Lots of great discussion points!
Some general comments: Remember that this study is an experimental model. It is left up to the individual reader to determine how applicable he or she feels the results are to their intended use of the product. Using this model, we were not able to demonstrate any differences between the odor-proof bags and ziplocs. This doesn't mean that there aren't any. As discussed in the article, the dogs work at close range and are trained to signal on even minute amounts of the substances in question.
On the other hand, I have yet to see any strong scientific evidence to suggest that the bags do work any better than ziplocs. Without that proof, and with reasonable evidence to the contrary, why should I blindly believe the theory?
In the end, I believe that our conclusion was appropriate and not overstated. The bags are not 100% odor-proof as advertised. Not only did the dogs find them, but even long after the bags had been removed, dogs would repeatedly signal on the open lockers that had previously held study bags. Scent had been transmitted to the immediate environment.
@Kyle- thank you for the kind words. I agree that proximity is a limitation of the study. However, the dogs' range is much longer than you think. The handlers like to talk about scent cones. Picture a 3 dimensional structure, heaviest at the source, and radiating outward in a cone-like pattern. The dogs are not actually searching for the object locker by locker as we would do, but instead run laps until they intersect the scent cone. It is very obvious when this happens as their behavior suddenly changes and they practically fall over themselves in their hurry to reverse direction and follow the cone. Study bags contained as little as 5 grams of scent, and as described earlier, this was vacuum sealed in freezer (foodsaver) bags, then wrapped in heavy canvas, then sealed in the study bag. So we are already talking about a trace quantity of scent that is doubly wrapped. Despite this, we saw the dogs hit scent cones from 10-15 feet away. Then they'd run straight to the appropriate locker. How big would the scent cone be when we are talking about 3-5 pounds of food, a stinky odor source approximately 320x larger than what we used and carried by the wind? I'm guessing it would be a pretty large cone.
@Luke and Jay- see my first post in this thread and the details above. Vacuum sealed freezer bags did not prevent odor transmission.
@John- we did not assess the nylafume bags, but may yet do so in an upcoming larger scale study. However, after this pilot study, my enthusiasm for doing so is low. I went into this first study not knowing what we would find (to the officers' credit, they never doubted the dogs for a second). Now that I've seen the dogs at work, I have little hope that a pack liner with a giant hole in the top would have any chance. We are hoping to look at some heat sealable mylar bags, but I will include a couple other odor-proof options if we can do so without making the project too cumbersome for the dogs and handlers.
@Rob, Bill, Stephen, and Herbert- Interesting discussion. We used the timed trials in our model to hopefully pick up on subtle differences between ziplocks and odor-proof bags, and were unable to detect a difference. To definitively answer your question, I think a large scale in situ study is probably needed.
@Redmonk- No false positives. When they knew, they knew, and they would bark and hit the lockers until acknowledged.
@Kirk- heat sealed mylar will be evaluated hopefully next month.
I appreciate all the comments and feedback even if, in the interest of space, I have not responded personally. Thanks again.