的方法,对黄柳4个天然种群的性比、空间分布格局以及空间关系进行了调查分析。结果表明:黄柳雌雄群体间在形态上的差异与种群所处环境条件密切相关;在调查样地内(50 m×50 m)种群性比(雌/雄)均极显著偏雌性(P0.01)。决定性比的主要组成部分为第I径级和第II径级雌雄个体的数量;随着年龄的增长,性比由原本的偏雌性转变为接近于1∶1。4个种群的雌雄群体在全部尺度范围内主要以聚集分布为主,偶尔在小尺度内呈随机分布。扎格斯台、饮马井、巴格来3个种群雌雄群体的空间关系均表现为相互独立,说明性别之间对资源的竞争不激烈;而宝绍岱种群两性的空间关系为相互排斥,即存在性别空间分离现象。研究结果揭示了天然黄柳雌雄群体的组成、结构及分布特征,为提高人工黄柳林群落稳定性奠定了理论基础。     

放牧干扰和种内竞争对伊犁河谷黑鳞顶冰花种群空间分布格局的影响

植物种群格局分析已有大量研究,但是大多数研究都是从格局推测过程,而很少根据已知的生态过程来检验格局。以新疆伊犁河谷分布的黑鳞顶冰花为研究对象,在放牧干扰和无放牧干扰地点分别建立了8个重复样地(样地面积2 m×2 m),分析放牧干扰和种内竞争两种生态过程对黑鳞顶冰花种群空间分布格局的影响。两个地点共测量了4677株,其中放牧地点819株,未放牧地点3858株。在放牧干扰下,8个样地的黑鳞顶冰花均在小尺度(r < 2 cm)上表现为随机分布,在大尺度(r > 2 cm)上表现为聚集分布,主要与放牧践踏形成的生境异质性有关;在无牧条件下,8个样地黑鳞顶冰花种群均在小尺度上表现出显著的均匀分布特征,在较大尺度则显示为随机分布特征,前者主要与种内资源竞争过程有关,后者主要与生境均质化有关。无论在无牧还是放牧干扰下,黑鳞顶冰花种群最大聚集规模与种群密度之间均存在显著的负相关关系。放牧干扰下各样地种群最大聚集规模介于20.26-168.71 cm²之间,平均值54.61 cm²,明显高于无牧条件的8.26 cm²。放牧干扰和种内竞争两种生态过程产生了明确且不同的空间分布格局,放牧导致黑鳞顶冰花个体出现更大程度的空间聚集,突出显示出干扰在种群空间分布格局形成中的重要性。     

西藏极小种群野生植物密叶红豆杉种群的性比及雌雄空间格局

雌雄异株植物种群具有性比和雌雄个体空间分布格局, 这对个体成功繁殖、种群生存潜力、天然更新能力和遗传多样性的维持都是重要的, 对于珍稀濒危雌雄异株植物种群尤其如此。密叶红豆杉(Taxus fuana)为国家I级重点保护植物, 是首批列入《全国极小种群野生植物拯救保护工程规划》的120种极小种群野生植物之一, 具有重要的生态和经济价值。但目前关于密叶红豆杉种群生态学方面的研究, 尤其是性比结构与空间分布格局在国内外鲜有报道。本文对西藏吉隆地区的6个天然密叶红豆杉种群(吉普、多甫、朗久、吉隆、开热和唐蕃)进行了实地调查, 研究其性比结构及其空间分布格局。结果表明, 调查的6个种群中总计雄株1,651株, 雌株1,231株, 仅吉普(雄/雌 = 1.89)与吉隆(雄/雌 = 1.39)两个种群有显著的性别偏倚现象且显著偏雄性(P < 0.001)。6个种群雌雄植株间的空间相关性不强, 整体趋于相互独立。不同径级间, 吉隆和开热种群的性比格局相似, 均在小径级上性比显著偏雄性, 而吉普种群则在中等径级上显著偏雄性。综上, 不同密叶红豆杉种群的大小、性比、雌雄个体的大小级结构以及空间分布格局等均表现出不同。因此, 需要结合各个种群本身的发展动态、受干扰的类型以及各区域环境因子的差异进行有针对性的保护。     

野生及栽培盾叶薯蓣（Dioscorea zingiberensis）的性别特征观察

对野生和栽培盾叶薯蓣(Dioscorea zingiberensis C. H. Wright)的性别特征及变异状况进行了观察,并对栽培盾叶薯蓣母根状茎和子根状茎萌发植株的性别特征进行了比较.观察结果显示,在湖北武当山6个样地717株野生盾叶薯蓣中共有155株开花植株,其中雌株、雄株和雌雄同株分别有60、94和1株,雌雄株比例1 ∶ 1～1 ∶ 3,平均雌雄株比例1 ∶ 1.57,每个根状茎一般具有1支缠绕茎.2004年至2006年的观察结果显示,2004年153株栽培盾叶薯蓣基株中雄株、雌株和雌雄同株分别有78、62和13株;2005年存活的基株中雌株和雄株数量明显减少,雌雄同株数量大幅增加,雄株、雌株和雌雄同株分别有35、19和88株;2006年成活的137株基株中有80株开花,其中雄株、雌株和雌雄同株分别有38、9和33株,雌株和雌雄同株数量大幅下降.随栽培年限的增加,每个基株的缠绕茎数由1～2支增加至5～6支.栽培过程中盾叶薯蓣的性别有较大幅度变异,雄株和雌株多数转变为雌雄同株,反向转变却较少,雄株和雌株间相互转变也较少.2005年存活并开花的基株中共有100株发生变异,在2006年有13株返变回原性别,有26株保持了2005年变异的性别,有11株变异为另一种性别;雄株、雌株和雌雄同株3年的总变异率分别达到80.77%、100.00%和46.14%.子根状茎萌发植株的性别与母根状茎有较大差异,总变异率达到约50%,其中母根状茎为雄性的其子根状茎萌发植株的性别总变异率最高(59.09%).研究结果表明,野生盾叶薯蓣可能以有性繁殖为主要繁殖方式,性系统简单清晰;栽培盾叶薯蓣复杂的性系统及大幅度的性别变异可能是由外部环境因子的影响造成的.     

基于形态学及生理学的黄檗性别差异研究

为了探究黄檗雌雄植株的形态学及生理学差异,对黄檗成树雌雄植株的形态学、叶片抗氧化防御酶活性及内源激素含量等指标开展对比研究,结果表明：黄檗雌雄植株形态存在显著差异,雌株叶宽比雄株大1.66±0.148 cm（P<0.05）、雌株枝间夹角比雄株大17.5±1.21°（P<0.001）;雌雄株叶片颜色存在差异,但叶绿素含量差异不显著;黄檗雌雄株超氧化物歧化酶（SOD）、过氧化物酶（POD）、过氧化氢酶活性（CAT）酶活性存在显著差异,CAT酶活性在7月最高,雌株含量大雄株;POD酶活性6月含量最高,雄株大于雌株;SOD酶活性8月最高,雄株大于雌株。内源激素赤霉素（GA3）含量雄株大于雌株;吲哚乙酸（IAA）含量为雌株大于雄株;玉米素（ZA）含量为雄株大于雌株,差异显著（P<0.05）;内源脱落酸（ABA）性别间无显著差异。黄檗雌雄植株在形态及生理学等指标上存在一定差异,说明黄檗雌雄植株为维持自身生长需要,在形态和生理方面都做出不同调整以适应生存环境。     

The Effects of Inflorescence Size and Flower Position on Biomass and Temporal Sex Allocation in <Emphasis Type="Italic">Lobelia sessiliflora</Emphasis>

To test the prediction of sex allocation theory that plants or flowers high in resource status emphasize the female function, we explored the variation in both biomass (the number of pollen grains and ovules) and temporal (male and female durations) sex allocation among and within plants of protandrous Lobelia sessilifolia in relation to plant size and flower position within plants. Among plants, the mean number of pollen grains and ovules per flower of a plant increased with plant size, whereas the mean P/O ratio (number of pollen grains/number of ovules ratio) decreased with plant size. The mean male duration, the mean female duration, and the mean ratio of male duration/flower longevity per flower of a plant were not correlated with plant size. Thus, large plants emphasized female function in terms of biomass sex allocation, which is consistent with the prediction of size-dependent sex allocation theory. The results for temporal sex allocation, however were inconsistent with the theory. Within plants, the mean number of pollen grains and ovules per flower at each position decreased from lower to upper flowers (early to late blooming flowers) and that of the P/O ratio increased from lower to upper flowers. The mean male duration and the mean female duration per flower decreased from lower to upper flowers, whereas the mean ratio of male duration/flower longevity increased from lower to upper flowers. The population sex ratio changed from male-biased to female-biased. Thus, later blooming flowers emphasized the male function in terms of both biomass and temporal sex allocation, consistent with the sex allocation theory, regarding the change in the population sex ratio.     

The true sex ratio in European Pseudocalliergon trifarium (Bryophyta: Amblystegiaceae) revealed by a novel molecular approach

Dioecious plants, including many bryophytes, rarely exhibit discernible sexual dimorphism before sexual maturity. Because many species and populations of dioecious bryophytes do not express their sex, it remains mostly unresolved whether expressing individuals reflect the ratios of genetically male and female plants. The present study assesses the population sex ratio of the wetland moss Pseudocalliergon trifarium in central and northern Europe. For the first time in a bryophyte, we estimate the sex ratio in a population by assessing directly both expressing and non‐expressing plants. Expressed gender ratio was assessed from herbarium specimens. Single shoots from non‐expressing specimens were sexed using a recently developed molecular sex marker. On the basis of the female and male frequencies in these two data sets and the overall proportion of expressing specimens, we estimate the European population sex ratio to be 1.93 : 1 (female/male). Expressed, non‐expressed, and population sex ratios are not significantly different from each other, suggesting that gender differences in rates of sex expression cannot account for the female bias. Earlier studies of P. trifarium failed to reveal gender‐specific growth rates or pre‐zygotic reproductive costs. Gender differences at the spore to protonemal stage, in mortality, or niche preferences could potentially explain the uneven sex ratio. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100 , 132–140.     

Floral sex ratio variation in hermaphrodites of gynodioeciousChionographis japonica var.kurohimensis Ajima et Satomi (Liliaceae)

Intrapopulation and interpopulation variations in floral sex ratio in hermaphrodites of gynodioeciousChionographis japonica var.kurohimensis (Liliaceae) were examined. The relative ratio of male flowers to total flowers (male and perfect flowers) decreased with plant size, suggesting size-dependent gender modification. The relative ratio of male flowers per population-basis is negatively correlated with the mean number of perfect flowers. Since the number of perfect flowers proportionally increased with plant size, populations showing low maleness consist of relatively bigger plants and are considered to be in high-quality environment. On the other hand, the relative ratio of male flowers per population basis is independent of female frequency in the population. Plasticity in gender expression probably plays an important role of maintenance of gynodioecy inC. japonica var.kurohimensis.     

Effect of growth substances on flowering and sex expression in isolated apical buds of Spinacia oleracea

Apical buds of Spinacia oleracea L. cv. Matador were isolated from 7-day-old vegetative seedlings and grown in sterile culture under inductive long, or non-inductive short photoperiods. Flowering of isolated apices was inducible in long days in approximately 75% of the plants, and in short days by gibberellin treatment (about 40%) or by raising the temperature to 30–32°C (13%). In long days the percentage of flowering was further increased by gibberellin treatment (up to 90%), while it was unaffected by abscisic acid and was strongly decreased by Amo 1618 (55%). In long days the ratio of male to female plants was near 1. The percentage of female plants in long days was increased by gibberellins, and the percentage of male ones decreased by kinetin; as a consequence, the ratio of male to female plants was lowered to about 0.50 in both cases. Abscisic acid and Amo 1618 had the opposite effect, probably because they decreased the flowering in female plants, so that the sex ratio was shifted to 2.6 and 6.8, respectively. Simultaneous treatment with GA₃ reversed the effect of abscisic acid on the sex ratio, but the reversal of the shift to maleness induced by Amo 1618 was only partial. In short days, gibberellins also stimulated the flowering in female plants more than in male. However, when the flowering was induced by higher temperature, most flowering plants were male and kinetin increased their percentage further. The above results have been discussed in terms of different requirements for flowering in male and female plants.     

Sex-ratio variation among Arisaema species with different patterns of gender diphasy

Arisaema species exhibit gender diphasy, or sex change, where individual plants produce either male, monoecious or female inflorescences depending on their size. Three basic sex-change patterns have been described in Arisaema. Type I species change between male and monoecious phases, type II species change between male, monoecious and female phases, while type III species change between male and female phases. Theoretical models suggest that sex ratios should be biased toward males, the sex with the lowest cost of reproduction. The goal of this study was to examine sex-ratio variation among Arisaema species that differ in sex-change patterns. Data from an extensive literature review, consisting of all available studies reporting Arisaema sex ratios, were combined with data from extensive field surveys of Arisaema dracontium and Arisaema triphyllum in southern Indiana, USA. This data set contains nearly 30 000 plants from 12 species. All species conformed to either the type I or type III pattern of sex change. There was little evidence for a distinct type II pattern of sex change, given that plants with monoecious inflorescences were rare relative to plants with pistillate inflorescences. The mean sex ratio in type I species (79.9% male) was significantly greater than in type III species (63.7% male). The data are consistent with the prediction that type I species are likely to have greater costs associated with female reproduction. We suggest that all Arisaema species have similar patterns of floral development, but differ in their ontogenetic patterns for male and female flowering.     

Sex ratio and sex expression in an urban population of the silver moss,Bryum argenteum Hedw.

ABSTRACT

Introduction. A well-supported pattern among dioicous bryophytes is male rarity. However, few assessments of bryophyte sex ratios have been made across environmental gradients to assess the role of environment in shaping population sex ratios.

Methods. We systematically surveyed 200 shoots from a 20?m² urban population of Bryum argenteum, and regenerated each shoot apex until sex expression occurred (up to 315 days).

Key results. Female shoots outnumbered male shoots 132 to 68, giving a sex ratio of 1.94♀: 1♂. The female bias was found in two transects in higher light environments but not in the third transect, which had a lower light level and an equal sex ratio. Female shoots took longer than male shoots to reach gametangial induction (122 vs. 60 days) and longer to produce 5 inflorescences (120 vs. 80 days). Male shoots produced an average of 10× the total number of inflorescences compared to female shoots (34 vs. 3.5 inflorescences). Despite producing more inflorescences, male plants also produced more regenerant shoots, thus contradicting the prediction that a higher prefertilisation reproductive effort in males trades off with vegetative proliferation. Female plants harboured significantly more associated microbes than male plants.

Conclusions. Our results support the role of light in influencing sex ratios in this species, suggest that trade-offs between reproduction and vegetative growth may not be strong for males, and indicate a potential role of a sex-specific microbiome in influencing sex ratios.     

Sex ratio in populations of Silene otites in relation to vegetation cover,population size and fungal infection

Abstract. In contrast to populations of most dioecious Silene species (which usually are female-biased), populations of Silene otites have been frequently reported to be male-biased. We describe sex ratio variation in 34 natural S. otites populations in Central Germany in relation to vegetation cover, population size and fungal infection. The overall sex ratio was unbiased in 1994 and only slightly male-biased in 1995. Sex ratio varied among the populations from 26.6 % to 72.6 % females. The sex ratio of small populations varied strongly due to stochastic processes. Furthermore, we found that populations in habitats with high vegetation cover contained a higher percentage of females. Hermaphroditic plants, theoretically, could increase male bias as they only produce male or hermaphroditic offspring. Their frequency in the populations, however, was far too low to affect sex ratio. In 1994 12.1 % and in 1995 17.0 % of the plants were infected by the smut fungus Ustilago major. Disease incidence in the population was not related to sex ratio, suggesting equal susceptibility of males and females. The sex ratio of partially infected plants did not deviate from the population sex ratio, both under field conditions and in a greenhouse laboratory experiment. The results suggest that the frequently reported male bias in Silene otites populations is not a general pattern, but is mainly caused by environmental conditions.     

Females make tough neighbors: sex-specific competitive effects in seedlings of a dioecious grass

If males and females of a species differ in their effect on intraspecific competition then this can have significant ecological and evolutionary consequences because it can lead to size and mortality disparities between the sexes, and thus cause biased population sex ratios. If the degree of sexual dimorphism of competitive effect varies across environments then this variation can generate sex ratio variation within and between populations. In a California population of Distichlis spicata, a dioecious grass species exhibiting extreme within-population sex ratio variation (spatial segregation of the sexes), I evaluated the intraspecific competitive effects of male and female D. spicata seedlings in three soil types. The sex of seedlings was determined using a RAPD-PCR marker co-segregating with female phenotype. Distichlis spicata seedlings, regardless of sex, were six times larger when grown with male versus female conspecific seedlings in soil from microsites where the majority of D. spicata plants are female, and this sexual dimorphism of competitive effect was weaker or did not occur in other soil types. This study suggests that it is not just the higher costs of female versus male reproduction itself that cause spatial segregation of the sexes in D. spicata, but that differences in competitive abilities between the sexes—which occur as early as the seedling stage—can generate sex ratio variation.     

Do secondary sexual dimorphism and female intolerance to drought influence the sex ratio and extinction risk of <Emphasis Type="Italic">Taxus baccata</Emphasis>?

Sex ratio and sexual dimorphism were studied in the dioecious tree Taxus baccata. We examined five populations of T. baccata in Poland and Ukraine to identify the differences between male and female individuals. The sex of all individuals, height and diameter, needle length and area, specific leaf area (SLA), the number of stomata rows, stomatal density, and content of carbon and nitrogen were measured to identify the differences between male and female individuals. The relationship between sex ratio and climatic conditions, age and population size were analysed using data collected from the field and the literature. Female trees were shorter than males, but needles of females were longer and had larger area. Although there were no differences among sexes in SLA, nitrogen and carbon concentration, we found a positive correlation between nitrogen concentration and SLA among females. The sex ratio changed with tree height within populations, and taller height classes were biased in favour of males. Regardless of population age, the percentage of females within populations was positively correlated with precipitation. Probably high reproductive effort caused female trees to lose in competition with males, and this loss may also be enhanced by lower drought tolerance in females and could contribute to risk of extinction for T. baccata. The continental geographic range of T. baccata may be restricted by limited occurrence of females, which demand higher water resources than males.     

Host population structure affects field sex ratios of the heteronomous hyperparasitoid, Coccophagus atratus

Abstract.

• 1 Second instar Filippia gemina de Lotto scale insects are the preferred hosts of female Coccophugus atrutus Compere larvae. These scale insects were found on their host plants, Chrysanthemoides monilifera Norlindh and Cliffortia strobilifera Mettenius, only at certain times during a 1 year sampling programme.
• 2 Late larval instars and prepupae of C.atratus, and a Metaphycus species, are the preferred hosts of male C.atratus larvae. These hosts, although they occurred on the same host plants as hosts for female C.atratus, were most numerous at different times during the sampling period.
• 3 The ratio of hosts suitable for C.atratus varied from a predominance of hosts suitable for females through to a predominance of hosts suitable for males. Sex ratios of adult C.atratus followed a similar trend but did not reflect, exactly, the ratio of available hosts. Differences in mortality between sexes and hyperparasitism may account for this anomaly.
• 4 Variable population sex ratios observed in C.atratus apparently result from the behaviour of individual females in which brood sex ratios are dependent on the relative availability of hosts for males and hosts for females. This behaviour, in turn, may result from variability in the host population structure but may also result from selection pressures operating at the time that heteronomous hyperparasitism evolved.