Three sampling operations were conducted over this area.

• On January 9 four samples were taken by the Gendarmerie, consisting of a dirt sample (P1), small quantities of surface soil (Q1) and depth soil below the crust (Q2), and an alfalfa sample.

• On January 23 the Gendarmerie took new samples at the request of the biological analysis laboratory. These vegetal samples were taken far away from the trace (about 20 meters away).

• On February 17 GEPAN took a new series of 8 vegetal samples. A new soil sample called P2 was collected outside of the trace area, 3 or 4 meters away.

The analysis began at CNES with the visual examination of sample P1 using a binocular enlarging lens. A selection of areas presenting an interesting appearance were later examined under a microscope.

The samples were then forwarded to various laboratories equipped for physical and chemical analysis in an effort to determine the element composition of the P1 and P2 samples, and to identify possible variations between the two samples. It was thought that such systematic comparisons might lead to the discovery of mechanical, thermal or radiation effects correlated with the phenomenon. (Samples Q1 and Q2 were not used at this stage). The following sections present the results of these independent analyses.

2. Visual and Microscope Analysis (CNES-Toulouse)

A series of photographs were taken using an optical microscope with low magnification. In these tests sample P1 was divided into two parts, respectively designated as P1a and P1b.

Sample P1a showed heavily compacted dirt with a crust 6 to 7 mm thick, predominantly composed of very dry limestone with only a few traces of dessicated vegetation in the form of moss. Curved striations are clearly seen on the surface, indicating that this dirt has been exposed to a rubbing effect that has resulted in the abrasion of some silicium grains (Figure 16). Further