OLFACTORY SENSITIVITY OF THE TURKEY VULTURE (CATHARTES AURA) TO THREE CARRION-ASSOCIATED ODORANTS

STEVEN A. SMITH • AND RICHARD A. PASELK Departmentsof BiologicalSciences and Chemistry, Humboldt State University, Arcata,California 95523 USA

ABSTRACT.--TheTurkey Vulture (Cathartesaura) is generally thought to rely on olfactory cuesto locate carrion. Becausevertically rising odorantsare dispersedrapidly by wind tur- bulence, we predict that Turkey Vultures should be highly sensitive to these chemicalsto detect them at foraging altitudes. Olfactory thresholdsto three by-productsof animal decomposition(1 x 10-6 M for buta- noic acid and ethanethiol, and 1 x 10-5 M for trimethylamine) were determined from heart- rate responses.These relatively high thresholds indicate that these odorantsare probably not cuesfor foraging Turkey Vultures. Odorant thresholds,food habits of Turkey Vultures, and the theoretical properties of odorant dispersion cast some doubt on the general impor- tanceof olfaction in food locationby this species.Received 23 September1985, accepted 3 March 1986.

THEsensory modality by which Turkey Vul- Companydiscovered that natural gas leaks could tures (Cathartes aura) locate carrion has been be tracedby injectingethanethiol into gaslines debatedby naturalistsfor nearly 140 years(see and patrolling the lines for Turkey Vultures Stager1964 for review). Most of the controver- that, ostensibly,were attractedto the metcap- sy concernedwhether olfaction or vision was tan (Stager 1964). Stager (1964: 56) concluded the more important sense,although other the- from anatomical examinations and field tests ories included an "occult" sense (Beck 1920), that the Turkey Vulture "possessesand utilizes the noiseof carrion-eatingrodents, or the noise a well developedolfactory food locatingmech- of carrion-eatinginsects (Taber 1928, Darling- anism." ton 1930) as attractingTurkey Vultures to their If Turkey Vulturesrely on olfactorycues to prey. find food, foraging altitudes and search pat- Cathartesaura is now known to possessan terns should be a function of their ability to anatomically well-developed and physiologi- respondto concentrationgradients formed by cally functional olfactory system.Comparative carrion odorants emitted during decomposi- anatomical studies indicate that the relative size tion. Numerous mathematical models predict of the Turkey Vulture olfactory bulb is the odorant concentrationsalong Cartesian coor- eighth largestof 108 avian speciesreported by dinates downwind from the odorant source. Bang and Cobb (1968). The olfactory tubercle Most commonlyused is the Gaussiangas dis- in this speciesis scrolled and lined with an persionmodel, which assumesthat the concen- epithelium innervated by the olfactory nerve trationgradient is normally distributedin three (Bang 1960, 1971; Stager 1964). Single-unit dimensions (Strom 1976). Odorant concentra- neural responseshave been recordedfrom the tions at a given point in space decreasewith olfactory epithelial receptors (Shibuya and respect to an increase in the parameters that Tucker 1967), and heart- and respiratory-rate affect odorant dispersion.Because wind is the changesare known to occurwhen vultures are principal dispersing agent for gases released presentedwith olfactorystimuli (Wenzel 1965, into the atmosphere(Bosseft and Wilson 1963), Wenzel and Sieck 1972). In 1938 the Union Oil the model predictsthat dispersionof a given odorant concentrationis greatestin the direc- tion of the prevailing wind (x-axis). Because xPresent address: Department of Wildlife and the interaction of wind shear with the ground Fisheries Sciences,Texas A&M University, College producesturbulence, dispersion is least along Station, Texas 77843 USA. the vertical axis above the emission source (z-

586 The Auk 103:586-592. July 1986 July1986] Olfactionin theTurkey Vulture 587 axis). Bossertand Wilson (1963) applied the amine, and I x 10-8 , I x 10-7 , and I x I0 6M for Gaussianmodel to an analysisof olfactorycom- ethanethiol; these concentrations were chosen from munication in animals and provided maxima a preliminary screeningof a wide range of odorant solutionsfor the extent of odorant dispersal dilutions]. Odorant dilutions were presentedin or- along the x- and z-axesas a function of an an- der of increasingconcentration, with a control odor- ant consisting of distilled water randomly inserted imal's olfactory threshold. These solutions de- within each dilution series. A minimum of 5 rain scribe the maximum downwind distance (xm•) elapsedbetween presentations of test dilutions.All and altitude (z•) for an odorant concentration three chemicalodorants were usedduring eachtrial, that is above the threshold level of an animal with the order of presentationdetermined randomly. responding to olfactory cues. To minimize mixing the odorants,an exhaustfan was Although it is generally acceptedthat Tur- usedfor a minimum of 15 rain betweenpresentations key Vultureslocate carrion from olfactorycues, of the different chemicals. A total of 66 trials was run olfactorythresholds to carrion-associatedodor- usingeach odorant dilution series;individual sample ants have not been reported for this species. sizes for the four vultures were 15, 16, 17, and 18 trials. Presumably,their olfactory thresholdsare low Heart rateswere determinedby electrocardiogra- enough for them to detect carrion odorantsat phy (ECG).Differences in heartrate were determined foraging altitudes. Solutions for x• and z• by comparingthe number of beats in 5-s intervals provide a model that can be used to examine both before and after the odorant reached the bird the likelihood of specificcarrion odorantsserv- (0 s). Net changeswere establishedby adding the ing as olfactory cues to foraging Turkey Vul- difference in the number of beats between the time tures. intervals 0-5 s and -5 to 0 s, to the difference in We determined approximate olfactory re- the number of beats between the time intervals 5-10 sponsethresholds to three odorants (butanoic s and 0-5 s. A Kruskal-Wallisanalysis was used to acid, ethanethiol, and trimethylamine) associ- compareheart-rate changes within treatmentsto de- termine whether the data from individual birds could ated with animal decomposition.Butanoic acid has the characteristic odor of rancid fat and is be pooled.The Wilcoxontwo-sample test compared heart-ratechanges associated with eachodorant con- a by-productof protein, carbohydrate,and fat centrationagainst changes recorded during the pre- decomposition(Wells 1907).Several thioIs, in- sentationof the control.Significant (P < 0.05)changes cluding ethanethiol, are formed from the in heart rate between the control and odorant dilu- breakdown of sulfur-containing amino acids tions were interpreted as indicators of olfactory (Wells 1907)and are the compoundsused com- thresholds. mercially to give natural gas its odor. Trimeth- The Turkey Vulturesused in this studywere adults ylamine has a "fishy" odor and is releaseddur- of unknown sexand age. Three of the four had sus- ing muscletissue decomposition (Grey and Lea tained permanent wing injuries and were obtained 1969). In addition to the association of these from raptor rehabilitationcenters or other research- ers; there was no evidence that these injuries had odorants with animal decomposition, both affectedthe olfactorysense of these animals. The ethanethiol and butanoic acid are known to fourthbird wasphysically sound and hadbeen raised elicit olfactoryresponses in the Turkey Vulture in captivity.All four birds were maintainedon a dai- (Stager 1964, Wenzel and Sieck 1972). ly diet of meatand were weighedbiweekly to mon- itor their health. Considerabletime was spent habi- MATERIALS AND METHODS tuating the vultures to the olfactometerand the ECG EXPERIMENTAL DESIGN equipment.Test sessions always preceded daily feed- ings. Birdswere not testeduntil minimal heart rates Responsethresholds to olfactorystimuli were de- had been achieved. To minimize odorant habituation termined by continuousmonitoring of a vulture's and stress, the birds were not tested more than once heartrate before and during the presentationof three every three days. concentrationsof ethanethiol (ethyl mercaptan),tri- Olfactometer.--Theolfactometer was a modification methylamine,and butanoicacid (n-butyric acid). Our of the simple injectionolfactometer described by method was a modificationof the techniqueem- Moulton (1973).Our systemvaporized dilute odorant ployedby Wenzel(1965) and Wenzeland Sieck(1972). into an air stream to achieve the desired odorant con- Eachof four TurkeyVultures was repeatedlytested centration. Diluted odorants were drawn from 15-ml with three dilutions of each test odorant [1 x 10-7, arepulesby a peristaltic pump at 2.9 x 10-6 1/s I x 10-6, and 1 x 10-5 moles odorant/liter air (M), through0.51-ram (ID) standardpump tubing and al- calculatedat STP, for butanoicacid and trimethyl- lowedto drip ontoa glass"T" via a 20-gaugestain- 588 SMITHAND PASELK [Auk,Vol. 103

lesssteel hypodermicneedle insertedthrough a rub- ber diaphragm. The glass T was wrapped with •A nichromewire and electricallyheated to give an in- stantaneousvaporization of the odorant.A regulated and calibrated airflow (3.9 x 10 2 I/s) connected to the glassT carriedthe vaporized odorantto a mixing chamber. -5 0 5 10 The mixing chamberwas constructedfrom a glass SECONDS powder funnel attachedto a 58 x 99-mm glasstube Fig. 1. Sample ECG recordsshowing (A) no net with a perforated Plexiglasbaffle placed at the junc- changein heart rate following stimulation with dis- tion of the funnel and glasstube. The mixing cham- tilled water, and (B) a net heart-ratechange of 4 beats ber was positioned through a hole in the front of a following stimulation with 1 x 10-6 M ethanethiol. closed compartment (63 x 33 x 30 cm, constructed from plywood) in which the test bird was placed,and adjustedfor optimal placementaround the vulture's head. A muffin fan exhausted odorants from the com- equal to 0.4 and 0.2 cma, respectively (see Sutton 1953), u is wind speed in cm/s, and n is a constant partment into an outside vent. equal to 0.25 for wind speeds between 100 and 500 After each test sessionthe mixing chamber,glass cm/s. Becauseempirical measuresof Q for decom- T, and animal compartmentwere washedwith soapy posing carcassesare unavailable, we assumedvalues water, rinsed with distilled water, and dried. The ap- of Q between 1 and 20 moles/day (7 x 10•s - 1.4 x paratus was then reassembled,and compressedair 102ømolecules/s) as plausible estimates.For butanoic was allowed to flow through the system for at least 60 min. acid (88 g/mole), Q representsa daily carcassmass lossbetween 88 and 1,760g. For rodent-sizecarcass- Heart-ratemonitoring system.--Heart rates were re- es, we suggestthat these estimatesfor emissionrates corded with an ECG assembledfrom two preampli- are likely to be high. BecauseXm• and z• vary di- fiers wired in series and connected to a strip-chart rectly with the rate of odorant emission,however, recorder.The first preamplifier filtered and amplified our estimates of Q result in excessive odorant dis- signals between 0.3 and 10 Hz; the secondpreampli- persion distancesand, thus, provide a cautioustest fier boostedthe filtered signal. Electrodesfor the ECG of our hypothesis.Similarly, wind speedwas conser- were constructedfrom 22-gaugex 25-mm stainless vatively assumedto be 100 cm/s becauseodorant dis- steel hypodermic needlessoldered to shielded wire. persion distancesvary inversely with wind speed. Two electrodeswere placed subdermally on either side of the pectoralregion near the axilla. A ground electrode was placed subdermally on the vulture's RESULTS AND DISCUSSION back. The rationale for inferring olfactory re- GAS DISPERSION MODEL sponsesfrom changesin heart rate is that the presentationof a novel stimulus to an animal The Gaussianform of the various atmosphericgas tested in a homeostatic environment results in dispersion equations (as modified by Bossert and a group of autonomic nervous system re- Wilson 1963) was used to evaluate olfactory thresh- old values as functions of maximal downwind and sponses(Wenzel 1971). These generally in- altitude distancesfrom a hypothetical carcass.Rele- clude changesin heart and respiratoryrates vant solutions for the maxima are: but may involve other autonomicactions (Uno and Grings 1969). Wenzel (1973) demonstrated that the heart-rate response in pigeons de- Xm• \K•rCyC•u/ creasessignificantly after bilateral sectioning of the olfactory nerve. Monitoring heart-rate and changesto study olfactory perception in birds has been used in several investigations(Neu- z• •\K•CyC•ue/' harts 1963; Wenzel 1965, 1973; Wenzel and Sieck 1972). Although these studies were not de- where x• and zm• are distances(cm) downwind and signed to determine threshold levels, Schmidt above the threshold level of an animal responding (1975)used heart- and respiratory-ratechanges to the odorant. Q is the rate of odorant emissionfrom to determine olfactory thresholds in three a sourcein molecules/s, K is the threshold response speciesof neotropicalbats. Similarly, respira- in molecules/cm3, Cyand C• are diffusivityconstants tory-rate changeswere used to evaluate olfac- July 1986] Olfactionin the TurkeyVulture 589

2600

50

• 4o- O x • Z •m10-

-10 i I [ 167 166 155 Fiõ. 3. The relationshipbetween Xm• and zm• for BUTANOIC ACID different odorant emissionrates (Q). Calculationswere basedon an olfactor• threshold (K) oœI x 10-6 M and a wind speed of 100 cm/s.

25 tory thresholdsin Black-billed Magpies (Pica pica)and pigeons(Columba sp.; Snyder and Pe- terson 1979). Zeaman and Wegner (1954) de- termined that heart-rate responseswere as ac- curate indicators of threshold levels as verbal responsesin auditory-perceptionstudies of hu- mans. 5- All presumptiveolfactory responses were as- sociatedwith an increasein heart rate (Fig. 1). Becausechanges in heart rate were consistent within treatments across all birds used in this study, the data were pooled for the pair-wise [ I 156 167 1•6 comparisons between controls and diluted odorants.A significant(P < 0.008)mean heart- ETHANETHIOL rate increaseof 6.8 beatsper minute (BPM) was c recorded at ! x 10 -6 M for butanoic acid, al- though nearly significant(P < 0.09) changes were observedat 10-v M (Fig. 2A). Threshold concentrationswere less ambiguousfor eth- anethiol and trimethylamine.A mean heart-rate increase of 10.3 BPM (P < 0.0001) indicated a thresholdresponse to ! x 10-6 M ethanethiol • 10- (Fig.2B). Trimethylamine concentrations of 10 s M resulted in a significant (P < 0.01) mean heart-rate increaseof 13 BPM (Fig. 2C).

Fig. 2. Mean heart-rate changesfor odorant con- centrationsof three by-products of carrion decom- 1•7 166 155 position. Solid circles represent mean heart-rate changes.Vertical lines through the means indicate TRIMETHYLAMINE 95% confidence intervals. C indicates controls. 590 SMITHAND PASELK [Auk, Vol. 103

TABLE1. Predicted olfactory thresholds(K) at various maximum altitudes (zm•) and odorant emissionrates (Q). Values of Q are given in different units for comparativepurposes. Values of Q above the line are assumedto representplausible emission rates. Those below the line are obviouslyhigh but are included for comparison.Calculations are for a wind speed of 100 cm/s.

Q K l/s Moles/day z• = 3 z• = 10 Zm• = 60 2.6 X 10 -• 1 1.5 X 10 -•0 1.3 X 10 -• 3.8 X 10 -•3 5.2 X 10 -4 2 3.0 X 10 -•0 1.3 X 10 -• 3.8 X 10 -•a 7.4 X 10 -4 3 4.3 X 10 -•0 3.8 X 10 -• 1.1 X 10 -•2 1.3 X 10 -s 5 7.6 X 10 -•0 6.8 X 10 -• 1.9 X 10 -•2 2.6 X 10 -3 10 1.5 X 10 -9 1.4 X 10 -• 3.8 X 10 -•2 5.2 X 10 -3 20 3.0 X 10 -9 2.7 X 10 -•0 7.6 X 10 -•2

2.6 x 10 -2 100 1.5 x 10 -• 1.3 x 10 9 3.7 x 10 -• 2.6 x 10 -• 1,000 1.5 x 10-7 1.3 x 10 8 3.7 x 10 -•0 2.6 10,000 1.5 x 10-6 1.3 x 10 • 3.8 x 10 -9

GAS DISPERSION MODEL OLFACTORY CUES AND FOOD LOCATION

The extent to which atmosphericturbulence Stager's (1964) anatomical investigations of near the ground impedes the vertical rise of the olfactoryorgan and innovative field studies odorantsat given concentrationsis reflected in of the possible ecological role of the Turkey the considerablysteeper slope of Zm• relative Vulture olfactorysystem has led to the general to Xm• (Fig. 3). For an olfactory threshold of acceptancethat C. aura relies to a great extent 1 x 10-6 M and what is probably a high odor- on this sense to locate carrion. There are, how- ant emissionrate (20 moles/day), z• extends ever, discrepanciesbetween the observed re- only to 0.17 m abovethe source.If Turkey Vul- sponseof Turkey Vultures to ethanethiol, as ture olfactory thresholdsto butanoic acid and reported by Stager (1964), and the predictions ethanethiol are 1 x 10 6 M, this analysis sug- of our model basedon the apparent threshold geststhat they should be able to detect these levels of this odorant. Stager (1964) observed odorants only at exceedingly low altitudes. Turkey Vultures respondingto ethanethiol at Furthermore, the odorant concentration at 1 m approximately61 m altitude and 183 m down- is above threshold only if the emission rate is wind from the odorant source.Our model pre- impossiblyhigh (2.6 l/s; Fig. 3). For butanoic dicts (Table 1) that thresholdsof roughly 1 x acid, this emission rate yields an exceptional 10 •2to 1 x 10-• M would be necessarybefore daily carcassmass loss of 8.8 x 10s g. a Turkey Vulture could detect ethanethiol at Solving the z• equation in terms of the this altitude, whereas our threshold data indi- threshold value and setting z• at various al- cate that responselevels to this odorant are titudes between 3 and 60 m suggeststhat ol- considerablyhigher at approximatelyi x 10-6 factory thresholdsbetween 1 x 10-•ø and 1 x M. In his experiment, Stagerforced pure etha- 10-'2 M are necessaryto detect any carrion-as- nethiol under pressure into the atmosphere. sociated odorant for emission rates between 1 Under this experimental design, ethanethiol and 20 moles/day (Table 1). Although these emissionrates (Q) were likely to have been ex- odorant concentrationsare considerablylower cessive,causing threshold-level concentrations than the threshold values for Turkey Vultures, to occur at greater distancesboth downwind they are within the perceptual range of hu- and above the sourcethan normally would be mans and some bats. Humans respond to bu- producedby a decaying carcass. tanoic acid and ethanethiol at concentrations In the context of the odorant-dispersion as low as 1 x 10-•' M (Phelps 1976) and to 1 x model presentedhere, our analysisof olfactory 10-•ø M trimethylamine (Zwaardemaker1926). sensitivity in the Turkey Vulture strongly sug- Schmidt(1975) reported olfactory thresholds of gests that the odorants used in our investiga- the vampire bat (Desmodusrotundus) to buta- tion are unlikely carrion cues and castssome noic acid as 1.5 x 10 -•ø to 1.5 x 10 -• M. doubton the probableimportance of this sense July1986] Olfactionin the Turkey Vulture 591 as a food-locatingmechanism. Admittedly, the ACKNOWLEDGMENTS Gaussianmodel is simpler than the phenome- We especially thank Dr. D. H. Brant for his advice na it describes. Nonetheless, odorant concen- and unfailing support.Technical assistancewas pro- trations are in some manner dependent on vided by R. Nocon, M. Reed, and S. Wycoff. The De- emissionrates. The absolutequantity and rate partment of Wildlife Management (H.S.U.), the San of odorantemission depends on the massof the Diego Wild Animal Park, the Alexander Lindsey, Jr., carcassand the rate at which decomposition Museum, and Dr. M. Fry (University of California, occurs.Unlike other cathartids,Turkey Vul- Davis)provided the Turkey Vultures. Commentson tures appear to feed primarily on small-bodied early draftsof the manuscriptwere kindly given by prey suchas rodentsand snakes(Stager 1964). K. Arnold, D. Cornmuzzle,J. Ensink, D. Hale, H. Hunt, There is some evidence to suggestthat they and P. Weldon; Drs. T. D. Johnstonand B. M. Wenzel provideduseful suggestions on the final draft. Partial also prefer freshly killed carcassesover those funding was furnishedby the BiologyGraduate Stu- that are putrified (Owre and Northington 1961). dent Association(H.S.U.). This paper represents, in Both of these considerationssuggest that Tur- part, a Master's thesis (by Smith) submitted to the key Vultures prefer to feed on carrion that Departmentof BiologicalSciences, Humboldt State would be expected to yield relatively little University. odorant. Detection of odorants from freshly killed, small-bodied animals even at low for- LITERATURE CITED aging altitudeswould not be expectedunless BANG,B.G. 1960. Anatomicalevidence for olfactory threshold levels were also low. function in some speciesof birds. Nature 188: The intent of this paper is not to categorical- 547-549. ly refute the generally acceptedrole of olfac- --. 1971. Functionalanatomy of the olfactory tion in C. aura. Rather, we have attempted to systemin 23 ordersof birds.Acta Anat. 79 (suppl. addressproblems associatedwith odorant dis- 58): 1-76. persionand altitude asthey pertain to olfactory --, & S. Co•. 1968. The size of the olfactory cues, and to establish a theoretical framework bulb in 108 speciesof birds. Auk 85: 55-61. by which future investigationsmight proceed. BECK,H. H. 1920. The occult senses of birds. Auk 37: 55-59. Accordingto the model we present, olfactory BOSSERT,W., &: E. O. WILSON. 1963. The analysisof sensitivity in Turkey Vultures should be com- olfactorycommunication among animals. J. The- parableto that of mammalsif olfactorycues are oret. Biol. 5: 443-469. to be detectedat foraging altitudes below 10 m DARLINGTON,P., JR. 1930. Notes on the senses of (Table 1). While our data do not support this vultures. Auk 47: 251-252. contention,we recognizethat thresholdsof this GREY,T., 8r C. LEA. 1969. Chemicaland organoleptic magnitude may exist for carrion-associated changes in poultry meat resulting from the odorants other than those used in this study. growthof psychrophyllicspoilate bacteria at IøC. Other compounds, such as putrescine, cadav- 6. Volatile aroma constituents.Brit. Poultry Sci. 10: 303-309. erine, several alcohols, fatty acids, and thiois, MOULTON, D.G. 1973. The use of animals in olfac- are also by-productsof carrion decomposition. tory research.Pp. 143-223 in Methodsof animal Any of these chemicals, either singly or in experimentation. IV. Environment and the spe- combination,may be importantas olfactory cues cial senses(W. I. Gay, Ed.). New York, Academic to Turkey Vultures. Press. Similarly, our approachto the Gaussiangas NEUHAUS,W. 1963. On the olfactorysense of birds. dispersionequation was far from exhaustive. Pp. 111-123 in Olfaction and taste, I (Y. Zotter- Modificationsof this equation are available for man, Ed.). Oxford, Pergamon Press. various substrate and atmospheric conditions OWRE,O. T., •: P.O. NORTHINGTON. 1961. Indication (Strom 1976)that, if appropriatelymodified for of the senseof smell in the Turkey Vulture, Ca- thresholddata, might increasethe flexibility of thartesaura (Linnaeus), from feeding tests.Amer. Midi. Natur. 66: 200-205. the model.Although numerousfield studiesby PANOFSKY,H., &: J. A. DUTTON. 1984. Atmospheric micrometeorologistshave confirmed the gen- turbulence.New York, John Wiley & Sons. eral utility of the Gaussian equation (see Pas- PASQUILL,F. 1974. Atmosphericdiffusion. New York, quill 1974, Panofsky and Dutton 1984) other JohnWiley & Sons. nonstatisticaltreatments (e.g. gradient-transfer PHELI'S,A. H. 1976. Odors.Pp. 307-339 in Air pol- theory) might better describethe concentration lution, vol. III (A. C. Stern, Ed.). New York, Ac- gradientsof carrion odorants. ademic Press. 592 SMITHAND PASELK [Auk, Vol. 103

SCHMIDT,U. 1975. Vergleichende Riechschwellen- WELLS,H. G. 1907. Chemical pathology. Philadel- bestimmungen bei neotropischen Chiropteran phia, W. B. Saunders. (Desmodusrotundus, Art•beus literatus, Phyllostomus WENTEL,B. M. 1965. Olfactory perception in birds. discolor).Z. Saeugetierkd.40: 269-296. Pp. 203-217 in Olfactionand taste,II (T. Hayashi, SHIBUYA,T., •a: D. TUCKER. 1967. Single unit re- Ed.). Oxford, PergamonPress. sponsesof olfactory receptorsin vultures. Pp. 1971. Olfaction in birds. Pp. 432-448 in 219-233 in Olfaction and taste, II (T. Hayashi, Handbook of sensory physiology. Vol. IV, Ed.). Oxford, PergamonPress. Chemical senses.Part I, Olfaction (L. M. Beidler, SNYDER,G. K., & T. T. PETERSON.1979. Olfactory Ed.). Heidelberg, Springer-Verlag. sensitivityin the Black-billedMagpie and in the 1973. Chemoreception. Pp. 389-415 in Avi- pigeon. Comp. Biochem.Physiol. 62A: 921-925. an biology, vol. III (D. S. Farher and J. R. King, STAGER, K. E. 1964. The role of olfaction in food Eds.). New York, Academic Press. location by the Turkey Vulture (Cathartesaura). --, & M. SIECK. 1972. Olfactory perception and Los Angeles Co. Mus. Contrib. Sci. No. 81. bulbar electricalactivity in severalavian species. STROM,G.H. 1976. Transportand diffusion of stack Physiol. Behav.9: 287-293. effluents.Pp. 401-503 in Air pollution, vol. I (A. ZEAMAN, D., •a: N. WEGNER. 1954. Cardiac reflex to C. Stern, Ed.). New York, Academic Press. tones of threshold intensity. J. SpeechHearing SUTTON,O.D. 1953. Micrometeorology.New York, Disord. 21: 71-75. McGraw-Hill. ZWA.ARDEMAKER, H. 1926. Odoriferous materials.Pp. TABER,W. B., JR. 1928. A theory of how the Turkey 358-361 in International critical tables of numer- Vulture finds its food. Wilson Bull. 40: 221-223. ical data,physics, chemistry and technology,vol. UNO, T., & W. GRINGS. 1969. Autonomic compo- I (E. W. Washburn, Ed.). New York, McGraw- nents of orienting behavior. Psychophysiology Hill. 1: 311-321.