surrounding tissue. The animal can then readily shed 

 the quill and escape its injured adversary. 



But battles with predators do carry the risk of 

 injury to the porcupine. I once examined the 

 skull of a porcupine at the University of Wiscon- 

 sin Zoological Museum in Madison that bore silent 

 witness to a violent encounter. The skull dates from 

 the 1890s, when wolves and wolverines shared the 

 porcupine's habitat. The skull was indented on top 

 and partly flattened by the jackhammer impact of 

 a canine's crunching down. Subsequent healing 

 shows that the porcupine survived, yet the margin 

 between life and death must have been thin. 



To avoid such battles, the porcupine issues warn- 

 ings, and its primary warning signal is olfactory As 

 a porcupine waits for an attack, quills erect, it pours 

 out a wave of pungent odor to signal that its foe 

 would do better to back off. 



The odor is generated by a patch of skin called the 

 rosette, on the porcupine's lower back. Specialized 

 quills growing out of the rosette help broadcast the 

 smell. Biologists have long noted modified hairs that 

 disseminate odors in other mammals: black-tailed 

 deer, the crested rat of East Africa, several bat spe- 

 cies, and others. Such hairs are called osmetrichia, 

 and they differ from ordinary hairs in having in- 

 creased surface area and in their ability to stand erect 

 when the animal is on alert. The greater surface area 

 holds more odorant molecules, and the erectability 

 helps disseminate the molecules into the air. In por- 

 cupines, the barb-covered section of the quills in the 

 rosette area is longer than it is on quills of the upper 

 back, and the rosette barbs themselves have the 

 greater overlap. Both effects increase the surface area 

 of the rosette quills. 



Just as the swat by Loretta got me hooked on how 

 quills exit the porcupine, another nighttime en- 

 counter propelled me on the path to identifying the 

 porcupine's warning odor. Passing under an apple tree 

 in the dark, I sensed an alarmed porcupine on a branch 

 above me simply by its wave of smell. That warning 

 smell has a penetrating quality somewhat similar to 

 the smell of goat or perhaps an exotic cheese. 



I asked David C. Locke, a chemist at Queens Col- 

 lege in New York City, whether he could help me 

 identify the warning-smell molecule. David operates 

 a gas chromatograph-mass spectrometer 

 (GC-MS). The gas chromatograph 

 sorts a gas mixture into its various 

 components, according to the rate at 

 which each component gas leaves the 

 system. Then the mass spectrometer helps 

 identify each component by its molecular mass 

 and fragmentation pattern. 



Retinaculum 

 (anchoring tissue) 



Root of a porcupine quill (left) is held erect and tightly in place 

 by a spool of connective tissue and a contracted piloerector 

 muscle when the porcupine senses imminent danger. If the 

 quill strikes another object, it is driven back into the porcu- 

 pine's body and through the immobilized spool (right), shear- 

 ing the attachment of the root to surrounding tissue. The quill 

 can then be readily detached from the porcupine. 



Before the instrument can work its magic, how- 

 ever, a smell must be captured. A portable air 

 pump draws the odor through a cartridge that con- 

 tains charcoal, silica gel, or some other odor-absorb- 

 ing compound, until the compound is saturated. The 

 odor is released, or "desorbed," by heating or adding 

 a solvent, and passed through the GC-MS. 



I set to work capturing a porcupine, securing it 

 in a picnic cooler, then sucking air through the cool- 

 er and through a cartridge. David advised drawing 

 the air for at least two hours to saturate the cartridge, 

 but that's much longer than I usually keep my guests. 

 Besides, this porcupine has other ideas. Alter a quar- 

 ter hour, it grows bored with its tight, dark enclo- 

 sure and begins chewing the plastic walls of the cool- 

 er to get out. I hurriedly release the animal, then 

 continue to pump air through the damaged cooler, 

 which reeks ol angry porcupine. 



David desorbs the cartridge and runs it through 

 the GC-MS. Now comes the first reality check. 

 What the printout reveals about the cooler's envi- 

 ronment is not what is perceived by the human 

 nose. Instead of the smell of angry porcupine, the 

 GC-MS detects a jumble of thirty compounds. 

 The biggest component is naphthalene — the active 

 ingredient in mothballs. Belatedly, I remember that 

 in the back room of the cabin, where I have set 

 up the cooler and air pump, there is a mothball- 

 filled clothes closet. Another major component is 

 identified as a plasticizer — released when the por- 

 cupine started demolishing the plastic lining of 



March 2006 NATUU.A1 MSTORV 



5 I 



