One of the most unknown yet numerous insect groups in the Arctic, as well as in the world, is a small soil dwelling insect called the springtail (Collembola in latin). It is one of the oldest insects with a fossil record dating back 400 million years. It has been successful in colonizing all continents and today we know that there are about 6 500 known species that occur in soils from the Arctic to the Antarctic. In the Arctic, springtails are the most abundant terrestrial insect with densities ranging between 10,000-40,000 ind/m² and have a range of adaptations to cope with harsh climatic conditions.

The biology and diversity of Arctic springtails is a subject that has interested scientists on expeditions to the Arctic since the end of the 19th century. Thanks to data from earlier expeditions we now know that many species of Arctic springtails have circumpolar distributions though almost nothing is known of their movements on the tundra today and how they have colonized the Arctic since the last glaciation. Due to their small size (0.5–5 mm), their dependency on the soil habitat and their winglessness, the possibilities of their moving over and between continents are limited. Individuals can actively move over small areas ranging from a centimetre to 100 metres in a day depending on their size and the adaptations they have made to survive the dry conditions on the soil surface. Given the distances between continents and islands in the Arctic this does not explain their circumpolar distribution for which a few hypotheses have been put forward:

• they may have been passively dispersed by wind;
• they may have arrived ashore via drift wood;
• they may have been carried in soil attached to the feet of migratory birds breeding on the tundra.

Aims

During the Tundra Northwest 1999 expedition, the aim of this project is to study the colonization and dispersal rates of springtails. Two main questions are asked:

1. What is the colonization rate of springtails on a continental scale, versus their active dispersal on a local scale?
2. Can data of gene flow between populations show directions and rates at which springtails recolonized the Arctic after the last glaciations?

With behavioural assays of dispersal rates we can estimate the rate at which a population would move in to and colonize an area. By using molecular markers to measure relatedness, we can estimate the exchange of genes between populations on a continental and local scale. Data on gene flow between populations that are based on mitochondrial DNA can also be correlated to a time scale, making it possible for us to estimate when different populations were isolated.

Methods used in the field and on the ship

We have chosen to study three species of Arctic springtails that are common in the Arctic, all have circumpolar distributions and are expected to have different short range dispersal behaviours according to their size and adaptations to life in the soil.

At all 17 field sites visited on the Tundra Northwest 1999 expedition we collected soil, mosses and lichens that were potential habitats for springtails. The samples were brought back to the ship and the springtails were extracted using a heat gradient extractor which makes them leave the heated soil and choose a cooler area where they can then be collected alive. We aimed at collecting three populations (each containing 20-50 individuals) of all three species from each site. Hypogastrura concolor was found at 16 sites, and O. groenlandicus and A. bidenticulata at 11 of the 17 sites visited.

Gene flow studies

To measure the relatedness of the springtails we will mainly use two techniques. One is to analyse the polymorphism of allozymes, which is the variation of production of enzymes encoded by different genes. The other is a PCR based technique for analysing DNA that we are currently developing and which will make it possible for us to study mitochondrial DNA sequences. These analyses will be carried out on all three species and from all sites where they were found. Work started on this in October 1999 and we will need a further year of experimental work to obtain sufficient data concerning the colonization patterns of Arctic springtails.

Dispersal experiments

We performed dispersal experiments on board the ship with Hypogastrura concolor collected during the expedition. An experimental arena, a circular plastic cylinder (ø 20 cm) was used with a layer of soil with mosses and Arctic heather. A number of springtails were released in the middle of the cylinder and after 24 to 40 h at 20°C the soil was divided in to circles of 2 cm radius and the springtails were extracted from the soil at the different distances.

The pattern found is that most of the springtails remain close to where they were released and a few disperse up to 10 cm in a day. These data were fitted to a theoretical mathematical model of diffusion that gives a diffusion coefficient (D) for determining the area time ratio (cm²/day) that the average population of springtails randomly disperses into. The diffusion coefficient of  H. concolor in Arctic soils was 2.2 ± 1.8 cm²/day. This means that in one year with a temperature of 20°C for 45 days they will colonize 0.01 m², so roughly it will take about 100 years for a population to colonize 1 m². Compared to D-values of other species that have been analysed before, H. concolor belongs to a relatively fast moving species of springtail.

According to our first results, active movement by springtails cannot explain their successful colonization of the Arctic. A real test for establishing whether passive dispersal of springtails occurs over the vast Arctic is to combine the results of the experiments on active movement with data on gene flow between populations. If the actual gene flow is greater than predicted from the behavioural experiments, it will be a challenge to identify the mechanisms involved in the passive dispersal of these small Arctic insects.