I noted the orientation of the vertebrae and the position of the arm bones. The remains were lying faceup, one arm crossing the chest, the other flung above the head.

Below the upper arms and chest lay a heart-shaped black mass with two fractured long bones projecting distal y. The pelvis. Beyond this, I could see the charred and fragmented bones of the legs and feet.

I felt relief, but some confusion. This was a single, ful y grown victim. Or was it? Infant bones are tiny and extremely fragile. They could easily be hidden below. I prayed I'd find none when I sifted through the ash and sediment.

I made notes, took Polaroids, then began to sweep away soil and ash using a soft-bristle paintbrush. Slowly, I exposed more and more bone, careful y inspecting the displaced debris, col ecting it for later screening.

LaManche returned as I was clearing the last of the muck that lay in direct contact with the bones. He watched silently as I took four stakes, a bal of string, and three retractable measuring tapes from my kit.

I hammered a stake into the ground just above the cranial cluster, and hooked the ends of two tapes onto a nail I'd driven into the top. I ran one tape ten feet south and pounded in a second stake.

LaManche held that tape at the second stake while I went back to the first and ran the other tape perpendicular, ten feet toward the east.

Using the third tape, I measured off a hypotenuse of fourteen feet one and three-quarter inches from LaManche's stake to the northeast corner.

Where the second and third tapes met, I hammered in a third stake.

Thanks to Pythagoras, I now had a perfect right triangle with two ten-foot sides.

I unhooked the second tape from the first stake, hooked it to the northeast stake, and ran it ten feet south. LaManche brought his tape ten feet east.

Where these tapes met I hammered in the fourth stake.

I ran a string around the four stakes, enclosing the remains in a ten-by-ten-foot square with ninety-degree corners. I would triangulate from the stakes when taking measurements. If needed, I could divide the square into quadrants, or break it into grid units for more precise observations.

Two evidence recovery techs arrived as I was placing a north arrow near the cranial cluster. They wore dark blue arctic suits, SECTION

D'IDENTITE JUDICIAIRE stamped on their backs. I envied them. The damp cold in the basement was like a knife, cutting right through my clothes and into my flesh.

I'd worked with Claude Martineau before. The other tech was new to me.

We introduced ourselves as they set up the screen and portable light.

"It's going to take some time to process this," I said, indicating the staked-out square. "I want to locate any teeth that might have survived, and stabilize