Bug Girl has introduced me to a fascinating piece of research that was published today in Proceedings of the National Academy of Sciences of the USA. A pair of researchers have turned their attention to the question of how aquatic organisms become trapped in amber (Schmidt & Dilcher, 2007).
Amber, as you may already be aware, is formed from resin exuded by trees hardening on contact with the air. While the resin is still fluid and sticky, small organisms such as insects may become trapped and covered. As the resin turns into amber, the engulfed organism is sealed off from the ravages of the outside world and preserved (image above is of an alate termite, and is from here). Good examples of such amber inclusions are arguably the best preserved fossils in the world, and provide a depth of information that is simply not possible with other preservation methods. Details can be preserved right down to the cellular level and smaller, while soft-bodied organisms such as nematodes and protozoa that are otherwise absent from the fossil record may be found.
As already indicated, exposure to air and drying is a vital component of the process of resin turning into amber, and so it has generally been assumed that amber must form terrestrially. Unfortunately for this assumption, aquatic organisms are not uncommon in amber. A number of suggestions have been made to explain how they could have gotten there - they could have been living in water sitting in hollows between branches (technically referred to as "phytotelmata"), or they could have been travelling between bodies of water, or they could have already died in the water before being blown by the wind into resin. As pointed out by Schmidt & Dilcher, none of these options is really a good enough answer. Terrestrial dispersal is just not an option for some obligate aquatic organisms found in amber, and wind dispersal of dead organisms wouldn't explain the numbers of aquatic taxa found (as well as seeming unlikely for soft-bodied organisms such as protozoa). Phytotelmata are also not common enough to explain the numbers found, especially as most amber deposits are believed to have been formed by conifers.
Schmidt & Dilcher decided to solve the problem by emulating amber formation in a patch of swamp forest owned by Dilcher. After "inducing artificial resin outflows from pine trees" (translation: hacking off pieces of bark with a handsaw), Schmidt & Dilcher observed what happened to the resin after it came into contact with water sitting in small pools on the ground.
Fig. 1. Resin in Dilcher's swamp forest east of Gainesville (Florida). (A) Pinus elliottii and other trees standing in the water. (B) Resinous exudate spreading at the water surface. (C) Small elongate resin pieces of up to 4 cm in length hanging at the water surface attached to the tree trunk. (D) Solidified resin piece at the water surface with trapped insects and anthers (arrows). (E) Subaquatic resin with enclosed detritus flowing downward on the tree trunk. (F) Subaquatic pillow-like resin piece of 5 cm flowing on the swamp floor. (G) Extensive subaquatic resin flow of 20-cm length on the swamp floor. (H) Pillow-like resin piece of 2.5 cm width in lateral view with trapped water beetle (arrow). (Scale bars: 1 cm.) (from Schmidt & Dilcher, 2007)
Some resin spread over the surface of the water in a thin film - while organisms could be trapped in this, it was not thick enough to be likely to preserve anything long-term. Blobs of resin could remain attached to the tree at the water surface, and organisms could become trapped on the underside of these. Finally, if there was enough resin to break the water surface, it formed pillow-like masses at the bottom of the water. Small organisms such as rotifers and mites could be trapped in the surface of the resin, burying themselves deeper in the mass as they struggled to escape. After a day or two, a thin hardened skin formed over the mass, generally excluding micro-organisms, but larger mobile animals such as water beetles could still break through the skin and become trapped (the authors did not comment on whether such animals were likely to be actively attracted, tarpit-wise, to food organisms already trapped). Resin masses could move and flow in the water, and small droplets of water that might also contain micro-organisms could become enclosed in the resin. As long as the resin remained liquid, bacteria and fungi could even grow in the mass, making them more likely to be preserved later.
As long as the resin remained covered by water, it didn't harden, but when the pool dried the mass hardened and solidified. As Schmidt & Dilcher explained in their paper, their observations have a number of taphonomic implications for using amber inclusions to reconstruct the environment they formed in. Inclusions would be biased towards fungi and bacteria growing in the resin and larger animals that could break through the skin. Because the window of opportunity for micro-organisms to become trapped before the skin formed was relatively small, such organisms were less likely to become inclusions.