Individuals of E. fetida were collected in a temporary heap of cow manure from a small farm near the University of Vigo (42○9′N, 8○41′W), in the northwest of Spain. Average annual rainfall and temperature in the zone range between 2500–3000 mm and 11–12 °C respectively, with lower rainfall and higher temperatures reached in the summer [11,12]. Nevertheless, during the sampling period, the total precipitation was 1820 mm, as is referred in Table 1. The population of E. fetida was limited to the manure heap, since this epigeic species is scarce or absent in soils without direct presence of organic amendments [7]. Earthworms were hand-sorted from five 0.25×0.25×0.20-m blocks around the heap in November (autumn) 1999, January (winter) and April (spring) 2000 to determine earthworm population dynamics. Summer samples were collected in August 2000 using only two blocks, since most of the manure heap was removed as every year in late spring to use it as source of organic matter in agricultural fields. In order to know the age distribution of the population, earthworms collected from each sample were separated in the laboratory into mature (with well-developed clitellum), preclitellate (with tubercula pubertatis), juvenile, and hatchling (0.01 g) stages. Earthworms were counted and weighed during each seasonal sampling. Cocoons were also counted and weighed from a subsample of 100 g of substrate. Average cocoon deposition per earthworm was estimated for each season from densities of cocoons and mature individuals. Mature earthworms were checked in the laboratory, using a dissecting microscope, for the presence of spermatophores, which indicates a recent mating [13]. Therefore, mating activity was estimated as the percentage of mature individuals with spermatophores.

Statistical analyses of earthworm data were done using SPSS (version 11.0.1) for Windows. Prior to the analysis, sampled values were normalized using a log(x+1) transformation. The effect of the season was analyzed by one-way ANOVA with the GLM procedure of SPSS. Post hoc comparisons between the different seasons were performed using the Tukey HSD test.

The density of E. fetida in the field population across all seasons was 4950±690individualsm−2, with a biomass of 1040±130g live weight m−2. Most of the biological traits of this population strongly changed through the year (Table 2). The number of earthworms increased significantly from autumn to spring, when the population showed its maximum density. In summer, the earthworm population density decreased until the lowest numbers. Earthworm biomass followed a similar seasonal trend, with the highest value in spring and the lowest in summer (Table 2). The density and the average weight of the cocoons were significantly higher in spring and lower in winter (Table 2).

The age structure of the population also changed over time. The mature-to-juvenile ratios were 11.4, 1.4, 1.7 and 0.1 for autumn, winter, spring and summer, respectively. The proportion of mature earthworms was almost constant during autumn, winter and spring and was very low during the summer, when almost 80% of the individuals were juveniles; the highest numbers of hatchlings were found in spring (Table 2). Mature, preclitellate and juvenile earthworms were larger in winter and autumn than in spring and summer (Table 2). Mature earthworms bearing spermatophores were present in the population during the whole year and were more abundant in spring (Table 2). Mating activity was significantly higher in spring and autumn. The number of cocoons per mature earthworm was similar throughout the year (Table 2).

Seasonality had a strong influence on the growth and reproduction of E. fetida and consequently on the age structure of the population. This population was characterized by a very high density and biomass with a high proportion of mature individuals throughout the year. This resulted in an unusual population structure where the young earthworms were not the most abundant stage, as in other earthworm species [5]. High densities are common in E. fetida populations when large quantities of organic matter are available, supporting up to 14 600 individuals m⁻² [7]. Moreover, E. fetida lives in environments where the food and the living substrate usually are the same [9] and this resource availability should improve the conditions for earthworm growth and reproduction, and promote the occurrence of high numbers of mature individuals.

In our study, the average size of the mature earthworms ranged between 0.30 and 0.40 g, as it was reported by Watanabe and Tsukamoto [10], although E. fetida can reach nearly 1.6 g in laboratory conditions [14]. The low mean annual temperature in the study area (11–12 °C) could explain the observed size of the mature earthworms [15]. Furthermore, the high population density likely had a strong effect on mature weight, since Neuhauser et al. [16] and Reinecke and Viljoen [17] reported a weight of adult individuals below 0.40 g for population densities equivalent to 2000 and 2700 individuals m⁻², respectively.

Unexpectedly, maximum mean weight of adult earthworms was achieved in winter, at the lowest temperatures, but these large-sized earthworms showed the lowest mating activity. Temperature regimes below 15 °C cause low growth rates [18], but also affect to reproduction diminishing cocoon production and increasing incubation time [15,19]. The low reproductive activity found in winter could lead to weight gain due to decrease of reproductive costs and reallocation of resources toward growth [4]. According to this, the mean weight of the mature earthworms decreased in spring with the increasing reproductive activity. Similar results for resource reallocation to growth were reported by Cluzeau et al. [20] when adult individuals of E. fetida were isolated or reared in groups after mating.

The decline of the mature stage in summer was another feature of the population of E. fetida. The marked decrease in density of mature earthworms was probably caused by the strong disturbance caused by the removal of most of the manure in late spring. After this event, small juvenile and newly hatched individuals became the most abundant stages in the summer population.

Mating processes and production of cocoons in the population of E. fetida occurred throughout the whole year, indicating that there were some favourable and stable conditions to reproduction, even in winter. According to this, in the studied population there was not seasonal effect in the number of cocoons relative to the number of adult earthworms, although the highest mating activity occurred in spring. This mating peak could be due either to the optimal environmental conditions that lead to a maximum in the population traits measured or to an increase in the number of available mates. In fact, the availability of mates is one of the factors involved in the size-assortative mating of earthworms [21,22].

Changes in the density of cocoons over time seem to be related with the density of mature earthworms, despite the lack of significance in their density between winter and spring. These results agree with those of Watanabe and Tsukamoto [10], who found that cocoons were recorded mainly in autumn and spring, being affected by both the high and low temperatures in summer and winter, respectively. We found that cocoons were smaller in winter and larger in spring, and it is interesting to note that the smaller cocoons were produced by the heavier mature earthworms. Then our data suggest that E. fetida is able to regulate cocoon size in response to environmental factors. Currently, it remains unclear whether the size of earthworms is the only factor governing cocoon size [14,23,24], but in E. fetida the weight of the cocoon influences the number of progeny per cocoon [10,23,25]. We suggest that this ability to regulate the size of the cocoons has important implications in the development of the population. Thus, although in our study there are no significant differences in the number of cocoons in relation to the number of adult individuals between seasons, changes in offspring production and population growth rates can occur anyway.