Over the three days we will be keeping a diary of finds, comments, discoveries.

Day Two, 1.00pm

Here we are at perhaps the most exciting archaeological site that Britain has seen in many years – and it has been raining more or less heavily since at least 3am. Despite this, our dedicated archaeologists and superb diggers labour on, wet but far from miserable, given the continuing stream of terrific finds.

In Trench 1, a room with an apsidal (semi-circular) end has emerged. The experts think it may be a bath house because of the evidence of burning already found - this would have been produced by the heating system.

Trench 2 has produced traces of the room behind the main corridor. But the wall of this doesn't abut directly on to the corridor wall. Rather it stands a slight distance apart, and the space in between is filled with rubble. This seems to mean that the wall was rebuilt at some stage. This is interesting because it shows how the villa changed over time.

However, it is the new Trench 4 that is causing the most excitement. Only six inches below the surface, the archaeologists have found very well-constructed walls and at least four rooms. They have found all four internal walls of one of these, which are covered in red-painted plaster.

The plaster itself has a lot to tell us. The type used inside this room is opus plaster – a pink mortar mixed with crushed bricks or tiles – which differs from the lime-based plaster that was used on the corridor walls. The use of opus plaster hints at heating – but no evidence of a heating system. In addition, there are two coats of plaster, showing that at some time the room was redecorated.

Right at the edge of this trench they have found an adjacent room with a stone flag floor. They are now extending the trench to reveal even more of these intriguing features.

Despite the rain, a Roman chef has been brought in to demonstrate how the former residents of this site once ate. See Recipes for the menu.

This is where Time Team have started to dig. All evidence is showing that our villa is much bigger than we thought.

Sunday morning, and Trench 4 begins to get some serious attention.

Here are all the important finds so far. They range from late first-century AD to the early fifth century.

A view of the site, Day 2 at 3pm.

Born in a barn!
Sue, Steve and little Milly (our Time Team graphics experts) under cover in the incident room.

La Tene III Nauheim derivative brooch from the first century AD

Victor Ambrus sketching a reconstruction of our villa.

Day Two, 5pm

The site, which now measures about 100m x 80m, could be (says Mick Aston) as big as Woodchester, certainly bigger than Chedworth. At the moment, it doesn't look as big as Fishbourne, but who knows what might happen?

We're finding red and yellow striped plaster and lots of box tiles from flues signalling central heating and/or a bath house, but only a series of small rooms and not a big one. And still no mosaics.

According to Mark Corney of the University of Bristol, the site could resemble the temple complex at Lydney, Glos, with lots of small rooms for individual worship. Another place it could resemble is Gadesbridge Park in Hertfordshire.

Trench 2 (Phil's) will now be closed. Ash and other burned material have been found at different levels – possibly a stoke hole to heat air for a hypocaust (Roman central heating). Another bronze brooch was found in the spoil heap by metal detectorist Tim Hand.

Trench 1 (Carenza's) has produced walls, plaster but, disappointingly, no mosaics. However, Carenza may have the answer to this. She believes that the main corridor was covered in high-class paving stones, which were subsequently 'robbed'. The corridor was also covered in plaster painted with panels of bright red.

The floor of this trench is now covered with stamped earth including plaster from the walls. This indicates that people lived there when the villa was deteriorating and plaster was falling off the walls and being ground in underfoot. Carenza thinks these much later residents were squatters, but Mick disagrees, saying that it sounds just like the possible archaeological results of his own poor attempts at DIY! This trench may be extended to reveal an entire room; then it will be closed.

Work is being concentrated on Trench 4, where there is a complete room with substantial walls and remnants of paving. The trench is already two to three times bigger than it was first envisaged, because the diggers keep unearthing intriguing things around the edges. Just found: an iron hammer and today's mystery find!

Work is progressing fast on the new Trench 5, on the downward slope (off the left-hand side of the Time Team grid). More news of this when I get down the slippery slope in the rain to have a look!

Jenni helps to uncover the plaster wall from what we suspect is a Roman bath.

Italian maestro, Aurello Spagnolo cooks a banquet for the Time Team. Aurello's name derives from the reign of Roman emperor Marcus Aurelius (161-180 AD). Anybody from one of the Roman provinces during his reign who received Roman citizenship would have added the name Aurelius to their own name. You can see Aurello's wonderful menu.

Day Two, 7.30pm

The new trenches, 5 and 6, aren't producing much of note so far, although a green glass bead was found in 6. The horse's skull found in Trench 3 is causing comment: some of the experts here have speculated that it might have been a ritual offering.

Trench 4 is the star at the moment. The plunge pool being excavated by Phil and his team was once lined with red-painted plaster, and the floor surrounding it was covered in plaster painted a delicate shade of aqua blue. Evidence of burning has been found in one corner. Neil Holbrook, the excavation leader, said that above this burning material might have been a grate, and on top of this may have been an iron or a bronze boiler. The hot water from this would have been carried into the pool by pipes; the waste would have been taken, again by pipes, down the hill.

But the most exciting find was the first evidence of a mosaic: two tesserae (mosaic pieces) have been found in the rubble packed into the pool. However, this rubble would have been taken from another room, so it is likely that whatever mosaic was present was not in this room.