What Died?

One day, in the Before Time, Heather, I, and our beloved Lilly were out taking a walk on the grounds of a certain defunct Massachusetts mental hospital. The state had abandoned the place decades earlier, and though the buildings stood ominous and dilapidated, the surrounding woods — themselves adjacent to a community park called Rock Meadow — were a wonderful place to train for SAR, ski, ride bikes, or just play.

I think that day we were doing the latter: just walking around, enjoying a beautiful New England autumn day. Lilly was little more than a puppy; Heather and I were little more than children. I’m not much given to wistful reflection on times past, but I think that the three of us were pretty damned happy with each other that day.

Anyhow, we were walking across a clearing alongside the old, unpaved road that ran back from the rear of the facility when Lilly’s body went taut, her tail came up, and she began frantically sniffing at the ground.

Curious as to what she was scenting, I walked behind her, looking at the ground. I saw a puzzling series of rectangular stones set flush with the earth, inscribed with mysterious numbers: P-17; P-33; P-48. Then the numbers changed: C-54; C-22; C-12. Like that.

It wasn’t until I approached a stone dais bearing only the bottom of a crumbled statue that the penny dropped for me: all that remained was the feet, but they were unmistakable: emerging from under a dress or robes, they stood upon a snake.

This Sicilian boy didn’t need to be reminded of his iconography: this had been a statue of the Virgin Mary [1]. And the numbers now made sense: Protestant number 17; Catholic number 22.

We were in a graveyard — the place where the hospital had buried patients who had died without family to take their remains. And though it took me about five minutes to piece that together, Lilly had known almost instantly.

I can’t be sure, either then or today, whether that last statement is actually true. After 18 years of working and training with search dogs, even I have trouble believing what I saw that day: Heather checked later, that graveyard hadn’t had a new tenant in the 20 years since the facility had closed.

Like most wilderness SAR handlers, we had cross-trained our dog to find human remains. But we’d been thinking in terms of finding a fresh, whole body for those tragic but inevitable times when we arrived too late — not detecting the bare bones of a two-decade-old Potter’s Field. I still don’t really know for sure, wonder if my eyes had tricked me somehow.

Well, today’s offering speaks to this question in a fairly direct way: Arpad Vass and homeys, from the Oak Ridge National Laboratory and the FBI’s Laboratory Division, were able to detect 478 unique decomposition-associated gasses from over human graves at the University of Tennessee Anthropological Research Facility. Even more interesting, they observed this nasty bouquet beyond 18,000 “burial accumulated degree days” — the number of days the body was buried times the average daily temperature in that location. Since that part of Tennessee gets 5,234 BADD per year, that means that at least some of the smelly compounds are still going strong three and a half years after burial, and after all the soft tissues are gone.

Pretty amazing, really.

The researchers did a couple of things that puzzled me. They ranked 30 of the compounds they found by “perceived importance,” but don’t seem to define that term in this paper. It appears that an earlier paper from 2004 sampled gases from a decomposing exposed body and ranked them somehow, but I haven’t been able to get hold of it yet. My guess is somebody smelled the test tubes and the rankings correspond with each compound’s contribution to the stink, but I need to get hold of that paper to be sure.

The guys also had a puzzling series of pie charts that show “differences in bone odor composition” in dogs, humans, deer, and pigs — but instead of the intuitive series of a pie chart for each animal showing the percent of aldehydes, amides, alcohols, and ketones in that beast, a la:

they have one chart for each chemical compound with percentages for each animal, thusly:

It shows amply that the four critters have major differences even at this broad level, I suppose, but I’m stumped as to what each percentage means in the latter — they don’t add to 100, for each animal, between graphs. May just be relative proportions reduced to percentages, but I find that a bit confusing.

Anyhow, they pick up only 19 of the 30 gases from cadavers on the surface. They don’t know whether the missing compounds, only seen in buried bodies, are the product of anaerobic decomposition that can’t take place on the surface, or are products of the interaction between the scent gases and the dirt and its microbes, so that’s a question for another day. They also identified 12 of the 30 that emanate from human bones.

This paper pokes a hole in a couple of old dog handlers’ tales: One, that pigs are chemically similar enough to humans to stand in for us as “training materials” for cadaver dogs: the pigs showed profoundly different scent signatures. The other — a rumor circulating among us scent wonks, if not the general dog-handler community — was that polyamines such as cadaverine and putrescine would prove central to “death smell,” and were pretty much the whole sum of certain commercial artificial scents. As I’d understood that these compounds were pretty much characteristic of decomposition, their total absence was a bit of a stunner for me. Gotta get hold of that earlier paper, because it might cast a different light, but right now the only thing even reminiscent of a polyamine on the list is methenamine, at number 28, and as a cyclic it isn’t really the same thing. Other expected stinkies, like the sulfur-containing compounds — dimethyl disulfide and –trisulfide, sulfur dioxide, carbon disulfide, etc. — are present and accounted for.

Again, I need that earlier paper to find out what the rankings actually mean — either the polyamines aren’t as important as I’d thought, or they’re present in a body at the surface but not underground. The latter would be remarkable, given that polyamines are the products of the kind of anaerobic decomposition you’d expect from a buried body especially. So I’ve got more work to do in understanding this one.

While the authors trot out the old trope of the “robotic scent detector” — one that doesn’t really impress me, as a wilderness handler, because the localization problem in my field is far more difficult than the detection problem — but also come down on the side of the angels in that they’re interested in producing standard, verified, published scent tools [2] that can be used to train and verify cadaver dogs. While I remain not-yet-entranced by the idea of artificial scents — you’d better be damned sure you’re giving the dogs something that really is representative of the target scent to train them, or you’re screwed — I recognize that real “training aids” suffer from a huge amount of quality variation [3]. I’m skeptical, but more than willing to see where the research leads.

Most amazing, though, is how this study really leaves the door open for how long these scents persist. I hope the researchers left their apparatus in place, and intend to come back in 10 years or so and see what’s coming out of these graves.

I still don’t know for sure that Lilly had been scenting 20-year-old graves. But I certainly can’t rule it out.

[1] I just love the poetry of this one: the saint whose bywords are nurture, forgiveness, gentleness is the only one who treads Satan below unprotected feet.
[2] Hear that, you companies with proprietary mixes who won’t even release data on how you verified their efficacy? We really aren’t too dumb to understand this stuff, you can keep your secrets but just tell us why you think it works.
[3] Very reminiscent of herbal medicines.