Variance in animal longevity: contributions of heterogeneity and stochasticity

Variance in longevity among individuals may arise as an effect of heterogeneity (differences in mortality rates experienced at the same age or stage) or as an effect of individual stochasticity (the outcome of random demographic events during the life cycle). Decomposing the variance into components due to heterogeneity and stochasticity is crucial for evolutionary analyses.In this study, we analyze longevity from ten studies of invertebrates in the laboratory, and use the results to partition the variance in longevity into its components. To do so, we fit finite mixtures of Weibull survival functions to each data set by maximum likelihood, using the EM algorithm. We used the Bayesian Information Criterion to select the most well supported model. The results of the mixture analysis were used to construct an age × stage-classified matrix model, with heterogeneity groups as stages, from which we calculated the variance in longevity and its components. Almost all data sets revealed evidence of some degree of heterogeneity. The median contribution of unobserved heterogeneity to the total variance was 35%, with the remaining 65% due to stochasticity. The differences among groups in mean longevity were typically on the order of 30% of the overall life expectancy. There was considerable variation among data sets in both the magnitude of heterogeneity and the proportion of variance due to heterogeneity, but no clear patterns were apparent in relation to sex, taxon, or environmental conditions.     

Genetic basis of between‐individual and within‐individual variance of docility

Between‐individual variation in phenotypes within a population is the basis of evolution. However, evolutionary and behavioural ecologists have mainly focused on estimating between‐individual variance in mean trait and neglected variation in within‐individual variance, or predictability of a trait. In fact, an important assumption of mixed‐effects models used to estimate between‐individual variance in mean traits is that within‐individual residual variance (predictability) is identical across individuals. Individual heterogeneity in the predictability of behaviours is a potentially important effect but rarely estimated and accounted for. We used 11 389 measures of docility behaviour from 1576 yellow‐bellied marmots (Marmota flaviventris) to estimate between‐individual variation in both mean docility and its predictability. We then implemented a double hierarchical animal model to decompose the variances of both mean trait and predictability into their environmental and genetic components. We found that individuals differed both in their docility and in their predictability of docility with a negative phenotypic covariance. We also found significant genetic variance for both mean docility and its predictability but no genetic covariance between the two. This analysis is one of the first to estimate the genetic basis of both mean trait and within‐individual variance in a wild population. Our results indicate that equal within‐individual variance should not be assumed. We demonstrate the evolutionary importance of the variation in the predictability of docility and illustrate potential bias in models ignoring variation in predictability. We conclude that the variability in the predictability of a trait should not be ignored, and present a coherent approach for its quantification.     

Twin study of the heritability of recognition thresholds for sour and salty taste

Seventy-four pairs of monozygotic (identical) twins and 35 pairs of dizygotic (fraternal) twins provided recognition thresholds (modified Harris-Kalmus test) for the sourness of citric acid and the saltiness of sodium chloride during the Twins Days Festival in Twinsburg, OH. Variance components (ACE) models were applied to the data: total variation = additive genetic (A) + common environment (C) + nonshared environment (E). The best-fit model of variation in recognition thresholds for sourness included an additive genetic factor, accounting for 53% of the variance, but no common environment component. This level of heritability, on par with that of sensitivity to the bitter compounds 6-n-propylthiouracil and phenylthiocarbamide, strongly suggests that genetic factors play a larger role than shared environment in determining individual differences in recognition thresholds for sourness. In contrast, the best-fit model for saltiness recognition included a common environment component, accounting for 22% of the variance in thresholds, but no additive component. This result suggests that environment plays a larger role than genetics in determining individual differences in recognition thresholds for saltiness.     

Mothering Style and Infant Behavioral Development in Yunnan Snub-Nosed Monkeys (Rhinopithecus bieti) in China

Considerable variation in mothering styles is found among primate species, which may be directly related to species-typical differences in social structure, dominance style, patterns of infant development, and rates of intragroup aggression. We predicted that, as egalitarian Asian colobines, mothers of Yunnan snub-nosed monkeys (Rhinopithecus bieti) would adopt a mothering style characterized by low restrictiveness and low rejection. We followed six mother–infant dyads in a provisioned group of Rhinopithecus bieti inhabiting the Baimaxueshan National Nature Reserve, Yunnan, China, and collected 717.2 h of observational data on maternal care and infant development. In the first month after birth, infants were completely dependent on their mothers for nutrition and movement. They began to locomote clumsily during their second month of life. Mothers restricted the movements of their infants only during their first 2 mo of life. Maternal rejection occurred infrequently and never exceeded a mean of two events per hour for a given infant over any 1-mo period. Most rejections were mild, and did not result in a cessation of nursing. Infants were not weaned when they were 12 mo old. Our study offers support for the contention that females of Rhinopithecus bieti adopt a relaxed mothering style in caring for offspring during their first year of life.     

Relative climatic,edaphic and management controls of plant functional trait signatures

Objective: To identify the relative roles of climatic, edaphic and management factors in controlling the weighted mean traits of vegetation. Location: Eleven sites in Europe and one in Israel undergoing transitions in land use. Material and Methods: Standardised methods were used to collect information on species traits and attributes from plots covering a range of land uses at each site. This was combined with abundance data to create a plot × trait matrix. Variance partitioning was used to identify the relative roles of climate, soil and management on the weighted and unweighted mean traits of the vegetation in the full data set, and the data set divided into vegetative traits (including life‐form, clonality, defence and a range of leaf traits) and traits linked to regeneration via seeds (including seed mass, dispersal and pollination mechanism). Results: Variance partitioning of the full data set showed that climate (18.7%), explained more variance in the weighted mean traits of the vegetation than climate and soil together (9.2), soil (6.9) and management (6.1). There was a similar distribution of variance explained for both vegetative and regeneration via seed traits, although more variance was explained for the latter. This restricted set of climatic, edaphic and management variables could explain 45‐50% of the variance in the weighted mean traits of the vegetation between plots. There were only small differences between analyses of the weighted and unweighted data. Conclusions: Despite large variations in climate and soils between sites, there was still a separate and recognisable impact of management on the mean weighted traits of the vegetation. There was also a degree of shared variation between the three groups of factors, indicating that the response of plant traits to one group of factors may not be predictable because they may be modulated by their response to other groups.     

Individuality in the vocalizations of infant and adult coppery titi monkeys (Plecturocebus cupreus)

As social animals, many primates use acoustic communication to maintain relationships. Vocal individuality has been documented in a diverse range of primate species and call types, many of which have presumably different functions. Auditory recognition of one's neighbors may confer a selective advantage if identifying conspecifics decreases the need to participate in costly territorial behaviors. Alternatively, vocal individuality may be nonadaptive and the result of a unique combination of genetics and environment. Pair-bonded primates, in particular, often participate in coordinated vocal duets that can be heard over long distances by neighboring conspecifics. In contrast to adult calls, infant vocalizations are short-range and used for intragroup communication. Here, we provide two separate but complementary analyses of vocal individuality in distinct call types of coppery titi monkeys (Plecturocebus cupreus) to test whether individuality occurs in call types from animals of different age classes with presumably different functions. We analyzed 600 trill vocalizations from 30 infants and 169 pulse-chirp duet vocalizations from 30 adult titi monkeys. We predicted that duet contributions would exhibit a higher degree of individuality than infant trills, given their assumed function for long-distance, intergroup communication. We estimated 7 features from infant trills and 16 features from spectrograms of adult pulse-chirps, then used discriminant function analysis with leave-one-out cross-validation to classify individuals. We correctly classified infants with 48% accuracy and adults with 83% accuracy. To further investigate variance in call features, we used a multivariate variance components model to estimate variance partitioning in features across two levels: within- and between-individuals. Between-individual variance was the most important source of variance for all features in adults, and three of four features in infants. We show that pulse-chirps of adult titi monkey duets are individually distinct, and infant trills are less individually distinct, which may be due to the different functions of the vocalizations.     

Genetic parameters of body weight in sheep estimated via random regression and multi-trait animal models

Variance components for five consecutive measurements of body weight in Polish sheep were estimated using random regression and multi-trait animal models. The data included between 7856 and 31694 body weight records at 5 age classes from birth to 150 days of age. The random additive genetic, maternal environmental and individual permanent environmental effects were fitted. All variance components were increasing over time (not at equal rates), which reflects increasing phenotypic mean and variance with age. Direct heritability tended to increase with age, whereas the effect of dam was reduced for older ages, and the proportion of permanent environmental component was relatively stable. Generally, similar tendencies were registered for estimates obtained via multi-trait animal model. The results confirm that there is a scope for genetic improvement in growth pattern in Polish sheep.     

Thermal environment and sudden infant death syndrome: case-control study.

OBJECTIVE--To compare the thermal environment of infants who died of the sudden infant death syndrome with that of age matched control infants. DESIGN--Case-control study. Infants who died were matched with two controls, one for age and one for age and birth weight. Thermal measurements were conducted at the death scene for cases and at the scene of last sleep for control infants, who were visited unexpectedly within four weeks of the index infant''s death on a day of similar climatic conditions. A follow up questionnaire was administered to parents of cases and controls. SETTING--The geographical area served by the professional Tasmanian state ambulance service, which includes 94% of the Tasmanian population. SUBJECTS--41 infants died of the sudden infant death syndrome at home; thermal observations at death scene were available for 28 (68%), parental questionnaire data were available for 40 (96%). 38 controls matched for age and 41 matched for age and birth weight. RESULTS--Cases had more excess thermal insulation for their given room temperature (2.3 togs) than matched controls (0.6 togs) (p = 0.009). For every excess thermal insulation unit (tog) the relative risk of the sudden infant death syndrome was 1.26 (95% confidence interval 1.05 to 1.52). The average thermal bedding value calculated from parental recall was similar to that observed by attendant ambulance officers (mean difference = 0.4 tog, p = 0.39). Cases were more likely to have been found prone (odds ratio 4.58; 1.48 to 14.11). Prone sleeping position was not a confounder or effect modifier of the relation between excess thermal insulation and the syndrome. CONCLUSIONS--Overheating and the prone sleeping position are independently associated with an increased risk of the sudden infant death syndrome. Further work on infant thermal balance and sudden infant death is required and guidelines for appropriate infant thermal care need to be developed.     

Rate of growth in length of Bangladeshi infants as a function of attained length.

We examine variation in the rate of growth in length of breast-feeding infants from rural Bangladesh. These data were collected between November 1985 and February 1986 from two rural sites. Eighty-eight infants, ranging from birth to 4 months of age at the start of the study and their mothers were measured monthly for 4 months. Length increased linearly with age over this 4-month period (infants' average bias-adjusted R2 = 0.90). The relationship between infant rate of growth in length and attained length was analyzed by two different methods: Oldham's (1962) method of regressing rate of growth on mean length and Blomqvist's (1977) method of regressing rate of growth on estimated initial length. The methods gave similar results. The rate of growth was negatively associated with mean infant length over the 4-month period (p less than 0.001); that is, shorter infants grew at a faster rate than longer infants. For every centimeter shorter the infant was, the rate of growth was 0.1 cm/mo faster on average; the effect was greater among males than among females. The average rate of growth was greater for males than for females and greater in financially solvent households and varied by site. Infant growth rate was slower among older infants than among younger infants, as expected. However, after adjusting for mean infant length, age was no longer significantly associated with infant growth rate, although mean infant length remained highly significant. Forty-one percent of the variation in infant rate of growth in length was explained by mean infant length, sex, sex by length interaction, household financial solvency, and site.     

Genetic heterogeneity of within-family variance of body weight in Atlantic salmon (Salmo salar)

Background

Canalization is defined as the stability of a genotype against minor variations in both environment and genetics. Genetic variation in degree of canalization causes heterogeneity of within-family variance. The aims of this study are twofold: (1) quantify genetic heterogeneity of (within-family) residual variance in Atlantic salmon and (2) test whether the observed heterogeneity of (within-family) residual variance can be explained by simple scaling effects.

Results

Analysis of body weight in Atlantic salmon using a double hierarchical generalized linear model (DHGLM) revealed substantial heterogeneity of within-family variance. The 95% prediction interval for within-family variance ranged from ~0.4 to 1.2 kg², implying that the within-family variance of the most extreme high families is expected to be approximately three times larger than the extreme low families. For cross-sectional data, DHGLM with an animal mean sub-model resulted in severe bias, while a corresponding sire-dam model was appropriate. Heterogeneity of variance was not sensitive to Box-Cox transformations of phenotypes, which implies that heterogeneity of variance exists beyond what would be expected from simple scaling effects.

Conclusions

Substantial heterogeneity of within-family variance was found for body weight in Atlantic salmon. A tendency towards higher variance with higher means (scaling effects) was observed, but heterogeneity of within-family variance existed beyond what could be explained by simple scaling effects. For cross-sectional data, using the animal mean sub-model in the DHGLM resulted in biased estimates of variance components, which differed substantially both from a standard linear mean animal model and a sire-dam DHGLM model. Although genetic differences in canalization were observed, selection for increased canalization is difficult, because there is limited individual information for the variance sub-model, especially when based on cross-sectional data. Furthermore, potential macro-environmental changes (diet, climatic region, etc.) may make genetic heterogeneity of variance a less stable trait over time and space.     

Fumonisin intake of the German consumer

In order to calculate the dietary fumonisin intake of the German consumer, a large survey was carried out on a variety of potentially contaminated products in the period between December 1998 and July 2001. A total of 1960 food samples comprising all known relevant groups of products were analysed for fumonisins. Furthermore, 272 of these samples were also analysed for hydrolysed fumonisins (HFB). For routine analysis enzyme immunoassay was used, confirmatory and control analyses were performed using HPLC-FLD after precolumn derivatisation, or by LC-MS/MS. Daily intake of fumonisins was calculated by combining fumonisin contamination data obtained in this study with available food consumption data for Germany. In a “mean case” scenario, median fumonisin levels in foods and mean food intake values were used. To generate a “bad case” scenario, the 90^th percentile of fumonisin levels in foods and mean food intake values were combined. The overall daily fumonisin intake by the German consumer was 1.1 μg in the “mean case” scenario, and 21 μg in the “bad case” scenario. It was concluded that in general there is no increased risk for the German consumer in aspects of exceeding the recommended tolerable daily intake of fumonisins (2 μg/kg body weight). However, certain products (and certain brands of products) were repeatedly found to contain elevated fumonisin levels, which in a “worst case” scenario (“high” food intake of maize-based products) could pose a potential risk for the consumer, in particular concerning foods for infants and young children. High fumonisin levels were found in infant foods in 1999, but contamination levels decreased strongly in the following years. HFBs (mostly HFB₁) were frequently found in processed cereals such as corn flakes, but in relatively low concentrations. According to our findings, the new European Union maximum levels for fumonisins are suitable to eliminate peak contamination levels of fumonisins in foods, but would lead to a regular excess of the TDI for infants and young children if these maximum levels would indeed be exhausted. Financial support: This work was financially supported by the German Federal Ministry for Nutrition, Agriculture and Consumer Protection, research grant 415-6080-1/60 (BMG alt).     

Estimation of genetic variation in residual variance in female and male broiler chickens

In breeding programs, robustness of animals and uniformity of end product can be improved by exploiting genetic variation in residual variance. Residual variance can be defined as environmental variance after accounting for all identifiable effects. The aims of this study were to estimate genetic variance in residual variance of body weight, and to estimate genetic correlations between body weight itself and its residual variance and between female and male residual variance for broilers. The data sets comprised 26 972 female and 24 407 male body weight records. Variance components were estimated with ASREML. Estimates of the heritability of residual variance were in the range 0.029 (s.e. = 0.003) to 0.047 (s.e. = 0.004). The genetic coefficients of variation were high, between 0.35 and 0.57. Heritabilities were higher in females than in males. Accounting for heterogeneous residual variance increased the heritabilities for body weight as well. Genetic correlations between body weight and its residual variance were -0.41 (s.e. = 0.032) and -0.45 (s.e. = 0.040), respectively, in females and males. The genetic correlation between female and male residual variance was 0.11 (s.e. = 0.089), indicating that female and male residual variance are different traits. Results indicate good opportunities to simultaneously increase the mean and improve uniformity of body weight of broilers by selection.     

Effects of Reproductive Status on Energy Intake,Ingestion Rates,and Dietary Composition of Female <Emphasis Type="BoldItalic">Cebus capucinus</Emphasis> at Santa Rosa,Costa Rica

The limiting factor for fitness in female primates is the acquisition of high-quality food, i.e., food that is high in energy and nutrients, such as protein, fat, and carbohydrates. Reproductive status can influence female feeding patterns, e.g., lactating females in some primate species consume greater quantities of food and specific nutrients than do nonlactating females. We examined the energy intake, ingestion rate, and composition of the diet in female white-faced capuchins in 3 reproductive states: lactating, gestating, and cycling. We observed 12 reproductively active females for 7 mo and compared their energy intake, ingestion rates, and intake rates of nutrients: protein, fat, sugar, and fiber. Lactating females took in significantly more energy per hour while feeding than pregnant and cycling females did. Lactating females also ingested significantly more food items per hour, but they did not spend more time eating than other females did, and they did not differ in the composition of their diet as measured by insect consumption and proportion of protein. Pregnant and cycling females did not differ from each other in any of the measures. We predicted that as the infants aged and began to move independently to forage and play, their mothers’ energy intake rates would decrease in concert with decreased energy demands by the infant. We found a statistically significant relationship between these 2 factors, with infant age explaining 75.4% of variation in mothers’ energy intake, supporting previous studies that found lactation to be the most energetically expensive reproductive state.     

Demographic and socioeconomic determinants of variation in food and nutrient intake in an Andean community

Understanding the sources of variation in a community's diet is vital for development work, as well as being a source of anthropological and cultural insights. Previous surveys in the South American Andes suggest that nutrient deficiencies may be widespread; however, such interpretations have remained tentative since variance in Andean populations' diet has not been thoroughly examined. In this paper we consider the variation in diet due to variation in age, sex, and socioeconomic status and variation attributed to inter- and intraindividual variation in the diet. One to six days of dietary data (mean = 3.1) were collected via 24 h recalls from 221 residents of a small, rural community in highland Ecuador. The contribution of various food groups to the diet varied with land holdings and age but not sex. For example, animal-derived foods contribute more and tubers contribute less to the diet of the households with ≥5 Ha, and sweets contribute more to the diet of children. The interindividual variation in energy and nutrient intake was low and the intraindividual variation high relative to developed countries. The consequence are twofold. First, because interindividual variability is low, group mean intake can be estimated relatively easily, facilitating group comparisons. Second, because intraindividual variation is high, individual nutrient intake cannot be easily estimated, which will decrease the ability to detect associations between nutrient intake and health measures. This knowledge of the sources of dietary variation can lead to better study and survey designs in the rural Andes and elsewhere in the developing world. Am J Phys Anthropol 105:407–417, 1998. © 1998 Wiley-Liss, Inc.     

The Roles of Sex,Mass and Individual Specialisation in Partitioning Foraging-Depth Niches of a Pursuit-Diving Predator

Intra-specific foraging niche partitioning can arise due to gender differences or individual specialisation in behaviour or prey selection. These may in turn be related to sexual size dimorphism or individual variation in body size through allometry. These variables are often inter-related and challenging to separate statistically. We present a case study in which the effects of sex, body mass and individual specialisation on the dive depths of the South Georgia shag on Bird Island, South Georgia are investigated simultaneously using a linear mixed model. The nested random effects of trip within individual explained a highly significant amount of the variance. The effects of sex and body mass were both significant independently but could not be separated statistically owing to them being strongly interrelated. Variance components analysis revealed that 45.5% of the variation occurred among individuals, 22.6% among trips and 31.8% among Dives, while R² approximations showed gender explained 31.4% and body mass 55.9% of the variation among individuals. Male dive depths were more variable than those of females at the levels of individual, trip and dive. The effect of body mass on individual dive depths was only marginally significant within sexes. The percentage of individual variation in dive depths explained by mass was trivial in males (0.8%) but substantial in females (24.1%), suggesting that differences in dive depths among males was largely due to them adopting different behavioural strategies whereas in females allometry played an additional role. Niche partitioning in the study population therefore appears to be achieved through the interactive effects of individual specialisation and gender upon vertical foraging patch selection, and has the potential to interact in complex ways with other axes of the niche hypervolume such as foraging locations, timing of foraging and diet.     

Energy intake and weight gain of very low birthweight babies fed raw expressed breast milk.

The energy intake and weight gain of low birthweight infants (under 1500 g) fed expressed breast milk were measured. Between the second and fourth weeks of life the mean energy intake was 577 kJ (138 kcals)/kg/day and the mean weekly increase in weight 119 g/week. Feeding energy-rich hind milk to two babies increased their energy intake but had little effect on their rate of weight gain. There appeared to be no correlation between energy intake and weight gain, probably owing to variation in the absorption of nutrients from expressed breast milk. This study forms a basis for a comparison of weight gain in babies fed alternative regimens of artificial milks.     

Relationship between oral health, nutrient intake and nutritional status in a sample of Brazilian elderly people

Objective: To evaluate the relationship between the oral health condition, the nutrient intake and the body mass index (BMI) in elderly people. Background: Impaired dentition has been associated with an inadequate consumption of key nutrients and with changes in nutritional status in elderly people. Materials and methods: The sample comprised 887 elderly people, aged 60 and over, of whom 816 underwent a clinical oral examination and were allocated into groups according to the numbers of teeth and number of posterior occluding pairs of natural teeth. Nutritional status was determined using the BMI. Dietary intake was assessed by a 24‐h diet recall interview and the data from these records were converted to nutritional intake using Diet Pro software. Differences between means were evaluated using anova , together with the Tukey test or Dunnet test, according to the normality of the data. Associations between categorical variables were tested using chi‐square analysis. Results: Ages ranged from 60 to 96 years (mean, 71.46 years), with 47% of the respondents in the 60‐to 69‐year‐old group. A high DMFT index (mean, 27.81) was observed with the missing component accounting for 88.8% of the index. Significant differences were observed between the mean intake of nutrients and the number of posterior occluding pairs of natural teeth (p < 0.05). No significant differences were found between the number of teeth and BMI. Conclusion: Nutrient intake is associated with the oral health status as defined by clinical measures.     

Infant care in the spectral tarsier ( Tarsius spectrum ) Sulawesi, Indonesia

I present quantitative and qualitative data on infant caretaking behaviors collected during a preliminary field study of spectral tarsiers (Tarsius spectrum),in a northern Sulawesi rain forest. The primary goal of the study is to identify the basic pattern of infant care in this species. I studied tarsiers at Tangkoko-Dua Saudara Nature Reserve in Sulawesi, Indonesia, from May to July 1992. I observed two infants, from two groups for a total of 96 hr using focal follows. During individual focal follows, ranging from 20 min to 6 hr, I recorded behaviors at 5-min intervals. I also recorded distances of group members relative to the infant at 5-min intervals. I subsampled the data at 35-min intervals to control for statistical autocorrelation between data points. Infants were alone between 40 and 50% of the time. The two subadults were more frequently in proximity to the infant than the adult males, the nonmatemal adult female, or the mothers were. This pattern of the subadults maintaining proximity to the infant continued when the mothers were absent. These results suggest that subadults may be guarding or babysitting infants. It is also possible that subadults are not traveling as far from the sleeping site as adults do and are therefore more likely to be found in association with the infant.     

Infant care in the spectral tarsier (Tarsius spectrum) Sulawesi,Indonesia

I present quantitative and qualitative data on infant caretaking behaviors collected during a preliminary field study of spectral tarsiers (Tarsius spectrum),in a northern Sulawesi rain forest. The primary goal of the study is to identify the basic pattern of infant care in this species. I studied tarsiers at Tangkoko-Dua Saudara Nature Reserve in Sulawesi, Indonesia, from May to July 1992. I observed two infants, from two groups for a total of 96 hr using focal follows. During individual focal follows, ranging from 20 min to 6 hr, I recorded behaviors at 5-min intervals. I also recorded distances of group members relative to the infant at 5-min intervals. I subsampled the data at 35-min intervals to control for statistical autocorrelation between data points. Infants were alone between 40 and 50% of the time. The two subadults were more frequently in proximity to the infant than the adult males, the nonmatemal adult female, or the mothers were. This pattern of the subadults maintaining proximity to the infant continued when the mothers were absent. These results suggest that subadults may be guarding or babysitting infants. It is also possible that subadults are not traveling as far from the sleeping site as adults do and are therefore more likely to be found in association with the infant.     

Multivariate interpretation of the foliar chemical composition of Norway spruce (Picea abies)

Twenty-four chemical elements were analysed by INAA, ICP-AES and CN in needles of Norway spruce (Picea abies (L.) Karst.). Branches were sampled from 54 trees on eight sites in Switzerland and South Germany. From each tree, twigs were sorted into the most recent four or five age classes and their needles analysed separately. All measured concentrations could be considered as log-normally distributed and statistical analyses were, therefore, performed on logarithms. Variance components were estimated by maximum likelihood and compared between elements. Non-essential elements varied more than essential nutrients (Mn was an exception). The sites and the age of the needles were the most important sources of variance. The interaction between site and age, the individual tree, the sampled branch and the residual variance were usually much smaller sources of variance. The effects of the most significant factors – site and age – were further described by principal components and cluster analyses. Mineral elements either increased or decreased with the age of the needles according to their mobility in the phloem. Two different components were identified in the effect of the sites: a geochemical component linked to soil pH and a climate component linked to altitude, temperature and precipitation. Multivariate statistics are discussed as a tool for the interpretation of complex interaction patterns between element concentrations in plants.