HOMING OF MAGNETIZED AND DEMAGNETIZED PIGEONS

Transcription

1 J. exp. Biol. 14, (1988) 27 Printed in Great Britain The ompany of Biologit Limited 1988 HOMING OF MAGNTIZD AND DMAGNTIZD PIGONS BY HARLS WALOTT, JAMS L. GOULD 1 AND ANTHONY J. LDNOR Laboratoiy of Ornithology and Section of Seurobiology and Behavior, ornell Univerity, Ithaca, AT 148, USA and department of Biology, Princeton Univerity, Princeton, NJ 844, USA Accepted 1 July 1987 SUMMARY Homing pigeon appear to ue the earth' magnetic field a a compa and perhap a part of their poition-finding ytem or 'map'. The enory ytem they ue to detect magnetic field i unknown, but two current poibilitie are ome mode of repone by the pineal organ or by the viual ytem, or it may be baed on the magnetite crytal found in their head. Three erie of experiment to tet the involvement of magnetite are reported here. The alignment of the permanent magnetic domain in the bird head wa altered by (a) demagnetizing the bird, (b) magnetizing them with a trong magnetic field and (c) expoing the bird to a trong magnetic gradient. None of thee treatment had a marked effect on the pigeon' orientation or homing under unny kie, but a few reult obtained under overcat kie ugget that demagnetizing the bird may have increaed the catter of their vanihing bearing. Perhap pigeon ue one magnetic enor for their magnetic compa and another for ome component of the map. INTRODUTION A variety of evidence now indicate that homing pigeon may ue a magnetic compa to determine direction when celetial cue are unavailable. For example, magnet (Keeton, 1971, 1972) or paired coil (Walcott & Green, 1974; Vialberghi & Alleva, 1979) attached to the head or back of homing pigeon dirupt their ability to orient on overcat day. On unny day, neither magnet nor coil have much effect (Keeton, 1972; Walcott, 1977). Two experiment ugget that the magnetic compa of bird may be bipolar and cannot ditinguih north from outh except on the bai of the inclination of the field line. The north-eeking pole of a compa needle free to rotate in three dimenion will point north and down in the northern hemiphere, while the outh-eeking end will point outh and up. When Wiltchko & Wiltchko (1972) and Wiltchko (1972) revered the north-outh polarity of the earth' field (o that the outh-eeking pole of a compa pointed north and down), the orientation of Key word: pigeon, homing, magnetim, navigation.

2 28. WALOTT, J. L. GOULD AND A. J. LDNOR robin in a tet cage wa unaffected. When, however, they rotated the field o that the field line pointed down toward the eat rather than the north, then the robin were reoriented by an equivalent amount, regardle of the polarity of the field. Similarly, Walcott & Green (1974) and Vialberghi & Alleva (1979) found that when pigeon were fitted with paired coil which generated field whoe magnetic north pointed down, the bird were correctly oriented toward home under overcat kie. If the outh pole of the induced field pointed down, the bird flew away from home. Vector addition of the earth and coil field (Walcott & Green, 1974) indicate that in the firt cae the direction of the dip angle ought to have been to the north while in the econd cae it hould have been to the outh. In addition to the relatively clear effect of trong (roughly earth trength) magnetic field on the direction-finding abilitie of pigeon under overcat kie, an increaing body of correlation ugget that mall (-1-1 %) change in the earth' magnetic field trength affect initial orientation, homing peed and bird navigation in general (Keeton, Larkin & Windor, 1974; Yeagley, 191; Schreiber & Roi, 1976, 1978; Southern, 1971, 1972; Moore, 1977; Larkin & Sutherland, 1977; arr, Switzer & Hollander, 1982). For example, pigeon eem to home more lowly and more pigeon are lot on day when there have been a large number of un pot. Furthermore, magnetic torm ranging from -1 to % of the earth' total magnetic field trength alter the orientation of both ring-billed gull and homing pigeon. Keeton et al. (1974) repeatedly releaed a group of experienced homing pigeon at the ame ite and found that the average vanihing bearing of thee bird hifted counterclockwie with increaing magnetic variability. Pigeon releaed at magnetic anomalie (place where the earth' magnetic field trength varie irregularly) how an increae in the catter of their vanihing bearing that i correlated with the amount of variability in the magnetic field (Walcott, 1978, 1982). Kiepenheuer (1982, 1986) ha recently alo found that pigeon releaed at a trong and irregular anomaly in the Rhine Valley were dioriented, wherea control bird releaed outide the anomaly were well-oriented toward home. Wagner (1976), Frei & Wagner (1976) and Frei (1982), working with an anomaly in Switzerland, report that deflection of the pigeon' vanihing bearing from the direction of home were related to the magnetic topography around the releae ite. But thee Swi anomalie were quite weak (of the order of 2 y) and not particularly irregular. At trong anomalie ( y or about 6% of the earth field) pigeon eem totally dioriented, although homing peed and ucce are normal. Although there i no direct proof that the pigeon' diorientation i due to the ditorted magnetic field, uch a uggetion i upported by the correlation between the degree of magnetic variability at the anomaly and the amount of catter of the pigeon' vanihing bearing. Thee effect of magnetic anomalie are evident on clear day when the un compa i available and, from clock-hift data (e.g. Schmidt-Koenig, 1961), take precedence. Moreover, the mall magnetic fluctuation at anomalie and from magnetic torm hould have little effect upon a magnetic compa ytem. It i tempting to uppoe that torm and anomalie dirupt the pigeon' ability to ene

3 Pigeon homing 29 where they are relative to their home, and that the 'map' (Kramer, 19) or 'type III navigation' (Griffin, 19) depend in part upon the mall, ytematic change in magnetic field trength over the urface of the earth (Gould, 198, 1982, 198; Lednor, 1982; Moore, 198; Walcott, 198, 1982; for a review ee Preti, 198; Wallraff, 198; Wiltchko, Nohr, Fuller & Wiltchko, 1986). Whether the effect of magnetic torm occur during tranport to the releae ite, at the releae point, or both, i not clear. Larkin & Keeton (1976) found that bar magnet applied jut before releae on unny day mimicked the effect of a maximal magnetic torm. The effect of localized magnetic anomalie appear to reult from a ditortion of ite-pecific information. Walcott (1977) found that paired coil with an earth-trength field activated jut before releae deflected the pigeon' vanihing bearing even on unny day. At the ame time, Kiepenheuer (1978), Wiltchko, Wiltchko & Keeton (1978), Papi et al. (1978) and Benvenuti, Baldaccini & Ioale (1982) all found that altering the magnetic field on the trip to the releae point had a ignificant effect on the pigeon' vanihing bearing. Since pigeon appear able to ene not only the direction of the earth' field but alo minute change in field trength, it eem likely that they mut have a pecialized magnetic detector ytem. An intriguing candidate for thi ene organ i the magnetite reported by Walcott, Gould & Kirchvink (1979) and Preti & Pettigrew (198) in the head or neck of pigeon and by Beaon & Nichol (1984), Beaon & Brennan (1986) in bobolink. Thee magnetite crytal act like permanent magnet; about 4% of the pigeon examined howed a net magnetic remanence. In the other 6%, magnetic remanence could be induced by a trong magnetic field. In a ubequent tudy, over 8 pigeon were examined by Walcott & Walcott (1982). Although the ame technique a previouly were ued, no conitent intrinic or inducible remanence wa found. However, erially ectioned pigeon head, tained with an iron tain and examined under the light microcope, revealed three ite where intracellular poitively taining granule were abundant: in the Harderian gland, at the bae of the beak jut poterior to the cere, and in cell lying cloe to a bony ledge jut ventral to the olfactory nerve and olfactory lobe. While the hitological reult clearly demontrate the preence of intracellular iron-containing granule, the difference in reult obtained uing magnetometry are puzzling. It i poible, a Gould, Kirchvink & Deffeye (1978) and Gould (198) ugget for honeybee, that magnetic detection could be baed on uperparamagnetic magnetite crytal, which are found in large number in bee and are too mall to have a permanent remanence at phyiological temperature. Gould (198) and Kirchvink & Gould (1981) decribe how uperparamagnetic detector might work and calculate their enitivity. In uch a cae, permanent-domain magnetite might not be neceary. It i alo conceivable that pigeon might make ue of a magnetic field detection ytem aociated with the retina (Leak, 1977). Semm (198), Semm, Nohr, Demaine & Wiltchko (1984) and Semm & Demaine (1986) have recorded electrical repone in cell of the nucleu of the baal optic root and the optic tectum to earthtrength magnetic field timulation.

4 . WALOTT, J. L. GOULD AND A. J. LDNOR If pigeon head are not alway meaurably magnetic it might eem that permanentdomain magnetite cannot form the bai of a magnetic detector. alculation made by Kirchvink & Gould (1981), however, ugget that thi i not the cae. A few hundred or thouand permanent-domain magnetite grain would be ufficient for a enitive magnetic compa (though not for a magnetic map ytem) yet could not be detected by the SQUID magnetometer ued to meaure the magnetization of pigeon head. The enitivity of thee magnetometer i about 1 e.m.u. and the moment of a relatively mall number of permanent-domain magnetite grain would be well below thi (1~ 9 e.m.u. for 1 ingle domain). The behavioural, magnetic and hitological evidence ugget to u four alternative hypothee. (1) The total propertie of a permanent-domain magnetite array are important in orientation. (2) The collective magnetic arrangement of permanentdomain magnetite particle i not important; rather it i the individual particle themelve that are reponible for the detection of magnetic information. () Magnetic enitivity i baed on uperparamagnetic domain. (4) The magnetite grain have nothing to do with the pigeon' enitivity to magnetic field. In an effort to tet the firt hypothei that the magnetic propertie of the total array of magnetite grain might be important we demagnetized homing pigeon in a trong alternating magnetic field. Thi proce revere the polarity of half the grain, thereby diabling a detector which depend on the total alignment of the grain. We teted the effect of thi procedure by releaing the pigeon at normal ite on both unny and overcat day and by releaing them alo at magnetically anomalou ite on unny day. We alo tried to align the permanent-domain magnetic grain in a trong magnetic field. Finally, we examined the effect on the pigeon' orientation of expoing the bird to a trong magnetic gradient that exerted a tranlational force on the permanent-domain magnetite grain. MATRIALS AND MTHODS xperienced, adult pigeon from our loft in Lincoln, MA, were trained a a flock along a line SS of the loft to a ditance of mile (8 km). They were then each given a erie of ingle toe along thi line until they were returning promptly from 8 km. One ingle to from 48 to 8km Wet of the loft completed their training. The relative alignment of the moment of the permanent-domain magnetite grain wa altered by putting the bird in the centre of a - 21 T 6 Hz a.c. oil decribed by Gould, Kirchvink, Deffeye & Brine (198) and fabricated by the Priceton Plama Phyic Laboratory oil Shop. Thi coil wa itelf located inide a 1-coil ytem (Mclhinny, Luck & dward, 1971) which wa ued to cancel the earth' field to within about - 2% of normal. From other data (Walcott et al. 1979; Kirchvink & Gould, 1981) we know that the field adequate to revere the polarity of the pigeon' permanent magnetic domain i about -7 T applied along the long axi of each grain. ach pigeon wa held in the centre of the coil a a 1-A current wa applied and then gradually reduced to zero.

5 Pigeon homing 1 Thi procedure wa followed three time with each pigeon o that the head wa demagnetized once in each of three mutually orthogonal orientation. A a reult every magnetite domain, regardle of it alignment in the pigeon, felt a minimum demagnetization field of -1 T. (-1 T i the maximum theoretical unblocking field required to revere a permanent magnetite domain.) The trong alternating field erved to flip the magnetic moment back and forth until the lowly declining current reduced the field below the trength jut neceary to flip the moment. At thi point, the lat polarity trong enough to effect reveral i retained indefinitely. xactly when a particular grain drop out i a function of it ize, hape and orientation, o that it ultimate polarity i determined by chance. Hence for elongate grain, half the moment will be revered. We call thi procedure 'degauing'. The null field around the a.c. olenoid wa maintained in order to avoid a - X 1~ 4 T bia on the moment of the grain which would tend to align them with the earth' field. Gould et al. (198) have demontrated that thi arrangement work on crytal roughly twice a magnetically 'hard' a thoe of pigeon. In an effort to align all the permanent magnetic moment, at leat roughly, we placed each pigeon' head between the pole piece of a gap magnet (dmund Scientific no. 71 1). The gap wa et at 2-4 or 2-8cm correponding to field of - 14 and -11 T, repectively. Thi field hould be ufficient to align roughly 68 or 64% of the permanent domain. A a reult, roughly one-third of the moment hould be flipped a compared to one-half in the degauing procedure. For mot experiment the pigeon wa oriented o that the field line paed tranverely through it head in a horizontal plane - that i, o that it eye faced the pole piece of the magnet. The bird' head wa lowly put between the pole piece, held there momentarily, and then lowly withdrawn. The 'Gauing' wa performed at the releae ite immediately before the pigeon wa releaed. To expoe pigeon to a trong magnetic gradient, they were taken to the Franci Bitter National Magnet Laboratory at the Maachuett Intitute of Technology in ambridge, MA. Here they were placed lightly off centre in a 12T olenoid. Thi arrangement expoed the pigeon' head and neck to a maximum gradient of about 8Tcm~'. With the pigeon firmly fixed in place, the current in the olenoid wa gradually and moothly increaed to it maximum value over a period of 4. It wa then gradually decreaed to zero. ontrol bird were placed in the magnet for an equal time, with the cooling water running, but without any current. In a econd erie, the ame procedure and ame peak current wa ued, but the current wa ramped moothly up to maximum over 2min, held there for 1 min and then decreaed to zero over 2min. According to calculation in Kirchvink & Gould (1981), the firt procedure would move a 1 A domain approximately 1 mm in a olution 1 time a vicou a water, while the econd would move uch a domain -8 mm. Pigeon were tranported from the loft to the releae point either in the back of a tation wagon or in the front compartment of a Volkwagen etate. The releae point ued and the direction and ditance to home were: Worceter Airport, 86, km; arver, 1, 71km; Hancock, 12, 84km; Lynnfield, 248, 2km; Providence,

6 2. WALOTT, J. L. GOULD AND A. J. LDNOR 27, 61 km. For ome experiment the gap magnet wa carried in a mu-metal box which erved to iolate the bird from the field of the magnet. At the releae point each pigeon wa equipped with a mall radio beacon (ochran, 1967) and tracked for 1min or until it vanihed from radio range. The bearing of bird whoe ignal vanihed in le than min were excluded from analyi. The direction to the pigeon at 1 min or when it vanihed wa recorded and i given in the diagram. Mean bearing were computed with the Rayleigh tet and teted for non-uniformity with the V'-tet. Ditribution of bearing were compared with the Waton U tet (Batchelet, 1981) and difference in the amount of catter were aeed with a nonparametric tet propoed by Wallraff (1979) and decribed in Batchelet (1981). Vanihing interval and homing peed were compared with the Mann-Whitney [/-tet (Zar, 1974). RSULTS If the magnetic enitivity of homing pigeon were dependent on an array of permanent magnetite domain of ordered polarity, then degauing a pigeon might caue it to become dioriented. If degauing caued an effect on unny day when pigeon are preumably relying on their un compae, thi would ugget that an ordered polarity detector might be involved in the map ytem. If, however, degauing only had an effect on overcat day when the un wa not viible, it would ugget that thi ame ort of detector wa involved in the magnetic compa ytem. To tet thee idea, pigeon were taken from their loft and degaued. Along with control bird they were taken to a magnetically normal releae ite, releaed and tracked. A Fig. 1A,B how, bird releaed at two ite under unny condition howed no ignificant deficit in their ability to orient toward home. Both control and experimental are well-oriented with the cutomary releae point biae in each cae. The experimental bird have omewhat greater catter than the control at Worceter, and omewhat le at arver. In neither cae are the difference ignificant. In both cae the degaued bird departed on an average bearing clockwie of the control. There wa no ignificant difference in time to vanih from the releae point, homing peed or the number of bird that returned. Data for thi and all other tet reported here are ummarized in Table 1. To ee whether degauing might be more effective if it were carried out at the releae ite, we moved the coil to Hancock, NH and conducted a erie of tet there. Half the bird were degaued immediately before releae under unny kie. A Fig. 1 indicate, thi made no difference to the reult; the performance of both control and experimental bird wa eentially identical. Apparently degauing ha little effect on either vanihing bearing of homing performance on unny day. The effect of releaing degaued pigeon under overcat kie, when they are probably relying on magnetic compa information, are hown in Fig. ID. The vanihing bearing of degaued bird were more cattered than thoe of the control and the mean vanihing bearing wa again rotated clockwie. Unfortunately, ince

7 Pigeon homing the number of day with olid overcat kie during the ummer wa mall, too few bird were releaed to make thee difference tatitically ignificant. The bet that can be aid i that the control bird, although random under the Rayleigh tet (P = -9) were ignificantly oriented by the V-tet (P = < -). The orientation of the degaued bird wa random by both tet. Time to vanih from the releae point, homing peed and the number of bird that returned were the ame for both group. Thu, if one can believe the reult from the few bird releaed under B O O o Fig. 1. Bearing of control pigeon and of pigeon degaued at the loft and releaed at arver, MA (A) or Worceter, Airport, MA (B). In bird were taken to Hancock, NH and degaued there. All bird were releaed under unny kie, except the bird hown in D, which were releaed at arver, MA under totally overcat kie. In thi and all the following diagram, each dot on the periphery of the circle repreent the radio bearing of a ingle pigeon. The open circle are the bearing of control pigeon, the olid dot, experimental pigeon. The dotted line inide each circle indicate home direction, each diagram being plotted with magnetic North toward the top of the page. The arrow originating at the centre of each diagram repreent the mean vector calculated by the Rayleigh tet. It i drawn to cale, with the radiu of the circle et to a vector length of 1. An arrowhead indicate the tet wa ignificant (P<) under the Rayleigh tet. An open arrowhead or dahed line indicate control bird, a olid arrowhead or olid line, experimental bird. Further detail of text reult are given in Table 1.

8 Table J. Summary of dala on initial orientation and homing peifonnaiice Homing No. No. on dav late ucce No. lot O <! r o r oo r O D > r rrl a O Site Worceter Airport arver I lancoek arver Lvnntield anomalv 1 lancoek Worceter Airport Hancock 1 lancoek Providence Providence Treatment DeGau at loft DeGau at loft DeGau at ite DeGau at loft DeGau at loft Gau Gau Gau (magnet tranported eparately) Gau (vertical) Gradient tet 198 Gradient tet 1981 Sun O/ S on MVB(.V) 82(28) 99 (26) 18(22) 24 (19) 1 (16) 161 (17) 27 (1) 2 (1) 128 () 72(22) 147 (16) 147 (16) 111 (22) 12 (21) 162(16) 149 (18) 16 (1) 166L (14) 11 (24) 19 (27) 2 (19) 16 (17) r P MVI 1- (12) 9-7 (1) 9- (7) 9-8 (9) 11-9(8) 11-(4) 12-6() 9- (2) 1- (1) 8-1(9) 12-4 (7) 8-9 (7) 7-2 (1) 7-(1) 9- (7) 11-1 (9) 11- (9) 11-(6) 1- (19) 14-6(14) 16- (1) 141 (14) A' MHS 11-7 (18) 11-2(22) 1-1 (17) 16-(11) 9-7 () 1-6() 12-1 (6) 1-4 (7) 6- (22) 7-2 (18) 14-6() 16-9 (9) 12-4(1) 1-9 (12) 16-9 (8) 21- (1) 16-4 (8) 1-7 (4) 1-6 (18) 16-4 (1) 1-7 (16) 12-9 (1) Sun ()/, tet were conducted under un (S) or overcat () kie. on,, control treatment;, experimental treatment. ( g g g p r, the mean vector length. l\ the probability that the ditribution of bearing i unifonn under the Rayleigh tet. MVI, the mean vanihing interval of bird vanihing in le than 1 mm; the ample ize i hown in parenthee. A'1, the number of bird remaining in radio range 1 min after releae. M1IS, the mean homing peed (in mile per hour) of bird returning on day 1, the day of the tet; the ample ize i hown in parenthee. Homing ucce, No. on day i the number of bird that returned on the day of releae; No. late i the number that returned after the day ol releae; and No. lot i the number of bird that never returned.

9 Pigeon homing ooo* Fig. 2. Initial orientation of pigeon degaued at the loft and control pigeon releaed at the magnetic anomaly at Lynnfield, MA. For convention ee Fig. 1. overcat kie, degauing eem to have had a greater effect than it did when the un wa viible. That degauing had no effect on unny day doe not necearily mean that it did not affect a magnetic map. It could be that the pigeon imply witched to ome backup ytem; that the map a well a the compa i redundant. To tet thi idea we releaed degaued pigeon at a magnetic anomaly. If degauing had detroyed their ability to detect the irregular field of the anomaly and they were uing a nonmagnetic back-up ytem, we would predict that the bird might be well-oriented. A Fig. 2 how, however, degaued pigeon releaed at the magnetic anomaly at Lynnfield, MA were jut about a dioriented a the control. It appear that even if degauing dirupt a magnetic compa ytem, it ha little or no effect upon the map. Of the pigeon examined by Walcott et al. (1979) in a magnetometer, only about 4% had a natural magnetic remanence. It eemed poible that increaing the alignment of the permanent magnetite grain by Gauing the pigeon would have more of an effect than degauing. We tranported bird to the releae point in a carrying cage next to which there wa a gap magnet encloed in a mu-metal box. One group of pigeon had their head placed between the pole piece of the magnet with their eye facing the pole. The control were imply releaed and tracked. Under unny kie at Worceter, MA and Hancock, NH, the experimental bird were, if anything, better oriented than the control (Fig. A,B). The difference were not ignificant, and there wa no ignificant difference in any meaure of homing performance. Since the gap magnet in the mu-metal box wa carried next to the pigeon on their outward journey, it eemed poible that the magnetic field around the pigeon might have been ditorted by thi procedure. To tet thi we carried the pigeon and the magnet in eparate vehicle. A Fig.,D how, there wa eentially no difference

10 6. WALOTT, J. L. GOULD AND A. J. LDNOR between Gaued and control pigeon. We alo tried another head orientation during Gauing uch that the magnetic field line paed vertically through the pigeon' head. A Fig. B,D illutrate, there wa no difference between bird Gaued horizontally and thoe Gaued vertically. It appear that Gauing ha remarkably little effect upon bird releaed under unny condition. The evidence o far ugget that neither Gauing nor degauing ha much effect upon the pigeon' orientation under unny condition. But if pigeon are uing particle of permanent-domain magnetite a part of a enory ytem to detect the earth' magnetic field, applying a ufficiently trong magnetic gradient might caue the particle to hift and interfere with their functioning. We applied a magnetic field gradient of 8 Tern" 1 to the pigeon' head and then releaed and tracked the bird Fig.. Orientation of bird Gaued immediately prior to releae under unny kie. (A) Reult at Worceter, MA; (B) reult of Gaued bird at Hancock, NH. At both ite both group of bird were carried cloe to a mu-metal box containing the gap magnet. xperimental bird were Gaued horizontally at both ite. () ontrol bird tranported to the ite in a eparate vehicle from that carrying the gap magnet, and then Gaued at Hancock; (D) bird Gaued vertically and releaed at Hancock. onvention a in Fig. 1.

11 Pigeon homing 7 Fig. 4. Reult of expoing pigeon to a powerful (-8 T cm ') gradient. (A) Tet done in 198; (B) reult obtained in onvention a in Fig. 1. under unny condition. A Fig. 4A how, the experimental bird were le accurately oriented than the control, but the difference in catter i not ignificant. Vanihing time wa greater and homing peed wa lower for the experimental bird but the difference were not ignificant (Mann-Whitney (7-tet, P>-). In the econd erie (Fig. 4B), the oppoite reult were obtained; the control bird were more cattered than the experimental group. The two group did not differ ignificantly in their initial orientation, time to vanih or homing peed. Unfortunately, no experiment were done under overcat kie. DISUSSION The reult of the Gauing and degauing experiment clearly rule out the poibility that, under unny condition, pigeon rely on a magnetic detector in which the polarity of permanent-domain magnetite crytal i critical. Any receptor baed on an orderly alignment of ingle permanent domain hould be dirupted if the moment of half the domain are revered by degauing. Under overcat kie there i a hint that the vanihing bearing of degaued pigeon were more cattered than thoe of control. It i unfortunate that there are o few data under overcat kie, becaue if pigeon were uing an ordered magnetite array for their compa one might expect little or no effect of degauing under unny kie, when old and experienced pigeon would be uing their un compa, but would only ee the effect under overcat kie when they were uing a magnetic compa. Kiepenheuer, Ranvaud & Maret (1986) report that expoing pigeon to a homogeneou magnetic field of 1 T (1 G) for 1 min caued both a ignificant bia in the direction a well a an increaed catter in their vanihing bearing. Thi effect perited for at leat 6 week after expoure. The field to which our pigeon were expoed wa about the ame trength and the length of the expoure wa about

12 8. WALOTT, J. L. GOULD AND A. J. LDNOR the ame, but we arranged for the pigeon to experience a ubtantial magnetic gradient. A ingle magnet in the uniform field will attempt to align itelf with the applied field but will not be ubject to a tranlational force. In a gradient, however, the magnet will not only tend to align itelf with the field but alo there will be a net force on the magnet. The familiar example i the effect of one bar magnet on another a they are brought together; if they are free to rotate, they align with one another o that they attract. It i urpriing that our treatment had o little effect; it wa altogether more dratic than Kiepenheuer'. It i poible that Kiepenheuer et al. (1986) were uing young pigeon who might be relying more on the magnetic compa than the old experienced pigeon we were uing. It i alo poible that the gradient o dirupted the receptor that the pigeon failed to ue it. Although it eem improbable that any permanent-domain magnetite-baed receptor could urvive the diruptive force of the gradient, a uperparamagnetic ytem would be far le affected. If the apparent enitivity of pigeon to earth-trength magnetic field (uitable for compa information) and to mall-magnetic field change (ued in ome way in the map) i baed on magnetite, then unle the pigeon' map i fully redundant with repect to magnetic cue, we can draw two main concluion from thee experiment. Firt, any magnetic map ytem, if it depend on magnetite at all, mut depend on uperparamagnetic crytal ince (a) many pigeon lack detectable permanentdomain magnetite and (b) Gauing and degauing had no effect on unny day. Second, the magnetic compa ytem could depend on permanent-domain magnetite ince (a) o little i needed that it could fall below the threhold of the detector and (b) the mall amount of data obtained under overcat kie indicated that degaued pigeon were more affected under overcat than under unny kie. Finally, the reult we obtained in thee experiment are alo conitent with the idea that pigeon can detect a magnetic field in ome part of their viual ytem. If that were the cae, none of the treatment we gave the pigeon would be expected to have any effect. We are grateful to Richard B. Frankel and hi colleague for their hopitality at the Franci Bitter National Magnet Laboratory at the Maachuett Intitute of Technology. We alo thank all the tudent for their help with the experiment and the NSF grant no. BNS for financial upport. RFRNS BATSHLT,. (1981). ircular Statitic in Biology. London, New York, Toronto: Academic Pre. BASON, R.. (1986). Magnetic orientation and magnetically enitive in migratory bird. In Biophyical ffect of Steady Magnetic Field (ed. G. Maret, N. Broccara & J. Kiepenheuer), pp Berlin, New York: Springer-Verlag. BASON, R.. & BRNNAN, \V. J. (1986). Natural and induced magnetization in the bobolink, Dolichoiiyx oryzivoru (Ave: Icteridea). J. exp. Biol. 12, BASON, R.. & NIHOLS, J.. (1984). Magnetic orientation and magnetically enitive material in the Bobolink (Dolichonvx oryzivoru), a tranequatorial migrant. Xature, Loud. 9, 11-1.

13 Pigeon homing 9 BNVNUTI, S., BALDAINI, N.. & IOAL, P. (1982). Pigeon homing: effect of altered magnetic field during diplacement on initial orientation. In Avian Savigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. ARR, P. H., SWITZR, \V. P. & HOLLANDR, \Y. F. (1982). vidence for interference with navigation of homing pigeon by a magnetic torm. Iotva State J. Re. 6, OHRAN, W. (1967) mHz beacon tag tranmitter for mall animal. BIA Information Module M-1. AIBS Wahington, D. FRI, U. (1982). Homing pigeon' behaviour in the irregular magnetic field of Wetern Switzerland. In Avian Savigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. FRI, U. & WAGNR, G. (1976). Die Anfangorientierung von Brieftauben im erdmagnetich getorten Gebiet de Mont Jorat. Rev. uie Zool. 8, GOULD, J. L. (198). The cae for magnetic enitivity in bird and bee (uch a it i). Am. Scient. 68, GOULD, J. L. (1982). The map ene of pigeon. Xalure, land. 296, GOULD, J. L. (198). Are animal map magnetic? In Magnetite Biomineralization and Magnetoreception in Organim (ed. J. L. Kirchvink, D. S. Jone & B. J. MacFadden), pp New York, London: Plenum Pre. GOULD, J. L., KIRSHVINK, J. L. & DFFYS, K. S. (1978). Bee have magnetic remanence. Science 22, GOULD, J. L., KIRSHVINK, J. L., DFFYS, K. S. & BRINS, M. L. (198). Orientation of demagnetized bee. J. exp. Biol. 86, 1-8. GRIFFIN, D. R. (19). Bird Navigation. In Recent Studie in Avian Biology (ed. A. Wolfon), pp Urbana: Univerity of Illinoi Pre. KTON, W. T. (1971). Magnet interfere with pigeon homing. Proc. natn. Acad. Set. LJ.SA. 68, KTON, W. T. (1972). ffect of magnet on pigeon homing. In Animal Orientation and Navigation (ed. S. R. Galler, K. Schmidt-Koenig, G. J. Jacob & R.. Belleville), pp NASA SP-262. KTON, W. T., LARKIN, T. S. & WINDSOR, D. M. (1974). Normal fluctuation in the earth' magnetic field influence pigeon orientation, jf. comp. Phyiol. 9, 9-1. KIPNHUR, J. (1978). Inverion of the magnetic field during tranport: It influence on the homing behavior of pigeon. In Animal Migration, Navigation and Homing (ed. K. Schmidt- Koenig & W. T. Keeton), pp Heidelberg: Springer-Verlag. KIPNHUR, J. (1982). The effect of magnetic anomalie on the homing behavior of pigeon: an attempt to analyze the poible factor involved. In Avian Savigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. KIPNHUR, J. (1986). A further analyi of the orientation behavior of pigeon releaed within magnetic anomalie. In Biophyical ffect of Steady Magnetic Field (ed. G. Maret, N. Broccara & J. Kiepenheuer), pp Berlin, New York: Springer-Verlag. KIPNHUR, J., RANVAUD, R. & MART, G. (1986). The effect of ultrahigh magnetic field on the initial orientation of homing pigeon. In Biophyical ffect of Steady Magnetic Field (ed. G. Maret, N. Broccara & J. Kiepenheuer), pp Berlin, New York: Springer-Verlag. KIRSHVINK, J. L. & GOULD, J. L. (1981). Biogenic magnetite a a bai for magnetic field detection in animal. Bioytem 1, KRAMR, G. (19). Die Sonnenorientierung der Vogel. Verb. dt. Zool. Ge. 192, LARKIN, R. P. & SUTHRLAND, P. J. (1977). Migrating bird react to Project Seafarer' electromagnetic field. Science 19, LARKIN, T. S. & KTON, W. T. (1976). Bar magnet mak the effect of normal magnetic diturbance on pigeon orientation..7. comp. Phyiol. 11, LASK, M. J. M. (1977). A phyiochemical mechanim for magnetic field detection by migrating bird and homing pigeon. Sature, Land. 267, LDNOR, A. J. (1982). Magnetic navigation in pigeon: poibilitie and problem. In Avian Savigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. MLHINNY, M. W., LUK, G. R. & DWARDS, D. (1971). A large volume magnetic free pace. Pureappl. Geophy. 9, MOOR, B. R. (198). I the homing pigeon' map geomagnetic? Sature, Land. 28, 69-7.

14 4. WALOTT, J. L. GOULD AND A. J. LDNOR MOOR, F. R. (1977). Geomagnetic diturbance and the orientation of nocturnally migrating bird. Science 196, PAPI, F., IOAL, P., FIASHI, V., BNVNIFTI, S. & BALDAINI, N.. (1978). Pigeon homing: ue detected during the outward journey influence initial orientation. In Animal Migration, Xavigation and Homing (ed. K. Schmidt-Koenig & W. T. Keeton), pp Heidelberg: Springer-Verlag. PRSTI, D.. (198). Avian navigation, geomagnetic field enitivity and biogenic magnetite. In Magnetite Biomineralization and Magnetoreception in Organim (ed. J. L. Kirchvink, D. S. Jone & B. J. MacFadden), pp New York, London: Plenum Pre. PRSTI, D. & PTTIGRW, J. D. (198). Ferromagnetic coupling to mucle receptor a a bai for geomagnetic field enitivity in animal. Mature, Land. 28, SHMIDT-KONIG, K. (1961). Die Sonne al Kompa im Heim-Orientierungytem der Brieftauben. Z. Tierpychol. 18, SHRIBR, B. & ROSSI, O. (1976). orrelation between race arrival of homing pigeon and olar activity. Boll. Zool. 4, SHRIBR, B. & Roi, O. (1978). orrelation between magnetic torm due to olar pot and pigeon homing performance. I Tran. Magn. 14, SMM, P. (198). Neurobiological invetigation on the magnetic enitivity of the pineal gland in rodent and pigeon. omp. Biochem. Phyio!. 76A, SMM, P. & DMAIN,. (1986). Neurophyiological propertie of magnetic cell in the pigeon' viual ytem. J. comp. Phyiol. A 19, SMM, P., NOHR, D., DMAIN,. & WILTSHKO, W. (1984). Neural bai of the magnetic compa: interaction of viual, magnetic and vetibular input to the pigeon brain. J. comp. Phyiol. A 1, SOUTHRN, W.. (1971). Gull orientation by magnetic cue: a hypothei reviited. Ann. X. Y. Acad.Sci. 188, SOUTHRN, W.. (1972). Influence of diturbance in the earth' magnetic field on ring-billed bull orientation. ondor 74, VISALBRGHI,. & ALLVA,. (1979). Magnetic influence on pigeon homing. Biol. Bull. mar. biol. Lab., Wood Hole 16, WAGNR, G. (1976). Da Orientierungverhalten von Brieftauben im erdmagnetich getorten Gebiete de haeral. Rev. uie Zool. 8, WALOTT,. (1977). Magnetic field and the orientation of homing pigeon under un. J. exp. Biol. 7, WALOTT,. (1978). Anomalie in the earth' magnetic field increae the catter of pigeon' vanihing bearing. In Animal Migration, Xavigation and Homing (ed. K. Schmidt-Koenig & W. T. Keeton), pp Berlin, Heidelberg, New York: Springer-Verlag. WALOTT,. (198). ffect of magnetic field on pigeon orientation. In Ada XVII ongreu Internationali Omithologici (ed. R. Nohring), pp Berlin: Deutche Ornithologen- Geellchaft. WALOTT,. (1982). I there evidence for a magnetic map in pigeon? In Avian Xavigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. WALOTT,, GOULD, J. L. & KIRSHVINK, J. L. (1979). Pigeon have magnet. Science 2, WALOTT,. & GRN, R. (1974). Orientation of homing pigeon altered by a change in the direction of an applied magnetic field. Science 184, WALOTT,. & WALOTT, B. (1982). A earch for magnetic field receptor in animal. In Avian Xavigation (ed. F. Papi & H. G. Wallraff), pp Berlin: Springer-Verlag. WALLRAFF, H. G. (1979). Goal-oriented and compa-oriented movement of diplaced homing pigeon after confinement in differentially hielded aviarie. Behav. col. Sociobiol., WALLRAFF, H. G. (198). Relevance of atmopheric odor and geomagnetic field to pigeon navigation: What i the "map" bai? omp. Biochem. Phyiol. 76A, WILTSHKO, R., WILTSHKO, W. & KTON, W. T. (1978). ffect of outward journey in an altered magnetic field on the orientation of young pigeon. In Animal Migration, Xavigation and Homing (ed. K. Schmidt-Koenig & W. T. Keeton), pp Heidelberg: Springer-Verlag.

15 Pigeon homing 41 WlLTSHKO, \V. (1972). The influence of magnetic total intenity and inclination on direction preferred by migrating robin {rithacu rubecula). In Animal Orientation and Xavigation (ed. S. R. Galler, K. Schmidt-Koenig, G. J. Jacob, & R.. Belleville), pp NASA SP-262. WlLTSHKO, W., NOHR, D., FULLR,. & WlLTSHKO, R. (1986).'Pigeon homing: The ue of magnetic information in poition finding. In Biophyical ffect of Steady Magnetic Field (ed. G. Maret, N. Broccara & J. Kiepenheuer), pp Berlin, New York: Springer-Verlag. WlLTSHKO, W. & WlLTSHKO, R. (1972). Magnetic compa of european robin. Science 176, YAGLY, H. L. (191). A preliminary tudy of the phyical bai of bird navigation II. J. appl. Phyic 22, ZAR, J. H. (1974). Biotatitical Analyi. nglewood liff, NJ: Prentice-Hall, Inc.

16