Proceedings ofthe United StatesNational MuseumSMITHSONIAN INSTITUTION ? WASHINGTON, D.C.Volume 122 1967 Number 3595
Ecology and Social OrganizationIn the Iguanid LizardAnolis lineatopus
By A. Stanley Rand *
IntroductionThis paper reports the findings of a fO-montli (August 1961 toJune 1962) field study of the ecology and behavior, particularly thesocial behavior, of Anolis lineatopus, in edificarian situations, inthe vicinity of Kingston, Jamaica.The geographical variation of this common Jamaican lizard hasbeen discussed by Underwood and Williams (1959) and by Grant(1940) but, beyond brief notes in these papers and in Barbour (1910),its natural history has never been described.There have been a number of detailed studies on free-living lizards,but most of them have dealt with temperate zone species in tem-perate environments. Few investigators have had the opportunityto conduct intensive and extensive studies on lizards in the tropics.The few exceptions include Evans (1951), Harris (1964), and Hirth
, (1963 a and b). No extended field study with an emphasis similar
1 to this one has been published on any tropical Anolis though theshorter papers of Evans (1938a) and Oliver (1948) report relevant
? Zoologist, Smithsonian Tropical Research Institute, Box 2072, Balboa, Pan-ama Canal Zone.
2 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122information for A. sagrei. Evans (1961) gives a quite completebibliography of lizard natural history.Throughout this study I have attempted to verify critical pointswith counts, censuses, or other objective measurements, and I haveused these extensively here. They have the disadvantage of ex-cluding all information which is not being measured and so present alimited picture and one which is sometimes difficult to comprehendby someone who is not famUiar with anoline behavior. To offsetthis and provide a frame of reference, I have included frequentabstracts from my field notes [direct quotes and paraphrasing bothare in smaller type?ed.] and have appended two short sections toprovide a picture of the general behavior of this species, one report-ing the behavior of an individual which was under observation foran entire day, and the other giving a composite and hypotheticalaccount of the usual course of the life of an A. liiicatojnis.I want to thank first Dr. Ernest E. Williams and my wife Patricia,the two people who contributed most to this study. Dr. Williams,principal investigator of the National Science Foundation grantunder which the field work was done, helped organize the project,advised me during our time in the field, and criticized the manu-script at various stages of completion. My wife helped with thefield work, did most of the editing and typing of the manuscript,and provided both moral and logistic support.A number of other people contributed importantly to the work.My father. Dr. A. L. Rand, supplied very valuable criticism of boththe ideas and form of the work. Prof. Garth Underwood discussedthe problem with me and gave me the benefit of his years of ex-perience A\dth the Jamaican anoles. Dr. P. E. Vanzolini advised onthe analysis of the data and criticized the manuscript. Discussionwith Dr. W. J. Smith provided additional insight into certain aspectsof the behavioral concepts. Drs. M. Moynihan, E. Willis, J. Eisen-berg, and W. Milstead read and criticized the manuscript.I am indebted to Prof. D. Steven, who allowed me to use the facilitiesof the Zoology Department of the University of the West Indies, andto the rest of the staff there, particularly Prof, Ivan Goodbody andMr. William Page for their assistance. I wish to thank them and thevisitors at the University, such as Mr. Malcolm Edmunds and Dr.Arthur Hughes, for listening to lengthy discom'ses on anoles and forcriticizing them most helpfully.I am also indebted to Mr. William Page and Mrs. L. Jones for theirhelp in determining the insects found in lizard stomachs.Finally, I must acknowledge the financial support for this projectfrom National Science Foundation grant number 16066. J
NO. 3595 ANOLIS LINEATOPUS?RAND 3Material and MethodsAnolis lineatopus shows considerable geographic variation \\4thinJamaica. Those from the vicinity of Kingston are medium-sizedAnolis, the adult males about 50 to 70 mm in snout-vent length andreaching about 8 to 9K grams in weight. The adidt females aresmaller, about 37 to 47 mm in snout-vent length and about 1% to 3grams in M'eight. Both sexes are brown to gray brown in color. Theadult males are marked with irregular black bars. The females aremore variable and may be almost unicolor, or they may have adiamond- or rhomb-shaped pattern on the back or a light middorsalstripe. There is only limited color change, from light to dark. Thefeet have the usual A7iolis form with moderately wide toe pads andan intermediate number of lamellae (29-34 under second and thirdphalanges on fourth toe). The dewlap or throat fan, large and lightyellow with an orange central spot, is well developed only in the male.These are called A. lineatopus lineatopus by Underwood and Williams(1959).Anolis are strictly diurnal and depend almost entirely on vision intheir social behavior and in locating food. They may respond tosounds but seem to pay no attention to odors.Jamaica, at 18? north latitude, lies well within the tropics and thereis relatively little variation in temperature in the two wet and twodry seasons. The Anolis are active year round and neither fighting,courting, nor copidation seems seasonal. There does seem to be someseasonal variation in the production of young: more in the wet season.Whether this is due to variation in egg laying or in egg survival, I donot know.In the dry lowlands on the south coast of Jamaica, A. I. lineatopusis widespread and abundant but much more abundant in some placesthan others. This is particularly striking when one is trying to collectsamples at predetermined intervals along a transect. In some casesthe differences in abundance are correlated with ecological differences,but in others I could see no difference in habitat.The areas where this study was made were all much disturbed byman. Those on which I concentrated were higtily artificial gardens,the campus of the University of the West Indies, and roadside hedgesand fence rows. Relatively little time was spent in natural habitatsand no detailed studies were made therein. Little undisturbedhabitat is left in Jamaica and what remains is mangrove swamp,scrub on dogtooth limestone and mountain forest, and all difficult ofaccess. Since the edificarian areas were easier to work in, closer athand and, most important, usually supported denser populations oflizards, I concentrated on these areas.
4 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Though I made repeated trips to the other parts of the island, theintensive observations reported here were made in two suburbs ofKingston, Barbican and Mona.In studying social behavior, I concentrated on A. lineatopus livingin small areas, marking them individually, mapping their positions,and describing their behavior. They were marked by clipping theterminal phalanx from one front and one hind toe, according to a code(removing one additional toe allowed higher numbers). The clippedtoes did not regenerate but were not visible from any distance. Anumber was painted on the back of each lizard with white paint(a quick-drying amylacetate base airplane paint). This was visibleat a distance but each A. lineatopus had to be recaptured and re-painted after each shedding. Though inconvenient, this gave us anopportunity to remeasure these lizards at intervals. The clipping ofthe terminal phalanx removed the claw but not the pad and did notseem to affect the lizard's climbing or running ability. The whitepainted numbers interfered slightly with shedding and certainly madethe lizard more conspicuous to me and presumably to other predators.The interference with shedding however, was temporary and theincreased conspicuousness unavoidable.The lizards were measured from the tip of the snout to the anteriormargin of the vent by holding them along a ruler. The measure-ments were recorded to the nearest millimeter, and repeated measure-ments on the same individual during a short period frequently variedby one millimeter but seldom by more. Throughout the paper theanole sizes given are snout-vent lengths in millimeters.Each lizard was marked in the field as soon as it was caught andreleased in less than five minutes in the area where it had been cap-tured. Two methods were used in catching the lizards. Somewere noosed during the day with a running noose of nylon filamenton the end of a slender stick. Others were caught by hand at nightwhile they were sleeping. The latter technique is very effective andis the one Cagle (1948) describes as being used by professional anolecollectors in Louisiana.The marking process disturbed the lizards, and, after being re-leased, a lizard usually remained quiet for an hour or so before re-suming its normal activities.Though the lizards learned to avoid the noose, they soon came toignore me completely, particularly if I were stitting quietly, even inplain view and only a few feet away. Those in the areas where Ispent hours watching them became very tame and would approachme without hesitation to catch an insect. One female even tookan insect from my fingers.
NO. 3596 ANOLIS LINEATOPUS?RAND 5Once the lizards had been marked and the study area mapped,observations were usually made daily. Many Lizards were not re-corded every day, but some were and most were seen repeatedly.In addition to this census, I made general observations, sitting quietlywatching one relatively small area and noting the interactions betweenlizards. Finally, there were continuous observations on individuallizards when we attempted to keep track of everything that an in-dividual did. These were mostly on the order of an hour or half anhour, but in one case we watched a lizard from before sum'ise to aftersunset.I supplemented observations on aggressive behavior with experi-ments with stake-outs. This technique, developed by Evans (1938a)and used by Ruibal (1961) in Cuba and by me in Puerto Rico, consistsof tying a lizard to the end of a stick, then placing it in the vicinityof another and recording the reaction of the resident.Throughout my time in Jamaica, even w^hen I was not engageddu"ectly in studying anoles, I was aware of them and watchedwhat they were doing. Many of my conclusions originated as generalimpressions based on these casual observations.Individual Ecological RequirementsThe social behavior of A. lineatopus can be considered as one ofthe adaptations of the species to its envu-omnent. The form of itssocial behavior is as strongly influenced by its envu'onment as isits morphology.In order to understand the social behavior of A. lineatopus, it isnecessary to consider the ecological demands of each individual andhow these are satisfied.Successful reproduction is a requu-ement of the population but it isnot necessary for the continued life of the individual. Social be-havior is so involved with reproductive behavior that discussion ofreproductive requh'ements is postponed until social behavior isbeing described.The demands of the individual are of two sorts: fundamentalrequu'ements and habitat requirements. The fundamental require-ments are those that an animal must satisfy to live. For A. lineatojmsthese include food, water, suitable temperatures, and protectionfrom predators.The habitat requu-ements are those structures in the environmentthat an animal uses to satisfy its fundamental needs. They areusually not necessary for the life of an animal, for a caged individuallives without them, but they are frequently those things that mustbe present if a wild indi\ddual is to settle in an area. For A. lineatopus
G PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122they include an observation and basking perch, cover near the ground,and a sleeping site. Fundamental RequirementsFood.?As is true of almost all lizards of its size, A. lineaiopusfeed primarily on insects and other small arthropods, occasionallyon very small vertebrates and snails.Anolis lineatopus waits for its prey on a perch a few inches to severalfeet above the ground rather than foraging for it. This huntingtechnique is common in visually oriented insectivorous lizards andcontrasts with the active searching technique that is typically used byolfactoril}^ oriented lizards such as skinks and teiids as well as somevisiuxlly oriented iguanids.Some prey A. lincato^yus can catch without having to move, butusually it must move and frequently to the ground or nearb};- foliage.Large individuals move farthest to catch something, sometimes goingfour or five feet. The approach is a quick run, often broken by oneor more pauses, and usually a pause just as the prey is reached.During a pause A. lineatojius cocks its head and looks at the prey withone eye and then shifts back to binocular vision. The binocularvision presumably uses the temporal fovea and allows better judge-ment of distance and du-ection, the monocular ^^sion probably usesthe central fovea, which may give better definition (Underwood,1951) . Sometimes A. lineatopus leaps several inches to seize an insect.A small insect is frequently eaten where caught but a larger one isusually carried back to a perch.The food items in the stomachs of 45 A. lineatopus are shown intable 1. The anoles were collected over a period of a week in anarea 10 x 50 3^ards in IMona. Thej^ were killed shortly after capture.Table 1.
?
Food items in the stomachs of 4o Aiiolis lineatopus
ANOLIS LINEATOPUS?RAND
measured, and their stomachs removed and placed in alcohoL vSiibse-((uently the contents were removed and examined. Some food itemswere found intact but many were represented by a scatterin^i; of more.resistant parts. A count of the minimum number of prey individualspresent was recorded.A covmt of prey items is perhaps the best count to indicate foodpreferences since each capture is weighted equally. It has the dis-advantage of ignoring the significant fact that one large insect maysupply more food than several small ones. Table 2, whicli shows theprey by size, makes some correction for this.Table 2.
?
Relation between prey size and lizard size (first fij^ure in each entryis number of prey individuals; figure in parentheses is number of stomachs)
8 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Anolis lineatopus feeds on a wide range of sizes of prey, the largerindividuals eating more large prey than do the smaller ones.The sizes of prey found in the stomachs of 45 lizards examined arerecorded in table 2 in the following categories: <5 mm; 5-10 mm;10-20 mm; and >20 mm. Entire insects were measured directly.Fragments were compared with entire insects, either from stomachsor in collections at the University of the West Indies. A few itemswhose size could not be estimated were omitted.The increase in number of large prey with increase in lizard sizeprobably reflects a greater ability of large lizards to handle large prey,since food is swallowed whole.The decrease in the number of small prey and the increase in rela-tive numbers of ants (which is marked only in the largest lizards)must reflect changes in the attitude of a lizard towards potential prey.Ants, unlike most of the insects A. lineatopus catches, forage up anddown the branches, tree trunks, and fence posts where the largerlizards regularly perch. To catch these ants, a lizard need moveonly a short distance, if at all. Apparently larger lizards differ fromthe smaller in taking small insects only when they can do so withminimum effort.The decrease in number of prej^ per stomach with increased lizardsize is probably largely a reflection of the increase in size of the preyitems; however, the smaller lizards are gro\^dng more rapidly than thelarger ones and so may require more food. In the smallest size class,both sexes are growing rapidly and, when the females' growth slowsdown, they begin to produce eggs, an activity that must requireconsiderable food. Dessauer (1955) calculated for A. carolinensisthat 'Tn the course of its 5-month laying season, the individualAnolis female must lay down the equivalent of her own total bodyprotein for egg production" (p. 12). The largest lizards are adultmales that are growing very slowly though expending considerableenergy in display. Harris (1964) foimd, however, that an Agamaagama took about the same weight of food per gram of lizard regard-less of lizard size.There are differences in micro-habitat associated with anole size(see p. 17) that may affect what is easily available for food to differentsized A. lineatopus.Whatever the reasons for it, A. lineatopus of different sizes living inone relatively small and quite uniform area differed in what they hadeaten. The smallest took large numbers of small prey, which werepredominantly not ants, and the largest took fewer prey, many ofwhich were large, plus an appreciable number of small prey, whichwere mostly ants. The change from one type of food to the otherseems to occur gradually.
NO. 3595 ANOLIS LINEATOPUS?RAND 9Harris (1964) showed conclusively that in Agama agama the dif-ferent size classes of lizards were taking largely different sized prey.Since adult males are larger than adult females and share the samehome ranges, this difference must reduce the possibility of intersexcompetition for food.Kennedy (1956) demonstrated differences between juvenile andadult Sceloporas olivaceous in size of prey and suggested that thisproduces "a minimum degree of intraspecific food competition betweenjuvenile and adult S. olivaceous which is highly advantageous in therapid development of juvenile lizards . . ." (p. 345). Blair (1960),however, in writing of the same species in the same area concludedthat the juveniles never approach the limit of their food supply.A difference in food size between adults and juveniles has been re-ported for lizards as varied as Eumeces fasciatus, Fitch (1954), andBasiliscus vittatus, Hirth (1963a).A habitat difference between juvenile and adults like that seen inA. lineatopus seems common in iguanids. It occurs in at least Basilis-cus vittatus, Hirth (1963a), Iguana iguana, Hirth (1963b), Uta stans-buriana, Tinkle et al (1962), Sceloporus olivaceous, Blah* (1960). Thishabitat difference must certainly reduce the possibility of one sort ofintraspecific competition for food. It also reduces the possibility ofthe adidts eating the young.Certain falcons are sexually dimorphic in size and the larger femalescatch larger prey (Cade, 1960). Storer (1952) has discussed thepossibility that this reduces intersex competition for food. Rand(1952) has reviewed a number of additional types of sexual dimor-phisms in birds that may also reduce intersex competition for food.In bii'ds as in lizards it is difficult to prove the applicability of theseideas to any particular case however plausible they are in general.To evaluate food as a limiting factor in the ecology of these anoleswould require a knowledge of both the food requirements of the ani-mals and the availability of food to them. Blair (1960) , for Sceloporusolivaceous adult females, and Milstead (1957a), for four species ofCnemidophorus, concluded that food was at least potentially limiting.Their arguments are convincing, but their proof is not rigorous. For
-1. lineatopus I suspect that food is potentially limiting but the dataare adequate only to demonstrate that in the study areas food isneither superabundant nor in very short supply.My impression is that insects of suitable size are common in Jamaicabut this idea must be treated cautiously for not all of the insectspresent are available to A. lineatopus as food. wSome insects are tooactive or too wary to be frequently caught. Others are noctm-naland still others too hard or protected by a nasty taste or smell.
10 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Of the 45 stomachs examined, only 1 was empty of food, suggestingthat most A. lineatopus catch enough food to keep their digestivesystems working most of the time; however, this can be said onlyof the period when they were collected?early in the dry season, aperiod when insects appeared abundant.A captive A. lineatopus will live at least several days to a weekwithout food, so presumably in the wild an individual could surviveshort periods of famine. I rarely saw A. lineatopus catch somethingto eat, and I have the impression that most days it catches only afew insects and some days perhaps none at all.In contrast to this slow rate of feeding is the response of A. lineatopusto food offered them. In my study area, these lizards became verytame and freely took insects tossed to them from a few feet away,running several feet to do so. In a typical session of 45 minutes oneafternoon, I tossed insects of moderate size into an area of aboutfom* square yards. During this time, 17 insects were captured byeight different lizards, two of them catching 3 insects and one catch-ing 4. This suggests that the slow rate of feeding usually seen isnot because A. lineatopus are satiated but because suitable insectsare not available.We saw in the field no thin or emaciated individuals among theselizards.The wide variety of prey taken by A. lineatopus means that availa-ble food is distributed throughout the habitat and is not denselyconcentrated in certain areas. Several lines of circumstantial evidenceindicate that probably few, if any, established resident A. lineatopusstarve to death but at the same time that a superabundance of foodis not available to them.Water.?Probably all the water A. lineatopus requires, beyondthat present in its food, it gets by licldng drops of dew or rain fromleaves and twigs with the tip of its tongue. The only report I knowof lizards having difficulty seeming water is Evans (1951), whoreports Ctenosaura pectinata making long excursions to drink.Rain is a frequent occmTence around Kingston. Even in the dryseason when no rain falls, dew forms almost every night and remainsin the shade until mid- or sometimes late morning. Water is thusavailable to A. lineatopus almost every day.Anolis lineatopus were only occasionally seen to drink in the fieldbut in cages they were seen to drink frequently.As I sat watching some marked A. lineatopus one afternoon, a lightrain began to fall. Only one, an adult male, reacted. He chmbed abranch to an orchid and licked several times at one of the wet leaves.An A. lineatopus, which woidd eagerly take tossed insects, com-pletely ignored the drops of water that formed when I watered the
NO. 3505 ANOLIS LINEATOPUS?RAND Hstudy area. Apparently even during this period in the dry season,when there was not enough rain to keep the grass green, A. lineatopuswere not short of water.Temperature.^?Anolis lineato'pus in the Kingston area has apreferred body temperature (or eccritic temperatui'e) range of 2S?-31? C. Sixty-one of 85 lizard temperatures taken under a variety ofenvironmental temperatures fall within this 4? range. Like manyother lizards which have been studied (Bogert, 1959; Fitch, 195Gb;Inger, 1959; Ruibal, 1961), this species maintains its temperaturewithin this range b}^ behavioral thermoregulation when environmentalconditions permit.I have continued to use the familiar term "preferred body tempera-ture" despite the demonstration of Licht, Dawson, and Shoemaker(1965) that observed field temperatures in some species differ fromthose chosen in gradient experiments and their suggestion that theterm "preferred temperature" be restricted to experimental studies.The A. lineatopus temperatures used here were taken over a widerange of air temperatures but always when the sun was shining.On each occasion a thermal gradient was available to A. lineatopusas it was to the lizards in Licht's and his coworkers' experimental set-ups. On most occasions, the environmental temperatures available,considering substrate as well as air temperatures, extended both aboveand below the observed A. lineatopus range. It seems likely that theobserved body temperatures approximate those that would be foundin an experimental gradient. Certainly my observations should bemore similar to those of an experimental gradient than to the resultsobtained by measuring all lizards regardless of weather.The cloacal temperatures of^l. lineatopus were taken with a Schul-theis quick reading 0?-50? C thermometer within a few seconds ofnoosing.The body temperature of an A. lineatopus is almost always close toand above the air temperature in its vicinity (fig. 1). Body tempera-tures are higher at high air temperatures than at low air temperatures,but the difference between au' and body temperature is greatest atlow ah' temperatures as the following figures (in centigrade) show
:
air tempemlure
12 PROCEEDINGS OF THE NATIONAL MUSEUMThe distribution of A. lineatopus temperatures is skewed so thatthere are fewer records (7) above the preferred body temperaturerange (28?-31? C) than below it (14) and the total range extends only1? above, but 4? below, the preferred range.35 ^C
25 27 29 31Air TemperatureFigure 1.
?
Anolis lineatopus cloacal temperatures plotted against air temperature inimmediate vicinity at time of capture (temperatures have been recorded to nearest 0.2degree and have been rounded to nearest degree for plotting).Heath (1964) in his report on the temperatures of beer cans hasshown that body temperatiu-es above the ambient ah' temperatures donot necessarily demonstrate thermoregulation. But the alreadydescribed relationships of body and air temperatiu-e combined withthe behavior patterns reported below are convincuig evidence thatA. lineatopus thermoregulates.Color change in A. lineatopus may play a role in thermoregulation.Casual observation indicates that in the early morning, when A.lineatopus are sunning themselves, they are darker than they areduring the heat of the day. Since this involves a gradual change in
ANOLIS LINEATOPUS?RAND 13
shade rather than an abrupt change in hue, it is difficult to measurein the field and I collected no quantitative data on the point.The lizard's behavior in moving into and out of the sun was moreconspicuous and more easily recorded than the color changes andprobably more important in thermoregulation.The following extract from my field notes shows the sort of behaviorfrequently seen
:
30 January?Barbican brush heap, 900 hrs. By this time the lineatopus aremostly sitting in the shade. I have the impression that they are avoiding hotperches more than the sun itself. A large male is on a branch which is shadedbut he liimself is largely in the sun.One adult female, on the other hand, is sitting in a small patch of shade on abranch in the brush heap. Most of her time is spent in these patches of shade,where she usually sits on top of the branch. When she does leave the shade fora sunny branch, she usually runs along the top, but when she stops she moves into the shade on the side. Sometimes when she runs along a sunny branch shewill stop and run very quickly back to the shade patch as if she had suddenlybecome too hot.Table 3 shows the residts of seven censuses of adidt male A. linea-topus along a stone aqueduct that runs north and south. Theselizards moved into the sun when they were cool and avoided the sunand hot substrates when they were warm. They seemed more careful
T.-VBLE 3.
?
Distribution of adult males on a stone aqueduct binder different weatherconditions (aqueduct runs north and south and all data was collected in themorning; temperature readings are given in centigrade)
Weatber
14 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122about avoiding very hot conditions than in seeking warm ones whenit was cool.Anolis lineatopus may remain active at body temperatures ap-preciably below its preferred body temperatiu^e range. Most lizardsstudied (Bogert, 1959; Fitch, 1956; Milstead, 1957) become inactiveat low temperatures but A. lineatopus do not do so, and I have seenthem on cloudy days and around lights at night feeding, fighting, andcopulating at environmental temperatures below 25? C when theirbody temperatures must have been circa ambient.In contrast to these observations of activity at low temperaturesare those made during the very early morning. Anolis lineatopus lefttheir exposed sleeping sites shortly after dawn but well before sunriseand moved into spots where they were concealed. They moved uponto their usual perches only after sunrise, when they began to bask.Studies on other lizards have shown that the preferred body tem-peratm'e is usually not far from the upper lethal temperature andthis is probably true in A. lineatopus. On analogy with other lizardsone would expect a thermal death point somewhere in the vicinityof 40? C Obviously A. lineatopus cannot occur in places where itstemperature would be forced above the thermal death point; it shouldbe noted that on the sunny aqueduct mentioned above (table 3)one of the stones of the aqueduct had a temperature of 41? C.The effects of temperatures below the preferred body temperatureare almost completely unknown not only in this species but alsoin most lizards. At very low temperatures, A. lineatopus becomessluggish and torpid, as the following illustrates. One male cooledin a refrigerator was sluggish when cooled to a body temperature of13? C though still active at 19? C.Presumably deleterious effects of moderately low temperature areassociated with the slowing of temperature-dependent physiologicalprocesses such as heart rate, oxygen consumption, rate of enzymaticaction (Bartholomew and Tucker, 1963, 1964; Bartholomew, Tucker,and Lee, 1965; Licht, 1961).Hardy (1962) has reported that in Cnemidophorus sexlineatus, atlower temperatures, defecation is delayed and individuals are lesseflScient at detecting prey, particularly motionless prey, and takeseveral times longer to dig tunnels. No similar changes were seenin A. lineatopus, but it was noted that A. lineatopus are shyer atlower temperatures and it was suggested that this might be a be-havioral compensation for the slowing of muscle or nerve reaction atlower temperatures (Rand, 1964b).Preferred body temperature range may be an important ecologicallylimiting factor, for this subspecies is absent from dark forest where
NO. 3505 ANOLIS LINEATOPUS?RAND 15
sunlight is not available and also from higher elevations where tem-peratures are lower.In the lowland edificarian and disturbed areas where these studieswere made, there are probably few places from which these anoles areexcluded because of generally low temperatures. They do not occurwhere there is no protection from overheating in the sun.Predators.?Every animal in Jamaica that preys on small landvertebrates or large insects probably eats A. lineatojms at least occa-sionally, but in the areas studied the predation pressure does not seemto be heavy.From my observation, the most important predators are domesticcats. One well-fed mother cat that was living with us brought herkittens at least one and sometimes several anoles every day, manyof them adult A. lineatoinis. However, this made no obvious diminu-tion in the population of anoles living around the house.I also saw dogs catch and kill A. lineatopus, and some of themprobably do this regularly. Chickens also eat these lizards. I foundan Anolis in the stomach of one of the few snakes I examined, a Dro-micus callilaemus. Even though this species of snake is moderatelycommon (one could almost always find one in a couple of hours ofsearching), it is a ground-li^'^ng species and probably not an importantpredator on any Anolis. Other snakes are rare in the study area.Birds are probably important predators on small Anolis; Wetmore(1916) reports that in Puerto Rico he found Anolis in the stomachs ofmost of the insectivorous birds, even as small as the tody, and thissituation probably applies to Jamaica also. The most common ofthe larger insectivorous birds around Kingston?mockingbirds, anis,and kingbirds?probably take young and female but few adult maleA. lineatopus.The birds that might be expected to feed on adult males, the largercuckoos, herons, hawks, and owls, are relatively rare in the studyareas and probably are thus of relatively little importance.The common toad, Bufo marinus (I counted 25 on the lawn oneevening), certainly could eat small to moderate-sized A. lineatopus,but, since the toads are nocturnal, they probably catch few.Anoles themselves eat lizards. Anolis grahami at least occasionallyeat young A. lineatopus and the larger A. valencienni and A. garmaniprobably regularly eat at least adult females and young males of A.lineatopus. Both these larger species are relatively common (an houror two search in the correct habitat would reveal at least one of each),but both are primarily hzards of tree crowns and consequently do notfeed in the places where A. lineatopus are most common.I only once saw an A. lineatopus, a 60 mm male, catch and eat oneof its own species, about 20 mm long, but I have several times seen an
16 PROCEEDINGS OF TIIE NATIONAL MUSEUM vol. 122
adult male chase and attempt to catch a young A. lineatopus and Ihave also taken a small A. lineatopus (just above hatchling size) fromthe stomach of an adult male. Cannibalism is probably relativelyrare, partly because the young lizards are too active to catch easilyand partly because they avoid the principal perches of the adultmales. The difference in perch between young and adult males maybe in part a direct result of chasing by the adult males.Predation probably is not an important factor in controlHng popu-lation density of adults resident in favorable habitats. Predation isprobably heavier on juveniles, on dispersing individuals, and on thoseliving in unfavorable habitats.Escape behavior.?Most of my information regarding the re-action of anoles to potential predators relates to their reaction topeople. It is possible but unlikely that they react differently tosmaller predators.I noted that the first reaction of A. lineatopus to an approachingdanger is usually to remain still, sometimes flattening against theperch. If the danger approaches closely, the A. lineatopus runsaround the perch to the other side and either up out of reach or downinto the vegetation at the base. Where possible, adult males usuallyrun upward, small A. lineatopus more frequently dodge around on thetrunk or even leave the tree to hide in the cover on the ground.Adult males, when they became famiHar with me, did not fleeat my close approach but displayed as they woidd at another male.A captured hzard usually tries to bite and, if successful, holdson. An adidt male, when seized, frequently shows most of his ago-nistic displays (see p. 38 et seq.), raising dorsal and nuchal crests,lashing his tail, opening his mouth, protruding his tongue, and some-times holding his dewlap open but never flashing it as does a maledisplaying to another male.A captured lizard frequently defecates; the feces produced areusually a pasty material, though they may be very watery or some-times a normal, dry, compact feces. This material is not pleasant tosmell nor presumably to taste but it does not smell very unpleasant.Elimination of wastes.?As many arboreal animals do, A.lineatopus defecate on whatever perch they happen to be and the fecesusually fall to the ground. As might be expected in a species whereolfaction is of little importance, the feces do not seem used in markingas Hardy (1962) has suggested they are in Cnemidophorus sexlineatus.The fecal pellets are roughly cylindrical, about 1 cm long, dark incolor, and dry and compact to touch. There is a small white cap onthe end extruded first; presumably the dark material is fecal materialfrom the intestine and the white cap is nitrogenous wastes from thekidneys.
NO. 3B95 ANOLIS LINEATOPUS?RAND 17Habitat RequirementsThe three main habitat requirements, perch, cover, and sleepingsite, must be satisfied in some way or other within the home range ofeach A. lineatopiis. These features of the habitat are essential to theoperation of the normal behavior patterns that enable ^4. lineatoiyusto satisfy its fundamental requirements and also important to thoseassociated with social behavior and reproduction. It is probablythese habitat requirements that are used as cues in habitat selectionby A. lineatopus.Harris (1964), writing on Agama agama in Nigeria, states: "Threeimportant structural components the environment must provide are:(a) suitable display posts, (b) a roosting place and (c) convenientlysituated refuges from predators" (p. 132). These categories arealmost identical to those used herein for ^4. lineatopus though thestructures in the habitat that satisfy these demands are quite difl"erent.Perches.?An A. lineatopus spends its days waiting on one oranother of its perches. From its perch the lizard sees most of theprey which it catches, most of the other lizards which it courts orchases, and most of the predators from which it flees, and it is fromits perch that most of its display is given. The perch also provides asunning site and usually shade as well.A Ande variety of objects are used as perches, among them trees,fence posts, rocks, walls of houses, bushes, and hedgerows. The largemajority of ^. lineatopus seen were within six feet of the ground, andrelatively few were seen on slender twigs and branches. In general,adult males perch farther from the ground and on larger diameterperches than do smaller individuals, which are usually seen in bushes,hedges, and brush piles very close to the ground and which seem toavoid the large trees and fence posts, etc., that the adults prefer.Collette (1961) describes shnilar intraspecific differences in perchpreference in A. sagrei and ^1. porcatus. Intraspecific difl'erencesbetween juvenile and adult habitats seem widespread in iguanids(see p. 9) (For a more detailed discussion of perches, see Rand,in press). The sorts of perches that are occupied most commonly areherehi called preferred perches, and those that were occupied lessfrequently are referred to as less or subpreferable.The same individual may be seen on the same perch da}^ after day;most, if not all, lizards use only one or a few perches as the centerfor theh' activity. Blair (1960) noted similar behavior in Sceloporusolivaceus, particularly in tlie females. Anolis lineatojms may spendmost of its time on only a small part of one percli, as O'Brien et al(1965) describe for Sceloporus undulatus. I have called tlie perch orperches where an individual spent most of its time its "usual" perch orperches.240-241?67 2
18 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Anolis lineatopus usually rests on the side of a vertical perch or onthe upper surface of a slanting or horizontal one, most frequentlywith the head pointed toward the ground, a habit associated, Isuspect, ^^ith the large amount of food taken on the ground, as Isuggested for A. cybotes, a species with similar habits, in Hispaniola(Rand, 1962). Though A. lineatopus may spend long periods on thesame perch, it seldom spends more than 15 minutes \nthout shiftingposition or without displaying. Even between shifts the lizard isusually alert, and in one 4-minute period when I watched an adultmale on his perch, he moved his head 12 tunes and the eye that I couldsee also 12 times (the eyes move independently of each other) . Duringthis 4-minute period, the longest time mthout any movement was45 seconds and usually only 10 or 15 seconds elapsed.The sort of perch an anole uses varies from species to species so thatsympatric species occupy different microhabitats (Collette, 1961;OHver, 1948; Ruibal, 1961; and Rand, 1962, 1964a). This probablyacts to reduce interspecific competition (Rand, 1964a) just as thedifferent foraging zones do for sympatric wood warblers (MacArthiu*,1958).Cover.?Cover near the ground is a more important habitatrequu-ement for female and juvenile A. lineatopus than for adultmales. Very few A. lineatopus maintain home ranges that includeno cover at all. Cover takes several forms: dense vegetation, grass,herbaceous plants or ferns that grow around the perch; a pile ofdead sticks; a layer of large dead leaves; or occasionally a pile ofrocks. These provide protection against temperature extremes(p. 14) and predators (p. 16).Sleeping sites.?Anolis lineatopus sleep dm'ing the night inexposed situations above the ground, generally at elevations of 1 to10 feet, usually at the ends of branches, the larger usually climbinghigher than the smaller individuals. They sleep on leaves, smalltA\dgs, and sometimes on grass blades that frequently are slenderenough to bend double under the weight of tlie anole. Many arborealand semiarboreal iguanids have sunilar sleeping habits, among themIguana iguana, Hirth (1963b), Basiliscus vittatus, Hu-th (1963a), atleast some Sceloporus olivaceus, Blair (1960), and Urosanrus ornatus,Cowles and Bogert (1936).Sleeping A. lineatopus are usually visible from outside the bushor tree and consequently are exposed to any flying predator huntingby sight at night; however, they are concealed from any predatorthat approaches from inside the tree and is forced to climb the branchesto reacli them. The small diameter of the sleeping perches presumablywould protect them from any large climbing predator, and even a
NO. 3596 ANOLIS LINEATOPUS?RAND 19
small predator would probably be unable to reach a lizard wdthoutslialdng the perch enough to awaken it.Though easily awakened at night by a light or by movement ofthe perch, the lizards are slow to leave their perches and usuallydo so only when they are shaken quite violently.The sleeping sites differ sharply in character from the preferreddiiu'nal perches; the sleeping sites are places this species seldomvisits during the day except when catching an insect or when engagedin a dispute mth another anole.Some individuals seem to use the same leaf or twig as a sleepingsite night after night. Most, however, are not so regular, thoughthey tend to sleep in the same bush, tree, or bunch of grass. There isdefinitely not the same attachment to a particular sleeping sitethat there is to the daylight perch.Anolis lineatopus left their sleeping places soon after it becamelight and usually did not settle do^\^^ to sleep until it \\'as too darkfor them to be seen even from a few feet away. Blah (1960) reportsa similarly long period of activity for Sceloporus olivaceus during thesummer as does Hhth (1963a) for Basiliscus vittatus. Perhaps thisis characteristic of lizards sleeping, as these do, in exposed sites ina warm climate.On one occasion, January 27 in Barbican, I watched a male A . lineatopus go tosleep. I first noticed him at 6:13 p.m. when he ran up the stem of the croton andstopped just below the terminal bunch of leaves. The sun had set but the hind-scape was still bright. A minute and a half later he chmbed up among the leavesand out along one of them and stopped in a typical alert position, head raised andneck bent. Two and a half minutes later he flattened out against the leaf, stillfacing its tip. Six minutes later he turned around facing the stem of the leaf andflattened against the leaf. The sky was still light but the landscape dark. By6:30 the first stars were out and it was almost full dark. With the flashlight Icould see that the lizard's eyes were still open but he was in his normal sleepingposition.Both Mertens (1939) and Rand (1962) noted in Hispaniola thatcertain species of Anolis sleep with their hind legs flexed while otherssleep with these legs stretched out along the body. Anolis lineatopusmay sleep in either position and quite frequently a sleeping A. linea-topus was found wdth one leg flexed and the other extended backward.Home Range and Activity RangeAn A. lineatopus seldom travels far and most of the area it visitsis visible to it from its usual percli.I use the term "activity range" for the area visited regularly duringthe day including the usual perches but excluding the sleeping sitewhen it is outside of this area. "Home range," as used here, includes
20 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122both activity range and sleeping site. The activity range is describedin terms of its horizontal extent and, since A. lineatopus are largelyarboreal, reference to its vertical extent is also included. I havefollowed the suggestions of Milstead (1961) and, in calculating ac-tivity range size, have omitted the occasional visits that certain A.lineatopus made to perches well outside of the area where they wereusually seen, but I have plotted these visits on the maps included inthis paper.Anolis lineatopus may sleep in the same area where it spends theday, perhaps climbing up and out onto one of the branches of itsprincipal perch. It may leave the area where it spends the day,however, if there is no suitable sleeping site, and travel several yardsto find one. Anolis lineatopus that have mutually exclusive activityranges may all use the same sleeping sites.Of 15 adult females living in Barbican [fig. 6] for which I have adequate dayand some night records, 7 were recorded sleeping only within their activity ranges,6 both in and outside of their activity ranges, and 2 only outside. The sleepingplaces usually were less than 5 feet away from where the female spent the day,but for one female they were 10 and 12 feet away. Of the 10 adult males [fig. 4],5 were recorded sleeping only inside their activity range, 3 both in and out,and 2 only outside their activity range. Two males regularly traveled 10 feet tosleep but for other males the records outside of the activity ranges were mostly3 to 5 feet away.The concept of an activity range wdth a definite size is useful sinceit gives a datum for comparison mth other animals. It is, at best, acrude approximation of the space actually utilized by the lizard.First, it takes only slight account of the vertical range of the animaland, for an animal that spends most of its waking horn's above theground, this is important. Second, the individuals do not visit allparts of their activity ranges equally nor do they always range outin a regular manner from a single center. Finally, an activity rangehas definite boundaries only where it adjoins the activity range ofanother A. lineatopus of about the same size (see p. 46 et seq.).Both the size and the shape of an activity range varies mth the sizeand sex of the individual and with the number and distribution ofperches available.For adult males the usual activity range varies between 3 and 10square yards, seldom probably is less than 3 and sometimes probablyas much as 20 square yards. It usually extends vertically from 1 to 3yards, again sometimes less but probably never less than 1 foot andseldom more than 5 or 6 yards.The shape is influenced strongly by the distribution of the perchesused. Where a single perch is used and the surrounding area isuniform, the activity range is roughly cu'cular -with the perch in thecenter. The activity range of the adult male (no. 4) mapped in
ANOLIS LINEATOPUS?RAND 21figure 2 approaches this. He Hved on an isohxted tree and in the 40times he was observed over a period of 21 days, he was on the trunkof the tree below 20 feet (only 6 times going higher than 8 feet or tothe ferns at the base, which extended only 3 feet out from the tree).
Figure 2.?Record of observations of adult male Anolis lineatopus (63 mm) on isolatedIndian almond tree, August 22 to September 15 (dots= observations on near side oftree, x's= records on opposite side; preponderance of records on near side primarily reflectsusual position of observer).
22 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122His activity range was 3 square yards on the ground and extendedupward about 3 yards.In areas with more perches available, the activity ranges are largerin horizontal dimensions. In the Barbican brush heap sho\\Ti infigure 3, the three resident males had areas of approximately 13, 8,and 6 square yards; the first of the males using a small tree regularlyclimbed to 6-8 feet, while the other climbed only to about 3 feet.The activity ranges of these males were ovals except that the largesthad an elongate projection along the fence.The other males at Barbican [fig. 4] had activity ranges that fall within the3-10 yard size range. The relationship of perches to activity range shape isconspicuous in figure 4. The activity range of no. 2 was located on the frontporch of my house. He used the L-shaped railing as a perch; the observationrecords of him outline the railing just as the concentration of observations alongthe fence indicates the position of the fence posts.On an abandoned stone aqueduct, the males were found living mostly underthe arches and usually restricting their activity to a single arch. Those 16 maleson which we have adequate data had activity ranges varying from 4 to 16 squareyards in area and mostly between 4 and 8 square yards. They extended verti-cally between 2 and 10 feet. The activity ranges were mostly quadrangular inplane view, reflecting the shape of the area under an arch.The size of the activity ranges of the adult females is usually con-siderably smaller than that of those of the adult males, and againstructure of the habitat affects size and shape. In general, the femaleactivity ranges averaged about 1 to 2 square yards, sometimes beingas much as 3 or 4 square yards and probably sometimes only 2 or 3square feet. In vertical extent the females usually ranged from 1 to 5feet and probably sometimes more.In the isolated tree shown in figure 2 there were 9 resident females. Of these,8 concentrated their activities below 3 feet, in the ferns at the base and the rockborder around it, occasionally climbing the tree to the height of 6 or 8 feet. Onefemale concentrated her activity on the trunk and the side branches between 5and 9 feet above the ground and only rarely came down into the ferns. Foreach of these females, the activity range was about 1 square yard in horizontalextent.In the Barbican brush heap [fig. 5], which was more complex in perch distri-bution, 3 adult females (nos. 59, 145, and 162) had activity ranges of 1, 1, and 2square yards and ranged vertically between }4 and 4 feet. The other females inBarbican [fig. 6] have activity ranges that are usually between 1 and 3 squareyards in area.The activity ranges of lizards below adidt size are somewhatsmaller than those of the adults of the same sex and increase as thelizards grow.Figure 3.?Activity ranges of three adult males (no. 1=59 mm; no. 6=63 mm; no. 159=58mm) in Barbican brush heap, January 23 to February 23 (hea\y continuous lines=approximate activity range boundaries, fine lines= physical features, remaining lines=boundaries of brush heaps).
ANOLIS LINEATOPUS?RAND 23
0*'"''/,
5 fi
24 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122
Figure 4.? Distribution of adult males larger than 50 mm seen four or more times atBarbican in fenced yard around author's house September 30 to November 7 (dot=dayrecord of each position where lizard was sighted, circle^ night record; small numberadjacent to dot=more than one record in same position; lines around records for eachlizard enclose their activity ranges but indicate boundaries only roughly; observations along distance from activity range are joined by line that shows only which lizard wasinvolved and not path it took to reach place observed). Three of the males were presentfor only part of this period. All but no. 91 were marked by October 1. He appearedOctober 19 on a small, previously unoccupied, tree and was seen there until October 24,when he was found dead in the carport (probably killed by our dog and perhaps carriedthere by him). No. 2 was found dead on November 1. No. 33 disappeared aboutOctober 30 and at about this time no. 45 moved along the wall to occupy no 33's area.These records of no. 45 following no. 33's disappearance have been omitted. (No. 1==59mm; no. 2=56 mm; no. 3=59 mm; no. 6=62 mm; no. 7=62 mm; no. 8=67 mm; no.11=61 mm; no. 13=65 mm; no. 33=64 mm; no. 45=67 mm; no. 46=63 mm; no. 58=67mm; no. 91=55 mm.)
Figure 5.?Activity ranges of females and juveniles in part of Barbican brush heap (seefig. 4 for activity ranges of adult males; activity range boundaries were constructed bylinking marginal records for each lizard; heavy solid lines= larger individuals, dottedlines= smaller lizards, thin solid lines= structural elements, remaining lines= boundariesof brush heaps; period covered= January 23 to February 23, except for nos. 162, 148, and163, for which boundaries represent activity ranges before shifts discussed in test (pp. 49and 52) occurred; in the following list, dates of measurement are given because certainlizards show appreciable growth: no. 59 female= 41 mm, January 23; no. 145 female= 43mm, January 9; no. 148 female=26 mm, January 9, to 28 mm, February 1; no. 156female=25 mm, January 10, to 28 mm, February 7; no. 161 female=29 mm, January 10; no. 162female=ca. 39 mm; no. 163 female=26 mm, January 24, to 28 mm, February 9; no. 164female=33 mm, January 26, to 34 mm, February 9; no. 166male= 22 mm, January 30; no.171 male= 31 mm, February 9, to 35 mm, March 11; no. 175 female=24 mm, February 9).
26 PROCEEDINGS OF THE NATIONAL MUSEUM
Figure 6.?Distribution of adult females in Barbican September 30 to November 7 (samearea and lineal elements as figure 5; all lizards seen more than twice have been mapped;no. 14=44 mm; no. 15= 44-45 mm; no. 16=41-42 mm; no. 17=42 mm; no. 18= 42-43mm; no. 20=42-43 mm; no. 26=43 mm; no. 32=42 mm; no. 37=43 mm; no. 39=39-40mm; no. 41=44?45 mm; no. 54=42 mm; no. 56=43 mm; no. 59=39-40 mm; no. 61=42mm; no. 62=44 mm; no. 72=43 mm; no. 80=45 mm; no. 120=42 mm; no. 125=41 mm).
The best data on young females come from the Barbican brushheap (fig. 5) , where the activity ranges of 6 small females were plotted.These varied in area between 1 and 16 square feet and their shape wasstrongly influenced by the distribution of the larger branches in thebrush heap. The larger activity ranges were mostly held by thelarge individuals.
ANOLIS LINEATOPUS?RAND 27The best data on young males come from a study of the Monahibiscus hedge (fig. 9, table 4) where the activity ranges of 7 werestudied. These ranged from about 4 to 18 square feet, being about4 to 10 for the smaller individuals and about 10 to 18 for the larger.Since even adult male A, lineatopus activity ranges seldom reach150 square feet, they are well below average size for land vertebrates.No birds or mammals are known with home ranges nearly as smallas this; however, at least one frog has a home range of about thesame size
?
Scaphiopus holbrooki with an average home range of108.4 square feet (Pearson, 1955).Temperate North American iguanids have small home ranges buteven the species with the smallest, Uta stansbitriana, with an averageadult male home range of 2600 square feet (Tinkle, et al., 19G2),has a range considerably larger than A. lineatopus. In contrast,two tropical lizards have home ranges that lie between that of Utastanshuriana and Anolis lineatopus?Basiliscus vittatus, Hirth (1963a),and Agama agama, Harris (1964). It may be generally true thattropical lizards have smaller home ranges than do their relatives intemperate regions.Typically A. lineatopus visits all of its principal perches every day,frequently more than once. The frequency of forays that it makesfrom the principal perches is inversely related to the length of theTable 4.
?
Activity range use in Mona hibiscus hedge (areas in square feet; seep. 28 for discussion)
Lizard
28 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122forays. To measure the utilization of activity range, I recorded at5-minute intervals the position of each A. lineatopus present in asmall area of hibiscus hedge during 10 periods of about Yi hour eachApril 18-23. These records for each individual were compiled and 3convex polygons drawn, one connecting the outermost points, the next,the smallest including 75 percent of the observations, and a third, thesmallest including 50 percent of the records. All the records (April-June) were added and a polygon enclosing them was drawn. The100 and 75 percent polygons are shown in figure 9. The area of eachpolygon was calculated. The results (table 4) show that each anolespent half his time in a very small area compared with that which itvisited. Fifty percent of the records fall in areas ranging from 4 to35 percent of the respective areas enclosing all of the census records.In 9 of the 11 cases it is less than 20 percent and in 6 it is less than 10percent.The addition of other records for AprU, May, and June made agreat increase in area in only 1 case. In the other 9 cases it lessthan doubled the area.The principal perches of A. lineatopus and the area immediatelyaround them form what might be called the core area of the activityrange, following the usage of Kaufmann (1962). As he found inthe raccoon-like coatimundis, the core area is not only the part mostheavily used but also the part of the home range with the least overlapwith those of others (see p. 46).The activity range of an adidt A. lineatopus seems relativelypermanent and certainly shows no seasonal variation. Subadults andparticularly young males are much more mobile.In the abandoned aqueduct, that I observed, 5 adult males marked on October19, were found the following June 1 and all were in the same places where theyhad been marked. In the yard at Barbican, of the 16 adult males seen more thanonce in September and October, 7 were still present in the middle of March. Ofthese, 3 had shown slight shifts to adjacent areas and 4 had shown none. Ofthose 9 which had not been seen in March, 2 were dead, but it is possible that theother 7 had shifted their areas outside of the study plot.Of the 24 adult females seen more than once during September and October, 13were seen in March. Of these, 3 had moved in the interim, all to adjacent areas.Again it is quite possible that some of the females not seen in March had movedrather than had died.Smaller shifts, which did not involve abandoning the original area, were morecommon. Perhaps the commonest were brief excursions lasting less than a dayand peripheral to the usual area. These were recorded in both adult males(up to 20 feet, 7-10 feet being more common), and females (up to 13 feet, 4-5feet being more common). Also, not infrequently observed were shifts involvingthe abandonment of part of the area and certain perches and extending the areato new perches in some other direction. Finally, at least 2 of the females madetemporary shifts during the period of September to November 7, abandoning theoriginal area for an adjacent one for a few days and then returning.
NO. 3595 ANOLIS LINEATOPUS?RAND 29A number of the adults marked in September through October but not seen inMarch probably were marked during temporary shifts and occasional excursionsinto the study area.I have no data for longer than 6 months but several members of thefaculty of the University of the West Indies have told me of A.lineatopus that have lived in the faculty's houses and used the samepieces of furniture for perches over several years. These stories,based on unmarked animals, are inconclusive but not improbable.The reasons for the shifts observed in the adults include changes inthe structure of the environment through the cutting and growth ofvegetation and interactions with other adults.It is possible that changes in available food supply affect the areautilized, a concentration of insects attracting a lizard to a new areaand a scarcity forcing a Uzard to abandon an area, but I have noevidence on this point.Both young males and young females may hold activity rangesseveral days or weeks and perhaps longer, but their activity rangestend to be less stable and less permanent than those of establishedadults.Twenty-five young males in Barbican [fig. 7] were marked from Septemberthrough November 7. Only 17 of these were seen more than once and 9 of thesewere seen again in March. All 9 had made changes in their areas. Unlike theadults, only 3 of these changes were to adjacent areas, the other 6 were to distantspots, over distances of 45, 48, 35, 35, 45, and 85 feet, respectively.Even during the period of intensive observation (September through Novem-ber 7), disjunct shifts were common. Of the 17 young males seen twice or more inthis period, 8 showed changes. Six of these were disjunct with distances of 54,65, 25, 52, 70, 35 feet, respectively. That these shifts average no farther overthe short September-November period than over the long September-Marchperiod is indicative only, I think, of the small size of the study area.The young females in this area show similar instability but less strikingly sothan do the young males. Twenty-nine young females were marked. Of these,14 were seen more than once and, of these, 5 were present in March. Of these 5,2 show changes, neither disjunct. Of the 12 females seen more than once duringSeptember through November 7, only 2 show changes and both of these aredisjunct, one of 54 and the other of 55 feet, respectively. The young femalesare very small and thus frequently missed, and, as the mapping technique was toocrude to show very small shifts, these records understate the amount of change.Even so, these young females seem to be more stable than the young males, atleast in terms of long distance shifts.Though the activity ranges of young A. lineatopus are more liable to changethan those of the adults, the lizards do not usually wander randomly. The brushheap area included the areas of 6 small females and 1 small male. BetweenJanuary 23 and February 23 4 of these 6 females shifted their activity rangesbut in each case only to an adjacent area.In one of these 4 shifts, the lizard was chased from her original activity rangeby another A. lineatopus. In the other 3 cases, the causes of the shifts are un-known but may relate to the changing requirements of the lizards with age andgrowth.
30 PROCEEDINGS OF THE NATIONAL MUSEUM
Figure 7.?Distribution of young males at Barbican (same symbols as fig. 5 for period ofSeptember 30 to November 7; x's= sightings of lizards several months later, March 8 to14; all individuals seen more than once during first period and all seen in both periods aremapped; unlike most adults in figs. 5 and 7, these young males showed importantgrowth and first and last measurements are given for them where available; no. 21=47mm, September 26; no. 28=31 mm, September 26, to 37 mm, October 31; no. 35=44 mm,September 27, to 59 mm, March 10; no. 38=32 mm, September 27, to 38 mm, October 24;no. 44=31 mm, September 29, to 38 mm, October 29; no. 48=44 mm, September 30;no. 65=42 mm, October 5, to 62 mm, March 13; no. 79=29 mm, October 9, to 30 mm,November 2; no. 81=44 mm, October 9, to 65 mm, March 13; no. 82=19 mm, October 9,to 47 mm, March 10; no. 83=18 mm, October 16, to 40 mm, March 11; no. 93= 23 mm,October 19, to 57 mm, March 12; no. 95=37 mm, October 19, to 59 mm, March 10;no. 112=22 mm, October 24; no. 113=45 mm, October 42, to 60 mm, March 14; no.114=37 mm, October 24; no. 118=28 mm, October 29, to 43 mm, March 9).
NO. 3595 ANOLIS LINEATOPUS?RAND 31In the Mona hibiscus hedge studied from April 10 through June 1 there were10 small males. Of these, 4 changed their activity ranges and the other 6 re-mained roughly the same.The data for Barbican indicates that the adults of both sexes havepermanent activity ranges and such shifts as they do nud^e are usuallyto adjacent areas. There is some data from the study area in Monathat contradicts this, though it is a contradiction in detail rather thanin substance. When I moved to Mona in mid-April, there were onlya few large males that were living on the larger trees, particularlytrees with vegetation near the base, a bed of ferns, or a bush. Only afew of these males were marked?those resident in the vicinity of thehibiscus hedge that were being studied in detail?but casual observa-tions were made on the others. It became obvious in the period ofmid- to late May that there were many more adult males present andthat they were living on trees which had previously been unoccupied.Many of them were smaller than the males which had been present inmid-April. Some of these in one area were marked and their move-ments watched for about two weeks (fig. 8). These data supportedmy impressions that there w^ere a number of unestablished adultmales moving about in the area. The data show also that there werecertain males that did not change their activity ranges dvu'ing theobservation period. The latter males are mostly larger than theformer. My interpretation of the data is that the smaller males hadrecently matured whereas the larger males were older. The oldermales had been occupying the preferred perches and were successfulin keeping the smaller individuals from establishing there, forcingthem to spend theu' time on less favorable perches. That the youngmales were living in subpreferable areas, coupled with the possibilitythat they have less site attachment, accounts, I think, for theirmovements. This is essentially the explanation offered for thesituation observed in Barbican. There is a striking difference, how-ever, in the size of the individuals involved in the two situations. InBarbican the males that made activity range shifts were mostly sub-adults the size of females; in Mona they were fully adult. I thinkthere is a temporal factor herein that explains at least part of thedifference. The study in Barbican was terminated in March and thatin Mona was not started until May. I think that, if I had remainedin Barbican imtil May, I would have seen an increase in the numberof small adult males as the marked juvenile males matiu'ed and thatthese either would have occupied the activity ranges that were notoccupied by adult males or woidd have wandered through the area.Alternatively, had I started my study in Mona in October, I believe Iwould then have seen a great many movements by young, subadultmales.
32 PROCEEDINGS OF THE NATIONAL MUSEUMThis study was not designed to investigate either seasonal changessuggested above or dispersal in these lizards; data, therefore, arefragmentary. Clearly both topics are important and need furtherinvestigation.6fi
./-//>
Figure 8.?Distribution of adult males in an area in Mona, May 24 to June 8 (dots= posi-tions of trees on which lizards were seen; lizards: A=68 mm; B=63 mm; C=62 mm;D^60 mm; E=60 mm; F=59 mm; G=S7 mm; H=57 mm; 1=56 mm; J=S1 mm;dates seen: l=May 24; 2=May 25; 3== May 26; 4=May 28; 5=May 29; 6=May 31;7= June 1; 8= June 3; 9= June 4; 10= June 5; 11= June 8).Homing.?Though A. lineatopus shows a great deal of site attach-ment, the only information I have on homing is negative. This isvery slight but in the absence of other data is worth recording.An adidt male was moved 200 yards and, after being caged forseveral days, was released on a small tree unoccupied by any otheradult male. He established an activity range there and during the
NO. 3595 ANOLIS LINEATOPUS?RAND 33two months under observation showed no sign of attempting to returnhome.Spatial relationships between activity ranges.?The size,shape, and location of an A. lineatopus activity range is influenced notonly by the distribution of the habitat requirements but also bythe distribution of other A. lineatopus.Even casual observations show thatx4. lineatopus are not distributedrandomly with respect to one another. This is particularly obviousalong fence rows, where, on almost every fence post, there is one largemale but very rarely two. A similar though less conspicuous relation-ship occurs in the distribution of the adidt females.The simplest situation of this sort that I studied was observed in aseries of poui trees planted along a road through the faculty housingof the University of the West Indies at Mona.The trees had been planted in a strip of grass at about 12 yardintervals, about 4 yards from a hibiscus hedge on one side and about2 yards from the paved road on the other. They ranged in size from4 to 12 inches (most of them 8 or 9 inches) in diameter and about20 to 30 feet high. The spacing of the trees was such that no lizardincluded two in his activity range. All of the females and most ofthe males probably had no other principal perches than the treesand the vegetation at their base.Dm-ing one afternoon (May 20), I examined the trees along bothsides of a kilometer of this road, looking carefully at each tree andexamining the vegetation growing at the base. I caught every A.lineatopus that I could and sexed, measured, and then released it.A total of 55 trees were examined and 72 A. lineatopus seen. Sixty-one were caught, sexed, and measured, plus five more I was able tosex confidently and estimate the size sufficiently to use the data.The four A. lineatopus that I could not size or sex accurately havebeen omitted from the following discussion as have the other A.lineatopus on the same trees with them. This reduced the dataanalyzed to 51 trees with 62 A. lineatopus.Of males, 32 were sexually mature (i.e. greater than 50 mm snout-vent length), 7 were probably not sexually mature (i.e., less than 50mm snout-vent length). Twenty-two of the 23 females were probablysexually mature (i.e., greater than 36 mm snout-vent length).Of the 51 trees, 18 had no adult A. lineatopus on them at all, 13had one, 19 had two, and 1 had three.The observed distribution departs from random distribution intwo ways: there are fewer trees with two adults of the same sexand more with two adults of opposite sex.Using binomial distribution, the expected number of trees with0, 1, and 2 lizards were calculated independently for each sex on the240-241?67 3
34 PROCEEDINGS OF THE NATIONAL MUSEUM
assumption that the presence of a lizard on a tree did not affect tlieprobability of another lizard being found on the same tree. Thesedistributions were then compared with the observed distributions(table 5) and the probability that the differences were due to chancewere calcidated, using an X^ test. These results show that there aresignificantly more trees with only a single male (at the .01 level)than would be expected if they were distributed randomly, andfewer trees with two and with no males.The observed distribution of the females differed from the expectedin the same direction (table 5) but not enough to be statisticallysignificant (at the .05 level), perhaps because of the smaller samplesize.Table 6 shows the relationship between adult males and adidtfemales. Trees with no adidts and trees with an adult of each sexoccur more frequently than M^ould be expected if the distributionof the sexes was independent, while trees with a single male or asingle female occur less frequently.Table 5.
?
Results of a count of the number of A. lineatopus on a line of trees com-paring the number of adults per tree with the expected distribution if the adultsare distributed independently of one another (the expected distribution is cal-culated from the binomial tlaeorem; the sexes are treated separately; thedistribution of the adult males differs from the expected at the .01 level, thatof the females does not)
NO. 3595 ANOLIS LINEATOPUS?RAND 35This line of poui trees was somewhat unusual in the small numberof subadults found, 7 males from 36 to 49 mm and 1 female of 29 mm.The single small female was found on a tree with an adult female andthe smallest of the young males. The largest of the j^ouug maleswas found on a tree wdth an adidt male of 62 mm and an adult femaleof 44 mm. The remaining 5 young males, 40-46 nmi, were all ontrees that had no other lizards on them.The distribution of A. lineatopus at Barbican during Septemberthrough November sho^^'s a picture similar to that described abovebut more complex.The distribution of the adult males is shown in figure 4. Theyoccupy areas that are nearly mutually exclusive though sometimes amale was seen in the activity range of another. The activity rangesof the males in part of this area are mapped in greater detail in figure 3.i^gain there is little overlap even though in the area where the malesmeet in the brush heap there are no physical barriers. In fact, all 3of these males regidarly slept in the same tree.The distribution of adult females in the yard at Barbican (fig. 6)sho^^'s a similar mutually exclusive distribution though v.ith moreoverlap than shown by the adult males. Of the 20 females mapped,all but 3 of them were within the observed activity ranges of adultmales.The distribution of the young males (fig. 7) again shows an almostmutually exclusive distribution, though small ones may overlap wdthlarger ones. The larger subadult males generally show an exclusionalso with females. Unlike the females, 7 of the young males lived inareas where there was no adult male.In contrast to the line of poui trees, the Barbican yard had morefemales than males and, whereas in the former area there were severalmales without females, in Barbican there were several females withoutmales.Figure 2 shows only a single large male living on an isolated tree,but there are 8 adult females and 3 young males also living there. Onefemale lived on the tree trunk and some of the lower branches andseldom visited the ferns on the ground. Seven of the 8 females livedin the ferns at the base of the tree, dividing the area among them sothat there was little overlap, though in two cases 2 females occupiedthe same space at different times. Many of these females also visitedthe lower tree trunk, where little spatial separation was evident, butnone of them used this as a usual perch.Of the 3 smaller males, 1 had an activity range in the ferns thatwas largely separated from that of females living in the same area.He also climbed up among the lower branches of the tree. The other2 small males had activity ranges on the branches of the tree largely
36 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122distinct from that of the first small male, from all the females, andgenerally from the large male. These 2 overlapped in space con-siderably but there is quite a difference in size between them, thelarger almost approaching adult sizeIn addition to the four areas described above, two more were studiedthat contained a number of young A. lineatopus of various sizes. Inboth of these the spatial relations were more complex than when onlyadults are considered.The first of these, Hibiscus hedge, had resident in it: 8 subadultmales of various sizes, 1 adult female, and 2 subadult females. Itwas also visited by 3 adult males. The activity ranges of these 11lizards (omitting the adult males) is given in figure 9. The first im-pression of this figure is one of utter confusion. Closer examination,however, reveals several patterns. Though the outlines of many ofthe activity ranges overlap, there is Uttle overlap between the areasin which the Uzards spent 75 percent of their time. Between lizardsof approximately the same size, little overlap occurs even at the outerlimits of the activity ranges. This exclusion of lizards of the samesize operates regardless of sex.The lizards can be separated into three size classes: the largest with1 male 49 mm long, which ranged over almost the entire area; themedium-sized Uzards, 3 males ca. 44, 42, 42 mm long, which weremutually exclusive; and the smaller lizards, 7 in number, 3 femalesof 35, 32, 30 mm, and 5 males of 40 to 30 mm, which were also mutuallyexclusive.The section of the Barbican brush heap shown in figure 5 had 1 adultmale, 4 adult females, and 7 smaller lizards (5 females and 2 males)present. Again their distribution (omitting the adult male) looksconfusing until they are sorted by size into two groups: a larger onecontaining 4 females of 43 to 34 mm and 1 male of 3 1 mm ; and a smallersized group of 6 females of 30 to 22 mm. Again, though there is con-siderable overlap between lizards of different sizes, there is httle over-lap within the size classes. Again, for this size range of lizards, sexis irrelevant. It should be noted that the size groupings used herein
Figure 9.?Distribution of females, juveniles, and young adult males in Mona hibiscushedge; this area also was visited by several larger adult males (circles= fence posts andcertain important bushes; several types of lines distinguishing different size classes oflizards based on censuses conducted April 18-23= activity ranges; stippled areas= placeswhere lizard spent 75% of its time (see pp. 28 for explanation); stippled area for no. 12overlaps those of nos. 8, 10, and 11 and has been omitted for clarity; no. 3 female=32mm, April 10; no. 4 male=36 mm, April 10; no. 5 male=42 mm, April 10; no. 7 female=35mm, April 10; no. 8 female=30 mm, April 10; no. 10 male=42 mm, April 10; no. 11male=:44 mm, April 10; no. 12 male=49 mm, April 10; no. 15 male=40 mm, April 11;no. 18 male=36 mm, April 11; no. 25 male=36 mm, April 17).
ANOLIS LINEATOPUS?RAND 37
I 11 H
38 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122bear little resemblance to those used in the first case and, in fact, anydivision seems to be arbitrary.A basic rule in distribution of A. lineatopus seems to be that no twolizards of the same size can have widely overlapping activity ranges.This general rule seems to apply at all sizes, from the smallest studied,22-24 mm, to the largest adult males, ca. 70 mm, and to apply regard-less of sex. Anolis lineatopus of different sizes, however, may havewidely overlapping activity ranges. Usually the activity ranges ofadult males overlap those of adult females, which frequently in turnoverlap those of small juveniles. The activity ranges of the smallmales frequently are like those of similar sized females, but also, andmuch more frequently than females, they are in areas that are unin-habited by otlier lizards, either male or female, and some of which,because of the structure or situation, seemed subpreferable. Thedistribution of adult A. lineatopus relative to one another resemblesone of the common patterns seen in vertebrates, that of territoriality.The distribution of the juveniles with respect to one another, particu-larly to the adults, is more novel.Social BehaviorThe regular arrangement of activity ranges with respect to oneanother is largely tlie result of direct interaction between indi\adualA. lineatopus. These interactions fall into two categories, agonisticbehavior and courtship. These interactions, and the display asso-ciated with them, are common and make up most of the social be-havior of the species. As is general in lizards, parental care, flocking,play, allogrooming, and alarm signalUng are absent. Perhapsimportant in certain circumstances but hardly true social behavioris predation by adult male A. lineatopus on juveniles of the species(see p. 15).An adult male A. lineatopus probably spends more time in displaythan in any other activity except sleeping and watching. One malethat I watched for about 11 hours gave 181 displays (averaging aboutone every 3}^ minutes). During this tune he was involved in twoconflicts with other adult males, copulated twice, and courted unsuc-cessfully several additional times. In the 8 hours from 8:00 a.m.until 4:00 p.m. there were only 5 periods longer than 10 minuteswhen no display was seen, the longest being about 30 minutes.Before describing agonistic behavior and courtship and discussingtheir effects on the spatial organization of the population, it is neces-sary to digress briefly into a description of the various displays ob-served in A. lineatopus.Display.?No analysis is given here because the displays of WestIndian Anolis, including A. lineatopus, are under study by Dr. R.
NO. 3595 ANOLIS LINEATOPUS?RAND 39Ruibal (in press). ''Display" is used here in its usual nontechnicalsense.The displays of A. lineatopvs can be described under the followingcategories: back jumping , step bobbing, bobbing, dewlapping, orien-tation, posture, jaw fencing.Back jump: The lizard suddenly and violently extends all four legsso that it pushes itself away from the perch and backward along it.A series of one to three of these may be given in sequence. I have seenit only infrequently and always in long disputes between adult males,most often shortly before they meet to lock jaws or between bouts ofjaw locking.Step bobbing: In this display the head and anterior part of thebody is raised stiffly in a series of short steps and then lowered in asimilar manner, producing a slow, jerky bob that is usually repeatedseveral times in sequence. Step bobbing is somewhat more com-mon than the back jump. It is given by juvenile and adult males(and perhaps females) usually in disputes wherein both lizards aredisplaying, occurring early as well as late in a dispute. It occursmore rarely in dispute situations wherein the opponent is not dis-playing and only very occasionally in situations wherein no opponentis visible. It can be combined with dewlapping.Bobbing: The head or head and shoulders are moved up and downrapidly several times. This is the most common A. lineatopus display.It is given in courtship by both sexes and in dispute situations bymales and females, both juveniles and adults. The displays includedherein are variable in form and, though I did not do so, it may bepossible with careful study to separate a male courtship bob fromthis category, as Greenberg and Noble (1944) did for Anolis caro-linensis and Kastle (1963) did for Norops auratus. A great deal ofbobbing by all A. lineafop>us is not obviously either dispute or coiu'tship.Similar bobbmg occurs commonly in iguanids and agamids. Interpre-tations of its significance are varied. Greenberg and Noble (1944)called it a subordination gesture in Anolis carolinensis; Noble andBradley (1933) called it an aid to vision, increasing depth perception.Carpenter (1962), in Sceloporus undulaius and other iguanids, cafls itassertion display, and Harris (1964) suggests tliat in Agama agamait functions to keep the group members aw^are of one another'sposition. Evans (1936a) suggested that in Anolis carolinensis itserved as a cue given by a female to a male to indicate her sexualreceptivity.I have recorded the behavior associated with 57 instances ofbobbing by marked females and subadults during two sessions oftwo hours each (tables 7, 8). The most striking correlation is withmovement. Anolis lineatopus usually bobbed after it had shifted its
40 PROCEEDINGS OF THE NATIONAL MUSEUMTable 7.
?
Relationship between head bohhing and shifts in position in females andyoung males
Apparent reason for shift
NO. 3595 ANOLIS LINEATOPUS?RAND 41given in coiirtsliip and agonistic situations. Like bobbing, it seemsalso to serve as an advertisement display.Orientation: Though orientation by itself is not a display, certainstereotyped changes in orientation are important components of almostall displays. In courtship a male orients toward the female, who isusually oriented away from him. He may turn his head to one sideso that his dewlap is more visible to her. In a dispute, A. lineatopusfrequently does not face its opponent directly but turns so that itpresents a side view. The amount of turning varies from turningthe head slightly to one side, through a position in which the lizardis at right angles to its opponent, to one in which the lizard is facingalmost directly away from its opponent. In disputes in which bothlizards are displaying, they frequently approach one another obliquelyrather than directly.In watching a prolonged fight between two males, it is possibleto predict, on the basis of the angle of each, which one will attacknext. Lizards that are parallel to one another will usually displayrather than attack. The closer a lizard is to facing his opponent, themore likely he will approach the other; the farther away he is facingfrom his opponent, the more likely he will retreat.Posture: Associated with lateral orientation are postural displaysthat increase the apparent size of the lizard when seen in side view.In lizards of all sizes the sides are pulled in or flattened so that theback is arched and the belly extended with the result that the bodyappears deeper. The throat is gorged : the base of the hyoid is pulleddown so that the throat appears swollen. Frequently the mouth isopened slightly and the tongue, which is short and broad, is raised andpushed forward so that the tip appears as a small ball between theends of the jaws.Finally, and in the males only and most conspicuously in adults,the tissue along the center of the neck and back can be raised intonuchal and dorsal crests. The total effect of these changes is tomake the lizard, when seen from the side, appear almost twice asbulky as it normally is. The postural displays are all associated withdisputes and do not occur in courtship.Jaw fencing: Though fighting is not properly considered display,combat between A. lineatopus adult males is usually so ritualizedthat is should be considered here. Each male approaches the otherslowly until the two are an inch or so apart, head to head. Theylunge repeatedly at each other's snout for a moment or so, as if fencingfor a hold. They then lock jaws, one biting the other's snout and thelatter biting the former's lower jaw. The two lizards then strainagainst one another, each apparently attempting to dislodge theother from the perch. When one succeeds, the other may fall to
42 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122the ground or dangle in midair for a moment. Alternately, one lizardmay attempt to escape, opening his mouth and scratching with afront foot at the other's snout. This wrestling with locked jaws maylast for several minutes and may be repeated two or three times.Usually one of these encounters is decisive and the loser retreatsimmediately after the bout. I have never seen a defeated lizard soinjured or exhausted that it was unable to run away.Damage sometimes results from these bouts and blood is drawn,though the wounds are restricted to the snout. Some males arefound with a swelling on one of the mandibles, probably wounds fromthis sort of combat that have become infected. (Schmidt, 1928,suggested the same thing for A. cristatellus in Puerto Rico.) Theinfected wounds apparently are painfid. One noosed male with aninfected jaw repeatedly bit and released my finger rather than bitingand holding as a male normally does.The preceding descriptions of display provide a foundation fordiscussing the role that social behavior plays in the spatial organizationof A. lineatopus. Two sorts of social behavior seem unportant:courtship and agonistic behavior.Courtship.?This behavior brings together male and female whenthey are ready to copulate. Evidence suggests that it also influencesmales and females to establish and maintain overlapping activityranges and in effect to form pair bonds.This pattern, a male with a home range shared by one or severalfemales that are his mates, is common among vertebrates. It isperhaps most common among bu'ds and mammals in which the maleand female share in care of the young. It seems widespread in lizards,particularly iguanids (Anolis sagrei, Evans, 1938a, Oliver, 1948;Sceloporus olivaceous, Blair, 1960; Uta stansburiana, Tinkle, et al, 1962;Basiliscus vittatus, Ilirth, 1963a; and Agama agama, Harris, 1964).The role of courtship in establishing this pattern is far from clear.Blair (1960) feels that the males of Sceloporus olivaceous seek out thefemales. In Agama agama, Harris (1964) has evidence that it is thefemales that make the choice, joining a male that has established ahome range. Hunsaker (1962) showed experimentally that in theSceloporus torqiiatus group it was the male bobbing display that wasattractive to females, and Hunsaker suggested that this influencedthem to settle near the male. Perhaps dewlap displays of the adultmale A. lineatopus are similarly attractive to females.Though the male A. lineatopus spends a considerable amount oftime chasing and bobbing and dewlapping to females, I observedcopulations infrequently. Chases, though frequent, are seldom per-sistent, and I never saw a male catch a female that was trying toavoid him.
NO. 3595 ANOLIS LINEATOPUS?RAND 43In the consummated courtships seen, there was relatively littledisplay and the whole process seemed very casual. All of the copu-lations I observed between marked lizards occurred within the homeranges of the lizards involved. For example, at Barbican (figs. 4, 6),an adult male (no. 1) was seen in copulation with three females (nos.16, 59, 145), all of whose home ranges widely overlapped his. Noneof these females was seen to copulate with any other male thoughthis may happen. Among the hzards on an isolated tree (fig. 3), onefemale was seen to copulate with the adult male and also, on anotheroccasion, with a smaller (ca. 47 mm) but obviously adult male. Theactivity range of the large male overlapped that of the female whilethe activity range of the small male overlapped that of the female tosome extent.The following description from my field notes illustrate theseconsummated courtships:29 October?Bo,rbican. No. 3, adult male, was sitting on a fence post andNo. 18, an adult female, on the adjacent fence post about 7 feet away. Theyhad been sitting in these spots for at least J-^ hour ignoring each other. The maledisplayed occasionally: bobbing first, then continuing to bob he began to flashhis dewlap, stopped bobbing, and continued flashing dewlap, then stopped; a fewbobs were given without any dewlapping. The female moved several times butI saw no display on her part. Then, with no obvious preamble No. 3 climbedalong the barbed wire from his post to hers. He stopped on the wire for a momentthen proceeded to her post, approached her from behind and straddled her withboth fore and hind legs and seized a tiny bit of the loose skin on the back of herneck and they walked together a few inches down the post. They stopped andhe twisted the base of his tail under hers and apparently inserted his left hemipenis.Shortly he released his hold on her neck. They remained stUl except for two shortseries of bobs by the female. After perhaps 2-3 minutes they separated, bothdragged their vents on the post and then the male returned along the wire to hisoriginal post.Once the male had left his original post he did not display and hisearlier displays seemed the usual advertisement displays. I had beenwatching the male but was aware of the female for about a half hourbefore the male approached her and I noticed no display behavior onher part. The only unusual thing she did was not avoid no. 3'sapproach. About an hour later these two were in copulation again.Before any of this happened, I had placed another female, tied toa string, on the male's fence post and, though she was in full view ofhim, he seemed to ignore her completely.The other 10 courtships that I saw followed this pattern almostexactly, differing only in that the male occasionally stopped in hisapproach to the female and dewlapped. In each of these I wasimpressed by the small amount of display on the part of the male andits almost total absence on the part of the female. The only move-ments by the female that could be considered display were noted in
44 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122two of the courtships; in these, just before the male seized her neck,the female moved her head from side to side.There may be an inviting or receptive display on the part of thefemale but I was unable to recognize it. Certainly the females do notbob to indicate their receptivity as Evans has recorded for female A.carolinensis.Copulations were observed from August 24 to May 30. They seemto occur throughout the year.The above description of a copulation mentions that the malecopulated with a female that was resident in his area even thoughthere was an adult female closer to him and tethered so that she couldnot escape. On another occasion, a different male approached about10 feet, apparently in response to my placing a female stake-out inhis home range. When he reached the vicmity of the stake-out andhad inspected her, he mated with a free female that was nearby andwith which I had seen him before. I tried tethering females withinthe home ranges of males on several other occasions and in no casedid a male attempt to mate with them.These observations suggest the possibility that the males recognizethe females that live within their home ranges as individuals. Theobservations can also be explained by saying that a female tied up isnot a suitable mate or that the tethered females were not receptivewhile the free-living ones were and that the male could detect this.There are two other data relevant here. One is that adult maleand adult female home ranges overlap intersexually more frequentlythan one would expect on chance (p. 33). Apparently adults of onesex or the other, and perhaps both, prefer to live in a home rangewhich overlaps that of an adult of the other sex.The final piece of evidence is the very simple and almost casualway in which copulation occurs and the small amount of display thatimmediately precedes it. A great deal of display is given by themales in the presence of the females to which the females do not reactovertly. If it has any effect on them at all, it must be a cumulativeeffect over a period of time. This is possible since all of the copula-tions that I observed involving marked lizards took place within thenormal home ranges of the lizards mvolved. The cumulative effectof the display may result in keeping the female aware of the male'spresence, in bringing her into sexual receptivity (as in many birds),or in reducing her fear of or aggression toward him. The studiesof Hunsacker (1962), showing that there are two displays in Scelop-orus, one which attracts the females to stay near the male and anotherwhich is precopulatory, suggest that the cumulative effect of displaypostulated for A. lineatopus may occur widely in iguanids.
NO. 3595 ANOLIS LINEATOPUS?RAND 45Since the lizards live together more or less permanently and thefemales usually mate with a single male (the male with the one ormore females that have home ranges within his), there exists whatmight be considered a pair bond. The evidence that the male A.lineatopus may recognize the females as individuals suggests thatthis pair bond is comparable to that in birds and mammals. A pairbond of some sort has been reported in the largely monogamous Utastansburiana, Tinkle, et al (1962), in certain populations of iS'ceZopor'j^smerriami, Milstead (1961), and Crotophytus collaris, Fitch (1956a),and in Cnemidophorus tigris, Milstead (1961). Kastle (1963) sug-gests that his captive Noi'ops auratus recognized one another asindividuals.The large amount of unconsiunmated courtship and the very smallamount of immediate precopulatory display then could be interpretedtogether as display that serves to help maintain the pair bond. Thisis a function of "territorial" display that is quite well establishedin bird literature.This discussion of the possible pah' bond in A. lineatopus is specu-lation but it suggests a very interesting line of investigation.The adult males, in addition to courting females, also courtedyoung males that were of female size. I was unable to distinguishbetween the approach of a male and his display to these young malesand that exhibited toward adult females. This contrasts with thesituation in A. carolinensis, wherein Greenberg and Noble (1944)reported that they could distinguish the approach of an adult maleto a female from his approach to any other male.In no case did I see a homosexual mating though I saw one malestart precopulatory behavior with an adult male stake-out. Homo-sexual matings are reported for caged A. carolinensis by Noble andBradley (1933). These matings probably were due to the crowdedconditions under which the Uzards were being kept.The only difference I could detect between females and youngmales in then* response to courtship is that the males always ran andthe females ran except when they were receptive. (There is somedifference in their shape, particularly when a female is carrying alarge egg.)On one occasion I saw an adult male interrupt a copulation asfoUows:18 May?Mona. This evening just about suppertime my attention wasattracted by rustlnig in the leaves. When I looked I saw a male holding a femaleby the neck; as is typical of the terminal stages of courtship, they were walkingalong slowly. Just then another adult male ran over and jumped up in a bushnearby and bobbed, then he jumped towards the pair and all three disappearedaround a flower pot. By the time I maneuvered to where I could see them
46 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122
again, the pair had separated. Clearly one male had been attracted by thecourtship of another and had successfully interrupted it.This was seen only once, probably because males by their aggressivebehavior keep other males out of the area where mating takes place.Agonistic behavior.?The information on the role of agonisticbehavior in the spatial organization of A. lineatopus populations ismore abundant and more du-ect than is that on the role of courtship.Two factors are most important in agonistic behavior: relativesize and loca,tion. First, A. lineatopus is most aggressive towardothers of similar size while much less so toward those that are eithermuch larger or much smaller. Second, A. lineatopus is most aggres-sive within its activity range and particularly on its usual perches.As a result, A. lineatopus of similar size have activity ranges withlittle overlap. Greater overlap in activity range is possible betweenindividuals with greater differences in size. Anolis lineatopus (parti-cularly females and subadults) behave aggressively toward anolesof other species just as they do toward their own species.Though the basic pattern is quite simple, the relationships amongA. lineatopus in the areas studied were complex. Because of thewide range of sizes present, each individual behaved somewhat differ-ently to each of its neighbors.Collias (1944) has said that individual aggression may be expressedas either of two types: dominance hierarchies and territoriality.This dichotomy has been widely used in studies on Uzards (C. C.Carpenter, 1962; Greenberg and Noble, 1944; Evans, 1951, etc.).The behavior of A. lineatopus can not be assigned to either of thesecategories because it has important aspects of each of them. Byeither a simple definition of territory as "a defended area" (Noble,1939) or a more complex operational definition?"an area or spacein which a particular individual dominates certain categories ofintruders who dominate it elsewhere. An individual dominatesanother when it drives it away, excludes it or supplants it at will"(Rand, in press)?every A. lineatopus holds a territory, defending itagainst neighbors of the same size. At the same time, each is amember of a straight line dominance hierarchy that consists of allthose anoles of different sizes whose home ranges overlap its ownhome range.Territorial behavior in some form or other is a common behaviorpattern in iguanids and has been reported from almost every speciesthat has been studied in detail. Adult males are always involved,and sometimes adult females as well. In a few cases, principallyBlair (1960), Fitch (1956a), and Oliver (1948), it has been reportedin juveniles.
NO. 3506 ANOLIS LINEATOPUS?RAND 47The establishment of dominance hierarchies is a common phenom-enon in caged iguanids: A. carolinensis, Evans (1936), Crotophytuscollaris, Greenberg (1945), Uro.saurus ornafus, Carpenter and Grubits(1960). Dominance hierarchies in the field also have been reportedin Ctenosaura pectinata, Evans (1951) and Sceloporus (jraimnicus,Evans (1946).Tlie brief report on the social behavior of A. sagrei given by Oliver(1948) suggests that that species is very similar to A. lineafopus inits social beha\dor.Anolis lineaiopus difTers from the usual territorial pattern in verte-brates chiefly in the uivolvement t)f a variety of sizes of juveniles thathold territories and defend them against one another and, in the caseof large suliadult males, against adult females. The laboratorystudies of Greenberg (1947) show that juvenile green sunfish mayhold territories and suggest that in tlie ^\ild they may behave as doA. lineaiopus. The basic pattern in A. lincatojms and some of tliepossible complexities can be described effectively by ])rcsentingabstracts from my field notes.In the following series of examples I have begun with interactionsbetween individuals of similar size living in adjacent activity ranges.In these first two cases, the disputes were brief. They occurred onthe mutual range boundary or at the point when one A. lineafopusentered the activity range of the other and ended when the formerleft. Either lizard invaded, always as a by-product of some otheractivity, but the resident always was both the aggressor and thewinner.(1) Hibiscus hedge, Mona [fig. 9]?Seven disputes between No. 4, a 36 mmmale and No. 7, a 35 mm female, with activity ranges which overlapped veryshghtly.13 April. No. 4 ch.ased a smaller lizard (No. 6) out of his own area and wellinto that of No. 7. No. 7 charged at No. 4 and he retreated back toward hisusual perch.18 April. Both lizards met at the boundary between the two areas anddisplayed repeatedly at one another. First No. 7 retreated into her area andNo. 4 remained where he had been displaying. After several minutes No. 7 cameback toward No. 4 and displayed. No. 4 returned the display and then retreatedto his usual perch. No. 7 did not chase him.20 April. Twice a larger lizard (No. 5) chased No. 4 who fled into the activityrange of No. 7. The first time, No. 7 charged at No. 4 and the second tiinecharged, then stopped and displayed. Each time No. 4 fled back to his ownactivity range. The second time, he stopped when he was well within it to dis-play back at No. 7. In neither case did No. 7 carry her chase beyond her usualactivity range.27 April. The two liz.'irds were s(H>n displaying to each other on tlu^ bound-ary area but a bit nearer one of No. 7's perches than to No. 4's usual perch.No. 4 retreated.
48 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122
1 May. The two lizards were again seen displaying to each other in the samegeneral area as on 27 April but a bit nearer No. 4's usual perch than last time.This time both retreated with No. 4 moving away first and then No. 7.(2) Hibiscus hedge, Mona [fig. 9]?A similar set of disputes between No. 4and No. 18; both 36 mm males. On 14 April both No. 18 and No. 4 at differenttimes were seen chasing the same smaller lizard, No. 6. Once No. 4 chased No. 6into No. IS's activity range and followed him. No. 18 charged at No. 4, who ranback into his activity range. Twice No. 18, in chasing No. 6, ran into No. 4'sarea. Each time No. 4 displayed at No. 18 who displayed back and thenretreated.In the preceding two examples the encounters were brief and nochanges in activity range boundaries resulted. These were the mostcommon type of disputes seen between similar sized individuals andseemed to serve to maintain the status quo. Occasionally disputeswere observed that produced changes in activity ranges. Thesewere longer and often renewed on successive days. Two examplesare given below; the first resulted in only a minor boundary change.(3) Barbican brush heap [fig. 3]?No. 1, 59 mm male; 6, 62 mm male withadjacent activity ranges with one small area of overlap, a stake that was a pref-erable perch on the North edge of No. I's activity range and the South edge ofNo. 6's.Two encounters illustrate the original situation, when neither malewas using the stake frequently.15 January, 0905 hours. No. 1 sitting on stake well up in view, bobbingoccasionally. 0910 lirs. No. 6 appears in the center of the brush heap and runsa foot or two towards No. 1, stops and bobs. No. 1 runs down from stake intobrush heap, No. 6 moves back a bit, but No. 1 continues towards tree.1005 Hrs. No. 6 comes over in brush heap to level of stake, apparently tochase No. 152 (a smaller non-resident adult male). No. 1 comes down from treethrough brush to about one foot south of No. 6, tail twitching, dewlaps. No. 6dewlaps, then retreats a couple of inches. No. 1 jumps towards stake, now on asmall twig west of stake. No. 6 retreats toward center of brush heap, bobbing.[The next day the situation changed:] 1540 hrs. No. 1 runs over intobrush heap, bobbing and tail twitching. He goes beyond the stake a short dis-tance [he was probably chasing a non-resident male which I did not see]. No.6 leaves his perch in the north end of the brush heap and runs toward No. 1, andbobs. No. 1 retreats toward stake and No. 6 moves a bit nearer and bobs again.1546 hrs. No. 1 jumps to stake, crest up.1548 " No. 6 dewlaps, his crests are not yet up.1549 " No. 1 bobs.1549 Yi " No. 1 dewlaps.1551 H " No. 6 bobs, then jumps east, moving at right angles to No. 1.1553 " No. 6's crests are up a little.1555 " No. 6 bobs.155 " No. 1 dewlaps; No. 6 dewlaps and back jumps, crests now full up.1608 hrs. No. 6 bobs.1609 " No. 6 turns and moves west about 6 inches, directly north of stake,side-on to No. 1.1610 hrs. No. 6 moves about 2 feet farther west; dewlaps.
NO.
50 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122No. 161 ignored all this going on within 2-3 feet of her. When No. 163retreated from her activity range she approached the fence where another smallfemale lives (No. 156). This female ignored No. 163 today.1400-1430 lu's. No. 163 made another attempt to enter her old activityrange and retreated once more before No. 148's approach.29 January, 815-830 lirs. No. 163 was back in her activity range this morningbut No. 148 soon chased her out again. No. 163 fled toward the fence againbut this morning No. 156, who had ignored her on 27 January, chased her backtowards the brush heap. No. 148 chased her back toward the fence. No. 156again chased her away. This time No. 163 moved to a small plant on bare groundmidway between the fence and the brush heap and spent the rest of the morningthere. No. 148 moved north out of No. 163's old activity range and spent themorning on the branches on the southeast edge of the big brush heap.Encounters between similar sized individuals sometimes involved aresident and another A. lineatojms that was a stranger to the area.Presumably most of these strangers are the subadult and youngadult males (p. 31). Probably most A. li7ieatopus that change theirhome range are attacked when they cross or attempt to settle in anarea that is already part of the home range of an A. lineatopus of similarsize.In some of the encounters between residents and strangers, thestranger fled quickly, as in the following two incidents.(5) 2 February?Barbican brush heap [fig. 4]. A stub-tailed adult maleappeared within the activity range of No. 159 (adult male). No. 159 approachedhim to within about 6 inches, both had crests up and both bobbed and dewlapped.The intruder ran by No. 159, leaving 159's activity range and into thatof No. 6 (adult male). No. 159 followed, actually invading the edge of No. 6'sactivity range, to display at the intruder, who ran a bit farther then climbeddown inside the brush heap and disappeared.After a few moments. No. 159 retreated into his own activity range.No. 6 had ignored this whole business from his perch on the other side ofhis territory.(6) 19 February?Barbican brush heap, 0832 hrs. An unmarked bobtailedadult male appears in the small brush heap. No. 1 is on rock and probablycannot see intruder.The bobtail hops into shade of log.0835 hrs. Bobtail hops onto log.0841'/^ " Bobtail moves to tree out of sight of No. 1.0846 " Bobtail moves into full view of No. 1 at 2 feet up on tree.0849;'2 " No. 1 begins bobbing, giving several series of step bobs.0850>^ hrs. No. 1 runs toward tree; when No. 1 is still 4 feet away, bobtailjumps to ground on the other side. No. 1 stops in brush heap, step bobs anddewlaps.0851 hrs. No. 1 jumps 6 inches further towards tree, bobtail runs 2 feetfurther away.0851K hrs. No. 1 moves onto tree. Bobtail leaves area completely, goingsouth.The intruding lizard, however, is not always defeated.(7) 6 September?Mona. No. 26 (49 mm male) came to ground [from the
NO. 3596 ANOLIS LINEATOPUS?RAND 51
tree where he usually was seen] and ran towards another tree nearby. No. 17(44 mm female) who was on the ground in his path jumped to the tree [whereshe usually was seen]. No. 26 changed direction and chased No. 17 around andup the tree. No. 17 fled and No. 26 came back down the tree to sit on No. 17'sfavorite perch.Some of the fiercest fighting that I saw occurred between a residentand an intruder that did not retreat when attaclved.(8) 29-30 August?Mona. A large male lineatopus, No. 31, was marked on26 August on a large tree. He measured 67 mm. in snout-vent length. He wasnext seen on 29 August on a tree about 15 yards away lighting with an unmarked,probably resident, male. Tlie fight included some jaw locking and No. 31 ap-parently won.The following day No. 31 and probably the same unmarked male were fightingon the same tree. "This is serious fighting: no display, just biting." Firstseen on tlie tree trunk, they soon locked jaws and fell to the ground where theyremained for a couple of minutes fencing and biting at each other, not only atjaws but at necks also. Both climbed back onto the tree trunk and the fightingcontinued, the lizards standing head to head and lunging sideways. They lockedjaws and fell again, the unmarked male catching a small plant with his hind legsand letting No. 31 swing past to crash against the trunk with an audible thud,then hang suspended from his jaws. They hung so for better than a minute andthen separated. The unmarked lizard ran up the tree and No. 31 up the onenext to it. They worked their way up the trees in installments, each male onhis move getting higher than the opponent on the adjacent tree. At each stopthey displayed, dewlapping and bobbing, until finally they worked their wayup into the crown and out of sight.On September 1 and 2 No. 31 was seen back on the tree where he was firstmarked.These accounts of disputes or aggTessive encounters between sunilarsized individuals have illustrated the forms these encounters usuallytake and the conditions under which they usually occur. It should bestressed further that some sort of dispute occurs on ahnost everyoccasion when two similar sized A. lineatojms meet. They may notfight when both individuals are outside of their activity ranges orboth are frightened. It is sometimes possible for a Hzard to remainwithin the activity range of another of the same size for some timebut only if the intruder remains out of sight of the resident. This ispossible only in areas where the habitat is structurally complex withmany hiding places. Even under these conditions I beheve thesituation is only temporary.At the other extreme from these aggressive encounters betweensimilar sized A. lineatopus are the relationships between individualsof very different sizes. These ntay have widely overlapping activityranges and show very little, if any, aggressive behavior toward oneanother.This is particularly true if one individual is an adult male and theother a female. In this case, the male chases the female in courtship.
52 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122The female usually ignores the male except for moving to get outof his way and avoiding his courtship unless she is ready to mate.There may be little aggression even if both lizards are not adult,provided they are quite different in size.(9) Hibiscus hedge?Mona [fig. 9] No. 11, 44 mm male and No. 8, 30 mmfemale, held widely overlapping activity ranges and were both seen frequentlyin sight of each other over a two month period. During this time only threeincidents were recorded which could be considered interactions and in eachof these No. 8 moved out of the way of No. 11 who happened to approach theperch on which she was sitting. A similar relationship existed between No.162 (a 40 mm female) and No. 161 (a 29 mm female) in the Barbican brush heap.Between the above-described extremes of tolerance and intoleranceare a number of cases in which a smaller lizard is tolerated by alarger one in parts of the latter's activity range, usually those areasthe larger individual visited less frequently, but the latter attackedin other areas particularly on its usual perches.(10) Barbican brush heap [fig. 6]. Nos. 145 and 162, 43 mm and 39 mmfemales, demonstrate this. No. 145 held an activity range centered on a smalltree. No. 162 was a 39 mm female who spent most of her time in the smallsection of the brush heap east of the tree, though a number of times I saw hervisit the tree itself.24 January?No. 162 jumps from east brush heap to tree where No. 145was sitting. No. 145 immediately chased No. 162, who ran up tree. No. 145did not follow. Twenty minutes later No. 162 runs back down tree and jumpsto east brush heap too quickly for No. 145 to do anything.25 January. No. 162 climbs tree; No. 145, who is in brush heap north oftree, ignores her. No. 162 climbs to 10 feet (unusually high for a female) andstays there for about 40 minutes. During this time No. 145 had moved fromnorth of the tree into the brush heap east of the tree. When No. 162 came backdown to the brush heap where No. 145 was, they bobbed at each other for about15 minutes before No. 162 retreated down into the brush heap.29 January. No. 145 comes down tree and into east brush heap. No. 162is in this brush heap about one foot from No. 145 but may not see her. No re-action, and about H hour later No. 145 returns to tree and then to north brushheap.31 January. No. 162 is hopping from east brush heap over towards tree.No. 145, who was in brush heap north of tree, comes over across ground to be-tween No. 162 and tree and bobs. No. 162 runs back to brush heap.Later No. 145 moves to east brush heap where No. 162 is; they ignore eachother.1 February. No. 162 moves from east brush heap to tree. No. 145 comesfrom north brush heap to chase her and No. 162 runs up to 9 feet in the tree.5 February. Both Nos. 162 and 145 are seen in east brush heap. No. 162bobs at No. 145 but No. 145 runs back to tree but not in response to No. 162.12 February. No. 145 chases No. 162 from tree. No. 162 runs across eastbrush heap to rock beyond it. Until today No. 162 has spent most of her timein the section of brush heap just east of the tree with very infrequent excursionsnorth and further east. Today No. 162 moved northeast and was involved in along and rather violent fight with a lizard well north of where she usually ranges.
NO. 3595 ANOLIS LINEATOPUS?RAND 5313 February. Both Nos. 162 and 145 are seen in east brush heap. Theyare about one foot apart, lateral to one another, with tliroats gorged and sidesflattened. Both bob. No. 162 moves back and forth at right angles to No. 145and bobs several times. No. 145 moves to the perch she uses most when shevisits east brush heap. No. 162 moves south away from No. 145, and then backpast her and out of this section of the brush heap entirely.Today again No. 162 spends quite a bit of time after this dispute with No. 145northeast of the section of the brush heap where it occurred and is seen chasing asmaller resident lizard on a fence post which No. 162 has previously seldom visited.15 February. No. 145 comes from the brush heap north of tree and chasesNo. 162 out of the brush heap east of tree. No. 162 flees northeast. This is thefirst time I have seen No. 145 come to this section of the brush heap expressly tochase No. 162.20 February. No. 145 comes from north of tree to the east brush heap todisplay to No. 162. No. 162 flattens sides and gorges throat but runs and thendown into brush heap.23 February, Last day of observations. No. 162 seen in small brush heap,No. 145 in brush heap north of tree.At the first of the study period, No. 145 chased No. 162 from the tree and theylargely ignored one another in the brush heap east of the tree, though sometimesbobbing at one another. About February 12-13 they began to display morevigorously to one another in the east brush heap and no. 145 was seen to come tothe brush heap to chase or display to No. 162. At about this time No. 162, whichhad concentrated her activities in the small section of brush heap, began to spendmore time north and east of it, chasing and fighting with the other lizards sheencountered. Until observations were discontinued, however, she still visitedthe east brush heap section. This coincidence suggests that the increased numberand intensity of disputes with No. 145 in the east brush heap was a factor inNo. 162's moving into new areas that No. 145 did not visit.I have described a variety of interactions that can occur betweendifferent individual lizards and under a variety of conditions. I wouldUke to present some of these accounts here and describe the relation-ships of a single lizard to the others around him.(11) No. 5 was a subadult male in the Mona hibiscus hedge [fig. 9], 42 mm longwhen first measured on 10 April and 46 mm long when last measured 8 May.Between these two dates he was seen on 24 different days and interactions recordedwith eight other lizards. He used as a principal perch a hibiscus bush and fre-quently visited a nearby fence post and the ground around both. During theobservation period no' changes in his activity range were noted.The activity range of No. 5 overlapped that of tliree smaller lizards and hisrelationship with each of these was slightly different.(11a) No. 3, a 32 mm female, lived entirely within the area visited by No. 5.She spent most of her time on the dead leaves on the ground but frequentlyvisited the hibiscus bush, which was No. 5's principal perch, throughout theperiod of observation. She usually made these visits when No. 5 was elsewhereand when he returned she sometimes fled back to the ground but more usuallyNo. 5 chased her away, not pursuing her on the ground but merely chasing herout of the hibiscus. Several times No. 5 came to the hibiscus from somewhereelse, the fence post, the ground, in order to chase No. 3 away; on other occasionswhen he returned for other reasons to find No. 3 there he would chase her.
54 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122No. 5 usually seemed to ignore No. 3 when she was on the ground, thoughonce when she ran towards the hibiscus bush he chased her away before she reachedit. I never saw him leave his perch to chase her except on this occasion. How-ever, if he happened to encounter No. 3 while he was on the ground for some otherreason, most frequently to catch an insect, he would frequently chase her for ashort distance. Sometimes she would move out of his way under such situationswithout his paying any attention, and I never saw her display to him under anycircumstances. The repeated chases from the hibiscus, and I recorded 15 of them,did not seem to deter No. 3 from returning to it.(lib) No. 5's activity range also overlapped that of No. 4, a 36 mm male, wholived primarily on the ground among the dead leaves and used a dead stick as aprincipal perch. Like No. 3, No. 4 visited No. 5's hibiscus but did so much lessfrequently. No. 5 chased No. 4 on these occasions and, more frequently than withNo. 3, chased No. 4 before he actually reached the bush. Both these differencesbetween No. 3 and No. 4 may have been because No. 4 had farther to travel toreach the hibiscus than did No. 3, but it may also have been because No. 4 waslarger and No. 5 more aggressive towards him. No. 5 also occasionally came fromthe hibiscus over to No. 4's perch and once, at least, seemed to do so in order tochase No. 4, who fled and then hid in the leaves. More frequently, when No. 5came in this direction to catch an insect he would stop on No. 4's perch before re-turning to his liibiscus. Sometimes he would encounter and chase No. 4 in theprocess, sometimes No. 4 would move out of the way without being chased, andonce No. 4 displayed at No. 5 when No. 5 neared him in the vicinity of No. 4'sperch. On this last occasion No. 5 ignored No. 4 completely. Before the end ofthe study period, No. 4 disappeared from the study area. Whether he died ormoved away I don't know and, if he moved, have no idea of the influence of No.5's chases.(lie) No. 7 was a 35 mm female who used as principal perches a corn plantand a fence post nearby. No. 5 also visited the fence post not infrequently butwas never seen on the corn plant, so that the overlap between these two was notcomplete. Relatively few encounters between these two were seen and none onthe fence post which they both used. No. 7 did not visit No. 5's hibiscus but wesaw three chases, one of them quite persistent, which seemed to start when No. 7approached the hibiscus. In only one of these could I see why. In this case,No. 7 was chasing a smaller lizard and the chasing carried her into No. 5's vicinitynear the hibiscus. In the final chase observed No. 5 was chasing a smaller lizardand his chase brought him close to No. 7. He stopped chasing the smaller lizardand chased No. 7 for a bit.(lid) The smallest lizard whom I saw No. 5 interact with was No. 6, a 29 mmmale [not shown in fig. 9 because it appeared at a later date]. During the periodwhen these two interactions occurred, No. 6 was living on the ground near thehibiscus and being chased frequently by No. 3 and particularly by Nos. 4 and 7.Once No. 6 approached No. 5's hibiscus and No. 5 ran towards him and dew-lapped; No. 6 fled immediately. The second time. No. 5 chased No. 6 when No. 4had chased No. 6 towards No. 5. The first of these encounters is curious becauseit was one of the two times when I saw No. 5 dewlap, the other was in responseto a small stake-out I introduced into his area. No. 5 was not sexually matureand consequently was not dewlapping either in disputes or in advertisement dis-play at this time and he is the smallest male I ever saw dewlap.(lie) On either side of No. 5 lived two lizards approaching him in size: No. 15,a 40 mm male who lived in an oleander bush on one side, and No. 11, a 44 mmmale, who lived in the next hibiscus on the other side. These three lizards seldomentered the activity ranges of each other and never were seen to visit each other's
NO. 3595 ANOLIS LINEATOPUS?RAND 55principal perches. Perhaps for these reasons few disputes were seen betweenthem, none with No. 11 and only two with No. 15. In one of these, No. 15 hadinvaded the dead leaves near No. 3 to catch an insect, as he sometimes did, andNo. 3 had displayed to him. No. 15 displayed back and No. 3 in turn displayed.At about this point No. 5 left his hibiscus and ran towards the fight. No. 15 fledback to his oleander but No. 3 remained still. No. 5 returned to his hibiscus.On the other occasion, both No. 15 and No. 5 had left their perches and run outonto the bare dirt beyond the dead leaves in order to catch an insect. Thisbrought them close together and they displayed at one another briefly beforeeach retreated to his own area.(llf) In addition to these smaller and slightly larger lizards, No. 5 also cameinto contact with three males larger than himself. One of these. No. 14, a 48 mmmale, had an activity range which did not overlap that of No. 5 but was near it.On one occasion No. 14, for some unknown reason, left his home range and visitedNo. 5's hibiscus, arriving when No. 5 was absent and sat there quietly. WhenNo. 5 appeared and climbed the hibiscus, he made several series of bobs at No. 14who ignored them. No. 5 did not attack No. 14 but after this few minutes ofdisplay ignored the motionless No. 14.(llg) The next large lizard with which No. 5 had contact was No. 24, a 57 mmmale. This male was seen off and on throughout the study period but as far asI could tell never established a permanent activity range during this time. Ionly saw one encounter between the two. I had tossed a grasshopper towardsNo. 5, who came to the ground for it. He made several attempts, chasing itacross the ground and biting at it. No. 24 came towards the grasshopper, scaringNo. 5 who seized it and ran back to his perch leaving No. 24 sitting on the ground.(llh) The final lizard with which I saw No. 5 interact was No. 26, a 58 mm male,who included No. 5's hibiscus in his activity range and for part of the study periodvisited it quite frequently. Once No. 5 fled at No. 26's approach when they wereboth on the ground. Twice No. 5 left his hibiscus when No. 26 climbed it. Threetimes No. 26 chased No. 5 when No. 26 was sitting in No. 5's hibiscus and No. 5climbed up into it, approaching him. In each of these cases No. 5 fled immedi-ately. In the final interaction. No. 26 was sitting in No. 5's hibiscus and No. 5was on the ground. No. 5 ran out to seize a large insect and started to carryit back to his hibiscus. On almost reaching it he apparently saw No. 26, for hechanged course and ran part way to the fence post. Before reaching it he stopped,turned and, with the grasshopper in his mouth, step-bobbed at No. 26.The largest male in the area. No. 13, a 60 mm adult, was not seen to pay anyattention to No. 5 or No. 5 to him but No. 13 seldom visited the hibiscus.The aggressive behavior that has been described has involvedso far two A. lineatopus; such encounters were the commonest seen.Aggressive encounters, however, were seen between A. lineatopusand other species of Anolis.There are a total of seven species of this genus in Jamaica, five ofthem occurring around Kjngston. These lizards all have somewhatdifferent structural or cUmatic niches (Rand, mss.) but they alloverlap to a greater or lesser extent. The commonest species afterA. lineatopus in the study areas is A. grahami, a green species thathves generally higher in the trees than does A. lineatopus.The adult males of the two species are about the same size butrather different in shape and in color. They are quite tolerant
56 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122interspecifically and frequently both species will live on the same tree.Part of the small amount of interaction is due to a difference inpreferred perches but sometimes two adult males sit quietly a footor two apart ignoring each other. Interactions are seen, usuallyno more prolonged or intense than the following.(12) 5 February?Barbican brush heap. No. 1 was on south stake. A malegrahami ran up stake, No. 1 jumped off and bobbed and dewlapped, his crestsraised. The male grahami dewlapped and then after a moment jumped off inanother direction and eventually climbed a tree.Interactions between females and subadults of the two specieswere much more frequent and pursued more vigorously than betweenthe adult males.(13) The best example of this sort of interaction involved a female grahami,No. 165, and several female lineatopus. The grahami lived on the small treein the Barbican brush heap and the female lineatopus frequently visited it. Oneof them, No. 145 {ca. 43 mm long) used the tree as one of her principal perches,as well as using the brush heap around it. The grahami repeatedly chased No.145 off of the tree over the month of observation but seldom followed her intothe brush heap and never came to the brush heap to chase her. The other twolineatopus females, Nos. 162 (39 mm long) and 161 {ca. 29 mm long) visited thetree less frequently. As has been described, the larger lineatopus female (No.145) chased the smaller ones from the tree, and the grahami did this also, oncefollowing the smallest female, No. 161, into the brush heap and chasing her aboutfor several minutes.The behavior of the grahami to the lineatopus was, as far as I could see,identical to her behavior toward other grahami except that she did not start tochase the lineatopus unless they were on the tree, but once left her tree to chasea female grahami that was approaching it and still 4 feet away.The interaction between the two species is reciprocal, as the follow-ing notes show.(14) 23 January?Barbican brush heap. No. 145 chased a small grahami,35?mm SV, making three charges at it. No. 162 at south end of log displayedto a small grahami which ran.(15) 10 April?hibiscus hedge, Mona. A small female-sized grahami camedown to the ground to drink from the rain water which had accumulated in acup-shaped dead leaf. A lineatopus of about the same size which was sitting ona stick a few inches away first bobbed and then jumped at the grahami who fled.Not only do A. grahami and A. lineatopus fight but the male ofthe latter occasionally courted the female of the former and pre-sumably the male A. grahami courts the female A. lineatopus.(16) 12 April?hibiscus hedge. No. 13 (60 mm male) chased a female sizedgrahami from tlie hibiscus next to his usual fence post to the next fence post,clearly courting it. He has also courted No. 11 (a 44 mm male lineatopus) twicethis morning.Though most observations were on reactions between A. lineatopusand A. grahami, I also saw reactions with A. opalinus and with A.garmani.
NO. 3505 ANOLIS LINEATOPUS?RAND 57(17) 17 May?south of Lloyds. In the course of making movies of displayI placed a male opalinus on a fence near another male opalinus and, incidentally,near a male lineatopus almost twice the size of the opalinus. Though tliese twolizards did not pay any attention to each other, they both reacted to the stake-out.The opalinus reacted first to approach and dewlap. The lineatopus then rantowards the two and the opalinus, apparently frightened by this, ran away.The lineatopus stopped a few inches away and dewlapped. After a few momentsthe opalinus ran back to the stake-out and both dewlapped and bit at it.(18) 12 October?reservoir, Mona. A female lineatopus displayed (bobbing)to and chased a male opalinus slightly smaller than she was. He fled witlioutado. Also saw a female garmani and a large male lineatopus side by side withinthree inches apart and in sight of each other without any indication of conflict.(19) 30 May?College Common. This afternoon I saw a male lineatopusdisplay to a male garmani considerably larger than himself. Both lizards wereon a two-inch branch of a poui tree about 12 feet above the ground, the adultmale lineatopus near the trunk and the adult male garmani well out among thesmaller branches. As I watched, the garmani began to move quite slowly alongthe branch toward the lineatopus. Wlicn it reached a point about 2 feet away,the lineatopus turned its head so that it was orienting its dewlap towards thegarmani and began to give dewlap flashes. This display was very like, if notidentical to, the initial display to another male lineatopus. The garmani stoppedand then immediately jumped from the branch to another, moving at rightangles to the lineatopus. It jumped again in the same direction and stoppedwhen about three feet from the lineatopus, then bobbed.The result of these interspecific interactions is to reduce overhip inthe activity ranges of hzards of the same size regardless of species,but least so in the cases of adult males. This suggests that thesignificance of the aggressive behavior may be different in femalesand subadult males than in the adult males. This point is discussedlater.Not only do A. lineatopus display to and chase anoles of otherspecies but also they sometimes display to objects that are not lizardsat all. Adult males, when they became familiar with me, wouldsometimes respond to my approach by bobbing and dewlapping asthey would to the approach of another male. This display wasnever carried beyond the initial reaction though, as mentioned under
"Escape Behavior" (p. 16), a male held in my hand would flattenhis sides and erect his crests, both of which are part of the aggressivedisplay pattern.The smaller A. lineatopus also display sometimes to objects thatare not anoles, as the following note shows.21 April?hibiscus hedge, Mona. A. lineatopus (40?mm female) was sittingon a two foot high bamboo stake about six inches from the top. A yellow-facedgrassquit lit on the end of the stake. The lizard gorged its throat and flattenedits sides. The bird flew away. The lizard's reaction was that typically given onthe approach of a somewhat larger lizard.Agonistic behavior: discussion.?Neither of the well-establishedconcepts, dominance hierarchy and territory, will describe the situa-
58 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122tion in A. lineatopus adequately by itself, nor is there a point inter-mediate between them that will describe it if one considers dominancehierarchy and territoriality to be end points of a continuum as Davis(1959) suggests. It would be possible to do as Greenberg and Noble(1944) did in their paper on A. carolinensis and describe certain aspectsof A. lineatopus behavior in terms of territory and others in terms ofhierarchy; even this approach is not completely satisfactory. Itseems to me less important to try to fit observed data into labeledpigeon-holes than to describe that data usefully.An approach to the description of the situation in A. lineatopus issuggested in a paper by Allee (1950), wherein he distinguished be-tween two types of hierarchies ("peck right" and "peck dominance")on the degi'ee of predictabiUty of the outcome of disputes. Extendingthis approach, one can consider agonistic behavior in terms of thekinds of information necessary to predict dispute outcomes. Thisseems to me to provide the basis of a more general system of descrip-tion and analysis than does a dichotomy between territory andhierarchy.Both territory and hierarchy can be described in these terms: ter-ritory as a system in which the outcome can be predicted on the basisof the place where the encounter occurs, and hierarchy as a system inwhich the outcome can be predicted on the basis of the results ofprevious encounters.To my knowledge, this approach has never been developed in quitethe way I am doing here, though every useful generalization about thesocial behavior of a species can be paraplirased as a prediction aboutwhat an individual wiU do under the specified conditions. Ethologists(e.g., Neil, 1964, in his study of Telapia) have used the predictive ap-proach, but they have been interested largely in predicting the nextaction of an animal during a dispute rather than in predicting theeventual outcome. The cues the ethologists use seem largely to beevidence that indicate the internal state of the animal, which in turndetermines what it will do. The cues discussed below seem to be in-volved in actually determining or affecting the internal states of thelizards.From the discussion in the preceding section, two factors suggestthemselves as bases for predictions in A. lineatopus: relative size ofanole and place of encounter.Among vertebrates generally these two characteristics, size andfamiliarity with the site, are undoubtedly among the most important.Probably the only factors likely to rival these are sex, breeding condi-tion, and, in some cases, age. Brown (1963) found that, in SteUer'sjays, where the encounter took place was the most important factor indetermining (or predicting) which of the two bu'ds would dominate.
ANOLIS LINEATOPUS?RAND 59But this could be modified somewhat by sex, state of reproductivecycle, and state of molt.In figure 10 the size of the winning A. lineatopus is plotted againstthe size of the losmg individual in each of 182 disputes. The figureshows that, m a large majority of these, 155, the larger lizard won;thus, the outcome of 85 percent of these disputes could have beenpredicted correctly on the basis of the size of the lizards alone, and,where the size difference between the lizards was more than 8 mm, alloutcomes could have been predicted. Those cases wherein thedift'erence was less than 8 mm are unpredictable on this basis; thisrange of differences form what conveniently can be called a zone ofuncertainty. (These and the following percentages of accuracy ofprediction are merely suggestive. A calculation of confidence Ihnitsis mathematically possible but, as the sample was not random, is notstatistically valid.)
"M" M 1 I I [?5M I I M I I I 1551 I I 1 M I M ^511 I 1 \70\mm S-ViOSf/?"W
FicuRE 10.?Relationship between size and outcome of aggressive disputes between Anolislineatopus (figures= number of disputes observed).
60 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Using only the place where a dispute occurs as a basis for predictingthe outcome is not as satisfactory as using relative size alone. Inpart, this is because the prediction is most accurate where only oneHzard is resident and an appreciable number of disputes has occurredbetween lizards that are either both resident or neither resident and,in part, because place influenced disputes only where the difference insize was small.Location, however, can be combined with size as basis for pre-diction. Because of the nonrandom nature of the sample, an increasein the percentage predictability is probably a less adequate measureof success in predicting than is a decrease m the zone of uncertainty(in mm of size difference)
.
Table 9 shows the effect of both size of anole and place of disputeon the outcome. It also shows that, where both were resident (table9a), the larger always won. Where neither was resident (table 9b),the outcome is doubtful when the difference between the lizards is lessthan 4 mm, which reduces the zone of uncertainty to about half forthese cases.Table 9c deals with disputes in which one lizard was resident, theother not. In these, the resident lizard always won if it was largerthan the nonresident, equal to it in size, or only 1 mm smaller. Thezone of uncertainty now includes those disputes involving a resident2 to 8 mm smaller than its nonresident opponent.Considering both relative lizard sizes and location of dispute, a basisfor prediction can be phrased as foUows: If the difference between thelizards is greater than 8 mm, the larger will win (81 of 81 disputes);if the difference is 8 mm or less and if neither or both are resident, thelarger will stiU win (26 of 30 disputes); if the difference is less than8 mm and only one is resident, the resident individual will win (62 of71 disputes). Using this set of rules, 169 of the 182 disputes (93percent) could have been predicted successfully.Obviously factors other than relative size and place of occurrenceare influencing the outcome of disputes in the zones of uncertainty.These probably include the past experience of both individuals(including previous disputes between them) and the psychological andphysiological condition of the indi^dduals at the tune of the dispute.Still, a system, such as the present, that gives a successful predictionlevel approaching 95 percent is a good description of the factorsinfluencing outcome.The nature of a dispute, i.e., the form it takes, is also stronglyinfluenced by relative size and place. The closer in size two A.lineatopus are, the more likely that a real fight with physical contactwill occur and the closer an A. lineatopus is to its usual perch, themore vigorous wiU be its displays. The most prolonged fights oc-
ANOLIS LINEATOPUS?RAND 61Table 9.
?
Effect of both relative anole size and place of occurrence on the outcome ofdisputes (numbers of disputes are plotted against difference in size of anoles;only those with a size difference of 8 mm or less are given)
Difference In
62 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Thus far, predictability of dispute outcomes has been discussedonly as a useful way of describing aggressive encounters in A. line-ato-pus and of using relative size of anole and place of occurrence asuseful information for predicting. Certain aspects of the behaviorof these lizards suggest that the latter two topics have a more funda-mental biological meaning.I have used predictability to mean that an outside observer couldforetell the outcome of a dispute. In a large majority of the aggres-sive encounters I have observed, the lizards behaved as if they them-selves were doing the same thing, i.e., predicting the outcome of thedispute. Most disputes were very brief and consisted of one anoleretreating as soon as it was threatened or attacked by another orretreating at the approach of another even though the latter showedno aggressive behavior. Disputes wherein both lizards displayed,i.e., behaved aggressively, seldom resulted in physical combat, andusually one lizard retreated without actually testing the strength ofits opponent. Even in the fiercest physical combats that I observed,prolonged fights between adult males, fighting was restricted almostentirely to the ritualized jaw wrestling. I have never seen a fightthat ended with one anole dead or even physically exhausted.I am, of course, not suggesting that these lizards are consciouslyweighing their chances of winning an encounter. The data, however,shows that they are behaving as if they had done so. Phrased dif-ferently: A dispute may not be decided by the dispute itself but bythe reactions of the individuals involved to circumstances surround-ing the dispute, and these reactions may be determined before thedispute starts. Because the lizards themselves behave as if they werepredicting the outcome, "predictability of dispute outcome" is morethan a convenient way of describing agonistic behavior; it seems tobe an important biological or psychological aspect of agonisticbehavior.Having suggested that A. lineatopus frequently behave as if theyare predicting outcomes, it is possible to ask if they are using thesame information that I have found useful. Phrased differently:Are the place where two lizards meet and their relative size importantstimuli in determining their behavior in the dispute situations? Ithink the answer is a qualified yes. Careful experiments would benecessary to demonstrate that this is true, but the field observationscertainly suggest it.That a Uzard is more likely to attack an opponent, less likely toflee, and will travel farther to attack when the opponent is in theresident's home range (and particularly when it is on the resident'susual perches) strongly suggests that place of encounter has animportant effect on the aggressive behavior of the animal.
NO- 3595 ANOLIS LINEATOPUS?RAND 63The same sort of informatiou suggests that relative size is alsoimportant, but here an additional Une of evidence is available. Inthe aggressive displays of these lizards, there are many elements thatact to increase the apparent size of the hzard (see p. 41). Theseelements are developed most highly in adult males but some of theelements, like lateral orientation and flattening, occur in all sizes andin both sexes.That selection has favored the incorporation of so many devicesfor increasing the apparent size of A. lineatopus into its aggressivedisplays suggests that it is to the advantage of a hzard not onlyto be bigger but also to appear bigger. This can only be importantif the reaction of an anole to an opponent is based in part on anestimate of its size. (In other words, this sort of display can beimportant only if apparent size is used by^the lizard to "predict"the outcome of the dispute.) Barlow (1963) has shown experimentallythat certain fish discriminate between fish of their own size and ofthose 10 percent larger or smaller. Anoles can probably do at leastas weU.In this discussion I have attempted to show that the outcomes of ahigh percentage of disputes in A. lineatopus are predictable on thebasis of the relative size of the lizard and on the place of occurrenceand that these two factors also influence the nature of a dispute andthe conditions under which it occurs. Further, I have attemptedto show that the lizards themselves behave as if they were predictingoutcomes and were using the same two factors in doing so.Adaptive Significance of Spatial DistributionBy "adaptive significance" of the spatial distribution of individualA. lineatopus in small areas I mean the possible selective advantagesthat this distribution can have for individual A. lineatopus and forthe population as a whole. This is equivalent to "function of terri-tory." C. E. Carpenter (1958) and Hinde (1956) have providedextensive lists of the functions of territory in various species of verte-brates. Some of these functions are clearly not applicable becausethey are related to parental care. The other functions, as Tinbergen(1957) has pointed out, must be divided into those that arise fromthe fact that individuals have definite activity ranges and those thatarise from the distribution of these activity ranges in relation toone another.One adaptive advantage of an activity range is that the individualcan learn thoroughly the structure of the area. Anolis lineatopusfrequently uses indirect pathways: an anole on a branch may runback up the branch and down the trunk to reach the ground, foUowinga pathway that initiaUy may take it away from its objective. The
64 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122
speed with which these pathways are selected suggests the lizardpreviously has learned the best routes. Anolis lineatopus males Uvingon trees almost invariably run up when frightened; those living onfence posts run down. In each case the anole selects without hesita-tion the escape route that takes hun most surely out of danger.Residence in an area makes it possible for an A. lineatopus to becomefamiUar with the fauna also, to learn good places to watch for insects,and to learn which of the larger animals in the area are potentiallydangerous. How important the former element is I do not know,but the changes in reaction to me as a potential predator were striking.Most lizards in the study areas were quite easy to noose the first timeI marked them. The next time they were much more difficult. Onemale became so shy that I could not catch him at all. Most maleswould allow approach within a few feet but, after several attempts tocatch this individual, he would run when I was 30 or 40 feet away.The effect of famiharity with an area has been reported in a numberof other hzards. Fitch (1956) reports a young Crotaphytus thatusually hid under a particular rock. When he removed the rock, thehzard ran to the same spot where the rock had been and seemedconfused at the absence of the rock.Residence in an area also provides an A. lineatopus with the oppor-tunity to learn to recognize other Hzards as individuals. Evidencethat they do this has been discussed. This famiharity w^ith otherindividuals over long periods provides the opportunity for pairformation and may reduce the severity if not the frequency of agonisticencounters.Beyond the advantages arising from the existence of activityranges, there are a number of advantages that arise from the distri-bution of these acti\dty ranges with respect to one another (Tin-bergen, 1957).The first of the two patterns in distribution of activity ranges, theoverlapping between adult males and adult females, has an obviousadvantage in mating. It helps to insure that there is a mate availablefor a receptive female and it enables the male to find mates withoutleaving the area with which he is familiar. It may also facilitatemating by allowing the indi\'iduals to become famihar with each otherand so produce some sort of pair bond.Interpretation of the adaptive significance of the second pattern,the minimal overlap between the activity ranges of A. lineatopus ofthe same size and the gradual increase in overlap as the difference insize between individuals becomes greater, is more difficult.There are, 1 think, thi-ee major effects of this pattern. Two areclosely interrelated and affect the whole population: the distributionof the population with respect to available food and the control of
NO- 3595 ANOLIS LINEATOPUS?RAND 65population density in favorable situations. The tliii'd, the distri-bution of the adult males with respect to potential mates, affects theadult males most du'ectly.Considering fii-st the lizard's relation to food supply, we have abeadyseen that A. lineatopus feeds largely within its home range and largelyon insects it sees from its principal perches. I have also shown thatthere is a tendency for different sized A. lineatopus to eat differentsized food. The spacing out in an area of indi^aduals of the same sizemeans that there are fewer times when two A. lineatopus attempt tocatch the same insect and consequently compete directly for food.Also, smce the lizards' activity ranges are nonoverlapping, the lizardscover the maximum area possible and a particular insect is mostlikely to land in the feeding area of one of them. The overlappingof feeding areas of A. lineatopus of dift'erent sizes means that a greatersize range of the insects within an area are potential food to the A.lineatopus living there. Thus, both the spacing out of individuals ofthe same size and the overlap of those of different sizes increases theprobability that a given insect will be eaten b}^ an ^4. lineatopus wher-ever the insect lands.From this discussion it appears quite possible that one of theimportant ecological residts of the social organization of -^4. lineatopusis in the decreasing of intraspecific competition for food and thehicreasing of the efficiency with which the available food supply isexploited.It has been suggested that territorial behavior (which is frequentlyso defined as to include the social system described herein) has an im-portant function in regulating density in favored areas. In A.lineatopus there is a lower limit set by the structm-e of the en\^ron-ment below which acti\'ity ranges cannot be compressed, i.e., thenumber of available perches. Particularly with adult males, buteven with smaller individuals, most perches are indivisible; they areoccupied by only one A. lineatopus of a particidar size. In areas wherethe only perches are scattered trees or fence posts, the social behaviorinteracts with the structure of the envu'onment to limit populationdensity. If there were more trees or fence posts, there could bemore A. lineatopus. But, if ^4. lineatopus peacefully shared perches,there could be more of these lizards in the same number of trees.In more complex structural habitats, the situation is not so clear.The general constancy of home range size in several rather differentplaces suggests that there is a size limit below Avhich home rangescannot be compressed. This in tm*n suggests that even in complexenvironments, social organization may act as a means of limitingpopulation density. There is another element in the structm-e of theen\ ironment other than available perches that may affect density and240-241?67 5
66 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122this is visibility. In veiy dense vegetation it might be possible fortwo A. lineatopus to have overlapping home ranges and seldom meet.I do not think this is of great importance, for most home ranges haveperches from which much of the area can be seen and it is on thesethat the lizards spend most of then- time. But the presence of alarge male in a brush heap for several days, even though he wasrepeatedly chased by the resident males, suggests that overlappinghome ranges are at least temporarily possible. The importance ofvisibility was emphasized for A. sagrei on Bimini by Oliver (1948),who reported that the territories became larger and that less overlapwas tolerated when a hurricane increased visibility by defoliating thehabitat.Agonistic behavior of the sort shown by A. lineatopus can regulatedensity only by forcing individuals to move away, since fights to thedeath are rare, if in fact they ever occur. Subadult, young adultmales, and young females do move considerable distances and maysettle, at least for short periods, in previously unoccupied areas(p. 29 et seq.). Further, a number of these young males were foundto be living in areas that for one reason or another seemed to be sub-preferable for the species. It seems likely that they had been forcedto move by aggressive behavaor on the part of other lizards andforced to settle in unoccupied places because the preferable oneswere already occupied.The evidence suggests that the social organization in A. lineatopustends to set an upper limit to the population density in a particularstructural environment and to force the excess lizards, particularlyyoung males and, to a lesser extent, young females, to disperse.How far these ^. li7ieatopus travel during dispersal we have no idea;consequently, it is impossible to judge the importance of social orga-nization in promoting panmixis in the population or in extending thedistribution of the species. Certainly that there is dispersal at allreduces the amount of close inbreeding. That the dispersal may takelizards into previously uninhabited areas indicates that it plays somepart in extending the distribution of the population.It has also been suggested that the spacing out of individuals in anarea may act to reduce both predation and disease. Both of these arepossible but seem unlikely in A. lineatopus. In spacing out individualsit is possible that fewer are found by predators, but A. lineatopus aremost conspicuous when they are displaying and fighting. The in-creased conspicuousness to a predator must at least partially offsetany advantages gained by over-dispersion.We know almost nothing about disease in lizards but it is possiblethat the spatial isolation that the social behavior produces may act toreduce the spread of infectious diseases. Two facts, however, argue
NO. 3595 ANOLIS LINEATOPUS?RAND 67
against its importance. Fii'st, only indi\ iduals of about the same sizeare overdispersed and diseases that are restricted to particular sizes\\ ithin the species must be uncommon. Second, although lizards maybe spatially isolated duruig the daj^, they frequently sleep in the sameplaces, so that, even if the spread of disease is reduced during the day,it would not be reduced at night.The fhial area in this discussion of adaptive significance involvesadult males. I have stressed that, while all sizes and both sexes areaggressive, the adult males are more aggressive than any other in-dividuals. The males travel greater distances to attack intrudersthan do other A. lineato'pus; they are more persistent in then- attacksand they fight more frequently and more fiercel}^ They have struc-tin-al modifications, nuchal and dorsal crests, used exclusively inagonistic behavior, that are lacking in females and young.This sort of difference in behavior is known in other iguanid lizards.In some, the males have larger home ranges from which they excludeother males even though, unlike the Anolis, the males are smaller thanthe females (Blair, 1960). In other species, only the males are aggres-sive at all (Blair, 1960). In ^i. carolinensis, males defend their homeranges most aggressively dming the weU-marked breeding season(Greenberg and Noble, 1944).Finally, in ^1. lineatoiJus adult males are less tolerant of individualsof then* own species and size than are the females and juveniles, butthey are more tolerant of males of other species.I think the general occurrence of aggressive behavior and the spacingout it produces in all sizes of A. lineato'pus can be explained by theecological advantages that have been discussed in the foregoing dis-cussion, but the greater aggressiveness of the adult males rec^uiresadditional explanation. I think the explanation lies in a function ofterritory discussed at length by Tinbergen (1957), which demonstratesthe selective advantage that is conferred on an adidt male if he caninsure himself exclusive mating rights to certain females by keepingother males away from them. If he can do this for a single female, heinsm-es that he will father at least some offspring, and the more femaleshe can keep isolated, the more offspring he wiU have and the greaterhis contribution to the gene pool of the next generation. This beingtrue, there must be a strong selection pressm'e for any mechanism thatwill insure a male exclusive mating rights to one or more females. Theaggressive behavior of adult male A. lineato'pus tliat keeps other malesout of the area in which females are permanently living is just such amechanism. I think that the selective advantage to the individualmale of having exclusive mating rights accounts for the greater aggres-siveness in the adult males of this species, while the other ecologicaladvantages to the population as a whole of spacing individuals of the
68 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122same size accounts for the existence of aggressiveness in all of theindividuals of the species.The Life of a LizardIn this section I want to describe what probably happens to a Uzardthroughout its life history in general tenns of its social relations andits spatial distribution in a densely populated area. There are manygaps in my information that I must bridge with hypotheses. Someof the imagery may be wrong but it is the best I can do to demonstrateas vividly as possible some of the gaps in our knowledge.A lizard hatches from an egg buried under a log, the edge of a rock,or in leaf litter and soon settles into a home range; how far it wandersbefore doing so we do not know. The initial activity range differsfrom those of the adults. The former is smaller, and the hatchlingavoids large perches and is not attracted by the presence of an indi-vidual of the other sex. The avoidance of large diameter perches isadaptive since these are frequently occupied by adult males that eathatchlings when the adults can catch them.These hatchlings are aggressive and soon space themselves out sothat their home ranges do not overlap, though the home range of ayoung lizard may be overlapped by those of several larger lizards.As the young lizard grows, it enlarges its home range. At the sametime, there seems to be a shift in perch preference with the result thatthe growing lizard begins to visit the usual perches of the largerindividuals.Initially, the young lizard is generally ignored by the larger neighborsand avoids them. But as it grows larger and begins to visit theirperches, they begin to chase it, both when it visits one of their perchesand when they encoimter it elsewhere. At first the young lizard tol-erates this behavior and flees the larger individual without changingits activity range. Such a situation may last for several weeks andperhaps several months. During this period the same lizard also ischasing from its own activity range others smaller than itself, and itmay expand its activity range at the expense of another by chasing itaway. The young Uzard also may be forced to move by anotherslightly larger lizard's moving into its area; I believe this action pro-duces some of the long distance shifts recorded for juveniles.Even if another juvenile does not displace it, this lizard's continuedgrowth soon brings it into more serious conflict wdth the adults whoseactivity ranges overlap its own?first, with adult females since theyare smaller than adult males. The decreasing diff'erence in size be-tween the adult female resident and the more rapidly growing juvenileseems to produce serious conflicts for two reasons. First, the female
NO- 3305 ANOLIS LINEATOPUS?RAND 69becomes less tolerant and begins to chase the young lizard, not onlyfrom her preferred perch but in other parts of the home range. Second,the young lizard begins to become aggressive towards the adult andto display back and attempt to chase her. This is probably a gradualchange but eventually the young lizard must either defeat the adultand chase her away or?much more likely since the young one is asyet smaller than the adult?be defeated and move. This changeseems to occur as the young females are approaching sexual maturitybut while the young m.ales are still immature. As either sex ap-proaches adult female size, they begin to attract the attention of theadult male, no longer as food but now as potential mates, and hebegins courtship chasing. For the maturing females this probablyhas no effect on their home range, but for the young males this maybe a factor in forcing them to move.When an^. lineatopus of tliis size moves, it may travel considerabledistances, at least 50 to 80 feet and probably more. During this timeit must continually come into conflict with other ^4. lineatopus of aboutits size, and probably it is chased repeatedly from the suitable activityranges that it visits. These A. lineatopus sometimes settle in placesthat are unoccupied because they are subpreferable, and we findyoung males living in unusual activity ranges, up in trees, on verysmall isolated perches, etc.Because there is a continual mortality among the adults, however,desirable activity ranges are always becoming vacant. If a youngmale ^. lineatopus is fortunate enough for this to happen to the femalewhose home range overlaps his own, then he may not be forced tomove at all. But probably and more frequently, he must travel andperhaps visit a number of suitable sites before he finds one from whichhe either can oust the resident or which he can occupy unchallenged.Probably another frequent occurrence is for a lizard to move into anarea that lies on the boundaries of two residents and establish himselfin an activity range that includes part of two other activity ranges.Such a course involves defeating both residents but only in parts oftheir activity range that they do not use frequently and, consequently,do not defend as vigorously as they would their preferred perches.For an adult female, this is probably her last move. If her newactivity range overlaps that of an adult male and contains the req-uisites in food, temperature, and cover, she probably wall remainthere the rest of her life, mating with the male, perhaps leaving thearea to lay her eggs if no suitable laying site is immediately available,perhaps shifting the boundaries or moving either temporarily orpermanently to an adjacent area, but making no further major moves.She will defend the area against intruders of her own size, at first
70 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122tolerating young lizards and then, as they grow?whether they are herOAvn offspring or not?gradually driving them out.The problems of a male the size of an adult female are not solvedeven Avhen he finds a suitable activity range. As he continues togrow, his activity range requirements change; he now prefers a largerperch. He has grown big enough so that no female can chase himfrom his established area, but he is also growdng big enough to challengethe adult male whose activity range may overlap his. Again tAvothings seem involved in the increasing conflict. The large male'sinterest in him changes from courtship chases to more vigorousaggressive attacks, and he himself becomes more aggressive towardthe big male. Though the big male may die or the younger maydefeat him and so remain and enlarge his area, it is more likely thatthe young male will be driven out. The younger male is now adultthough small. With the activity range requirements of an adult male,he finds that usually the best, most preferable home ranges are alreadyoccupied by larger adult males that wUl not tolerate him on any oftheir perches. Again he may have to travel considerable distances,establishing himself temporarily in undesirable sites, perhaps stayingwithin the activity range of a large adidt but avoiding him by stayinghidden, probably moving several times before he finally finds a suit-able place where he can establish ; and he, like the female, settles heremore or less permanently, making minor boundary changes or oc-casional shifts to adjacent areas or forays to nearby perches. But heis now an established resident who will court all the female-sizedlizards, mate with receptive females, eat the hatchlings, including hisown children, if he can catch them, and drive out any intrudingadult male. Daily RoutineIn previous sections I have discussed by topic various aspects ofthe ecology and behavior of A. lineatopus. To bring these aspectstogether and present a picture of the daily routine, this sectiondescribes the behavior of a single adult male that I and anotherobserver, working in shifts, kept under surveillance for one full day.We watched him from before sunrise until after sunset and lost sightof him only for two 10-minute periods, though he was out of sightfor a few seconds a number of times. During the day, wc saw himcatch and eat food t^A-ice, copulate twice, drive another male from hishome range, and dispute with, a nearby male over the common bound-ary of their home ranges.This male, no. 1, was a marked individual (60 mm S-V) that wehad kept under observation for several months. He had a homerange (fig. 3) which included a small tree about 20 feet tall that
NO. "505 ANOLIS LINEATOPUS?RAND 71branched at the ground into six trunks, each about three inches indiameter. It also included part of an adjacent brush heap and asmall log and three nearby fence posts.We started observations at 6:05 a.m., when it was just light enoughto make notes. The nocturnal frogs were still calling but the cockshad started to crow and mockingbirds were singing. It was stillcool (19? C at 6:15). I could see no anoles yet, though I knew fromprevious nights that many were asleep in the small tree, stretchedout along the twigs or leaves near the ends of branches.At 6:17 a.m. I located the first A. lineatopns, an adult male, stillin his sleeping position on top of a leaf at the end of a branch butalready awake with his eyes open. It was light now and, threeminutes later, he left his sleeping site for the inner branches of thetree. During the next 20 minutes, there was considerable movementin the tree as anoles moved about and departed for the adjacentbrush heap. The individual we had decided to observe, no. 1, ap-peared at 6:22 a.m., moving down one of the tree trunks and thenacross to the brush heap. After about 10 minutes he moved downinto the brush heap so that he was almost completely concealed bythe tangle of branches above him. He ignored the other anolesmoving about him, including a large male that came down the treeand jumped into the brush heap.By 7:00 a.m. both tree and brush heap were quiet again and noneof the A. lineatopus were sitting up on the perches where we usuallysaw them during the day. They had all left their exposed sleepingsites for more concealed spots, where they were waiting quietly.The sun rose and at 7:16 a.m. its rays struck the brush heap, butno. 1 was still in the shade. He waited only a minute and a halfbefore moving out into the sun and, three minutes later, climbed uponto the surface of the brush heap to sit in partial sunlight for a fewmoments before moving toward the tree and into full sunlight. Aboutthe same time other A. lineatopus also began to move, and the areathat had appeared empty moments before was now fuU of anolesbasking in the sun. Two large males began displaying to each otheracross the boundary of their home ranges in the brush heap. Thoughthey continued this for about 20 minutes and were less than five feetfrom no. 1, they were outside of his home range and he ignored them.By 8:00 a.m. it had warmed up considerably (24? C at 8:10 a.m.)and no. 1 was moving about frequently and displaying, both dew-lapping and bobbing. Though I saw him bob once shortly after hefirst appeared (6:23 a.m.), he did not display again for over an hourand did not begin displaying frequently until ho had been sitting inthe sun for some minutes.About 8:30 a.m. no. 1 dashed for the tree, ran up and, with pauses
72 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122for display, ran out toward the end of one of the higher branches.He dashed around among tlie foliage for a confusing few momentsand then another male jumped from the tree into the brush heap anddisappeared. No. 1 did not follow but moved to one of the largerbranches and displayed. The other male had probably slept in thetree and had been late in leaving it. No. 1 finally had seen himand chased him out. During the chase, no. 1 seemed to be trying toget above the other male and to drive him down from the tree.Now fully active, no. 1 spent most of the rest of the day in the tree(a total for the whole day of 7 hours and 55 minutes) though he visitedthe brush heap three more times (a total of 3 hours and 30 minutes)and made one sortie to a small log near the base of the tree, spending 11minutes on it and the ground nearby. One small branch of the treeseemed particularly attractive and he visited it eight times, spending3 hours and 48 minutes there. This branch was about five feet abovethe ground and from it he could overlook his entire home range; hehad used this same perch day after day. He spent another hour and57 minutes on other parts of the trunk from which this small branchgrew, 1 hour and 47 minutes on a second trunk, 10 and 7 minutesrespectively on two more trunks, and 16 minutes in the cro^\^l foliage.Two trunks he did not visit at all.The next few hours were routine. No. 1 sat on one perch for a whileand then shifted to another to sit there, moving then either to a newperch or back to the first. Though he did nothing I could interpretas searching, he seemed always to be watching his surroundings alertly.Even when on the same perch he made frequent small shifts in posi-tion, up, down, or around the perch, and with even more frequenthead and eye movements.One of the things for which no. 1 apparently was watching was food.During the day we saw him catch only two items. In both cases heran down the trunk of the tree from his perch and picked up some-thing small from the bark and swallowed it. He may have mademore captures during the few periods when he Avas out of sight but,if so, they must have been few and small because large prey is con-spicuously chewed before being swallowed.Most shifts in position were followed by some sort of display, eitherbobbing or dewlap flashing. During the day there were only eightperiods of 10 or more minutes \dthout display, the longest being about20 minutes.Most displays did not seem to be directed at anything in particularand probably served to advertise his presence to any potential in-truder or prospective mate, though the bobbing also may have afunction in improving his depth perception.Though most of the observed display did not seem to be directed at
NO. 3595 ANOLIS LINEATOPUS?RAND 73any particular lizard, several times we saw him unsuccessfully courta female, chasing her and displaying to her.No. I's acti\dty reached its highest peak in the earlier part of themorning, perhaps because he was moving frequently to take advantageof patches of sunlight and shadow to thermoregulate. During theheat of the day all the A. lineatopus were less active than they hadbeen in the morning, and they spent longer periods sitting in the sameplaces in the shade avoiding both the direct sun and the perches thathad been warmed by the sun. Later in the afternoon, they all againbecame more active but perhaps not as active as they had been in themorning.It was midafternoon before no. I's routine of waiting and AvatchingM as markedly interrupted. At 3:48 p.m. he mated with a female thatalso used the tree as one of her perches. He moved around the treetrunk and came upon her as she rested head downward. He movedup along side of her and attempted to seize the skin on the back ofher neck with the tips of his jaws; she moved her head from side toside as if avoiding tliis but did not run. Shortly after he secured thegrip and twisted his tail under hers to bring their cloacas into contact.They copulated for about three minutes and then separated. Asusual the successful courtship was casual and with little display.No. 1 returned to his usual routine until 5:00 p.m., when he movedto the brush heap to display \igorously to another large male thathad moved to a stake on the boundary between their home ranges.The resulting dispute lasted about 20 minutes; both Uzards dewlapped,oriented laterally, and bobbed at each other. Eventually bothretreated, leaving the stake unoccupied. Several days later, no. 1was able to expand his home range to include this stake but only aftera prolonged dispute with the other male. (Details of this dispute aregiven on p. 48.)Soon after this dispute the sun set at 5:44 p.m. and shortly after-ward no. 1 climbed up the tree into the higher branches. He movedout onto the smaller branches and appeared to be selecting a sleepingsite.Suddenly he came back to one of the main trunks and bobbed at asmall unmarked adult female and ran toward her. She jumped to abranch, he followed and seized her by the skin on the neck and walkedwith her a few steps. They then copulated for two and a half minutesand separated.No. 1 displayed and then moments later climbed out among thesmall branches and foHage. At 6:11 p.m. he was stretched out on avery small branch about 10 feet above the ground, apparently readyto go to sleep. It was already quite dark and by 6:13 it was too darkto see him or to make notes.
Literature CitedAllee, W. C.1950. Dominance and hierarchy in societies of vertebrates. Pp. 157-181in Grassed, P. P. (ed.), Structure et physiologic des societes ani-males. Coll. Intern. Centre Natl. Rech. Sci., no. 34.AsPREY, G. F., and Robbins, R. G.1953. The vegetation of Jamaica. Ecol. Monogr., vol. 23, pp. 359-412.Barbour, T.1910. Notes on the herpetology of Jamaica. Bull. Mus. Comp. Zool.,vol. 52, pp. 273-301.Barlow, C. W.1963. Role of size in courtship of the orange chromid (Etrophis maculatus;Pisces-Cichlidae) . American Zool., vol. 3, no. 4, p. 527.Bartholomew, G. A., and Tucker, V. A.1963. Control of changes in body temperature, metabolism, and circulationby the agamid lizard Amphibolurus barbalus. Physiol. Zool.,vol. 36, pp. 198-218.1964. Size, body temperature, thermal conductance, oxygen consumption,and heart rate in Australian varanid lizards. Physiol. Zool.,vol. 37, pp. 341-354.Bartholomew, G. A.; Tucker, V. A.; and Lee, A. K.1965. Oxygen consumption, thermal conductance, and heart rate in theAustralian skink Tiliqua scincoides. Copeia, 1965, no. 2, pp.169-173.Blair, W. F.1960. Tlie rusty lizard: A population study. Austin: Univ. of TexasPress, 185 pp.BOOERT, C. M.19.59. How reptiles regulate their body temperature. Sci. Amer., vol. 200,no. 4, pp. 10.5-120.Brown, Jerram L.1963. Aggressiveness, dominance and social organization in the Steller'sjay. The Condor, vol. 65, no. 6, pp. 460-484.Burt, W. H.1943. Territoriality and home range concepts as applied to mammals.Journ. Mammal., vol. 24, no. 3, pp. 346-352.Cade, T. J.1960. Ecology of the peregrine and gyrfalcon population in Alaska. Univ.California Publ. Zool., vol. 63, no. 3, pp. 151-290.Cagle, F. R.1948. A population of the Caroline anole. Nat. Hist. Misc., vol. 15, pp. 1-5.Carpenter, C. C1960. Aggressive behavior and social dominance in the six-lined racerunner(Cnemidophorus sexlineatus) . Anim. Behav., vol. 8, pp. 61-66.1 901. Patterns of social behavior in the desert iguana, Dipsosaiirus dorsalis.Copeia, 1961, pp. 396-405.1902. Patterns of behavior in two Oklahoma lizards. American Midi.Nat., vol. 67, pp. 132-151.74
NO. 3505 ANOLIS LESTEATOPUS?RAND 75Carpenter, C. C, and Grubits, G., Ill1960. Dominance shifts in the tree lizard (Urosaunis ornalus-lgimnidixe)
.
Southw, Nat., vol. 5, pp. 123-128.Carpenter, C. R.1958. Territoriality: A review of concepts and problems, pp. 224?250.In A. Roe and G. G. Simpson (eds.), Behavior and evolution.New Haven: Yale Univ. Press.Carr, a.1940. A contribution to the herpetology of Florida. Univ. Florida Publ.Biol. Sci. Ser., vol. 3, no. 1, pp. 1-118.COLLETTE, B. B.1961. Correlations between ecology and morphology in anoline lizardsfrom Havana, Cuba and southern Florida. Bull. Mus. Comp.Zool., vol. 125, pp. 137-162.COLLIAS, N. E.1944. Aggressive behavior among vertebrate animals. Physiol. Zool., vol.17, pp. 83-123.CowLEs, R. B., and Bogert, C. M.1936. The herpetology of the Boulder Dam Region (Nev., Ariz., Utah).Herpetologica, vol. 1, no. 2, pp. 33-42.Davis, D. E.1959. Territorial rank in starlings. Anim. Beliav., vol. 7, nos. 3, 4, pp.214-221.Dessauer, H. C,1955. Seasonal changes in the gross organ composition of the lizard Anoliscarolinensis. Journ. Exp. Zool., vol. 128, i^p. 1-12.Eisenberg, J. F.1962. Studies on the behavior of Peromyscus maniculalus gambeln andPeromyscus californicus parasiticus. Behaviour, vol. 19, pp.177-207.Etheridge, R.1959. The relationships of the anoles (Reptilia: Sauria: Iguanidae) : Aninterpretation based on skeletal morphology. Unpublished Ph.D.dissertation, Univ. of Michigan, 235 pp.Evans, L. T.1935. Winter mating and fighting behavior of Anolis carolinensis as inducedby pituitary injections. Copeia, 1935, pp. 3-6.1936a. A study of a social hierarchy in the lizard Anolis carolinensis. Journ.Genet. Psychol., vol. 48, pp. 88-111.1936b. Behavior of castrated lizards. Journ. Genet. Psychol., vol. 48,pp. 217-221.1936e. Social behavior of the normal and castrated lizard, Anolis carolinensis.Science, vol. 83, p. 104.1937. Differential effects of the ovarian hormones on the territorial reactiontime of female Anolis carolinensis. Physiol. Zool., vol. 10, pp.456-463.1938a. Cuban field studies on the territoriality of tlie lizard Anolis sagrei.Journ. Comp. Psychol., vol. 25, pp. 97-125.1938b. Courtship behavior and sexual selection of Anolis. Journ. (,'omp.Psychol., vol. 26, pp. 475-497.1946. Social behavior of the lizard Sceloporus grammicus microlepidotus
.
Anat. Rec, vol. 94, pp. 53-54.
76 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Evans, L. T.-?Continued1951. Field study of the social behavior of the black lizard Ctenosaurapedinaia. American Mus. Nov., no. 1493, pp. 1-26.1953a. New facts bearing upon the behavior of the male lizard Anoliscarolinensis. Anat. Rec, vol. 117, p. 606.1953b. Tail display in an iguanid lizard, Liocephalus carinatus coryi. Copeia,1953, pp. 50-54.1961. Structure as related to behavior in the organization of populations inreptiles, pp. 148-178. In W. F. Blair (ed.). Vertebrate speciation.Austin: Univ. of Texas Press.Evans, L. T., and Clapp, M. L.1940. The relation of thyroid extract to territorial behavior and to anoxemiain Anolis carolinensis. Journ. Comp. Psychol., vol. 29, pp. 277-281.Fitch, H. S.1954. Life history and ecology of the five-lined skink, Eumeces fasciatus.Univ. Kansas Publ. Mus. Nat. Hist., vol. 8, pp. 1-156.1956a. An ecological study of the collared lizard {Crotaphytus collaris).Univ. Kansas Publ. Mus. Nat. Hist., vol. 8, pp. 213-274.1956b. Temperature responses in free-living amphibians and reptiles ofnortheastern Kansas. Univ. Kansas Publ. Mus. Nat. Hist., vol.8, pp. 417-476.1958. Home ranges, territories, and seasonal movements of vertebratesof the Natural History Reservation. Univ. Kansas Publ. Mus.Nat. Hist., vol. 11, pp. 63-326.Grant, C.1940. The herpetology of Jamaica: II, The reptiles. Bull. Inst. JamaicaSci. Ser., vol. 1, pp. 61-148.Greenberg, B.1943. Social behavior of the western banded gecko Coleonyx variegatusBaird. Physiol. Zool., vol. 16, pp. 110-122.1945. Notes on the social behavior of the collared lizard. Copeia, 1945,pp. 225-230.1947. Some relations between territory, social hierarchy, and leadershipin the green eunfish (Lepomis cyanellus). Physiol Zool., vol. 20,pp. 267-299.Greenbero, B., and Noble, G. K.1944. Social behavior of the American chameleon (Anolis carolinensisVoigt). Physiol, Zool., vol. 17, pp. 392-439.Hardy, D. F.1962. Ecology and behavior of the six-lined racerunner Cnemidophorussexlineatus. Univ. Kansas Sci. Bull., vol. 43, pp. 3-73.Harris, Vernon A.1964. The life of the rainbow lizard. Hutchinson Tropical Monograplis.New York: Hutchinson and Co., 174 pp.Heath, J. E.1964. Reptilian thermoregulation. Evaluation of field studios. Science,vol. 146, no. 3645, pp. 784-785.Hinde, R. a.1956. The biological significance of the territories of birds. Ibis, vol.98, pp. 340-369.
NO. 3595 ANOLIS LINEATOPUS?RAND 77HiRTH, H. F.1963a. The ecology of two lizards on a tropical beach. Ecol. Monogr.,vol. 33, no. 2, pp. 83-112.1963b. Some aspects of the natural liistory of Iguana iguana on a tropicalstrand. Ecology, vol. 44, no. 3, pp. G 13-6 15.HUNSACKER, D., II1962. Ethological isolating mechanisms in the Sceloporus torquatus groupof lizards. Evolution, vol. 16, pp. 62-74.Inger, R. F.1959. Temperature responses and ecological relations of Iwo Borneanlizards. Ecology, vol. 40, pp. 127-136.Inger, R. F. and Marx, H.1961. The food of amphibians. Explor. Pare National de TUpemba, fasc.64, pp. 1-86.Kastle, W.1963. Zur Ethologie des Grasanolis {Norops aural us) (Daudin). Zeitschr.Tierpsych., vol. 20, no. 1, pp. 16-33.Kaufmann, J. H.1962. Ecology and social behavior of the coati, N^asua narica, on BarroColorado Island, Panama. Univ. California Publ. Zool., vol.60, no. 3, pp. 95-222.Kennedy, J. P.1958. Food habits of the rusty lizard, Sceloporus olivaceus. Texas Journ.Sci., vol. 8, no. 3, pp. 328-349.LiCHT, P.1961. Effects of temperature on heart rates of lizards during rest andactivity. American Zool., vol. 1, p. 155.Light, P.; Dawson, W. R.; and Shoemaker, V. H.1965. Body temperature relations of Australian lizards. [Paper read byP. Licht, June 17, 1965, at the 45th Annual Meeting of the AmericanSociety of Ichthyologists and Herpetologists, Lawrence, Kansas.]MacArthur, R. H.1958. Population ecology of some warblers of northeastern coniferousforests. Ecology, vol. 39, no. 4, pp. 599-619.Mayhew, W. W.1963. Reproduction in the granite spiny lizard, Sceloporus orcuUi. Copeia,1963, no. 1, pp. 144-152.Mertens, R.1939. Herpetologische Ergebnisse einer Reise nach der Insel Ilispaniola,Westindien. Abh. Senckenb. Naturf. Ges., vol. 449, pp. 1-84.MiLSTEAD, W. W.1957a. Some aspects of competition in natural populations of whiptail lizards(genus Cnemidophorus) . Texas Journ. Sci., vol. 9, no. 4, pp.410-447.1957b. Observations on the natural history of four species of whiptail lizard,Cnemidophorus (Sauria, Teiidae) in Trans-Pecos Texas. Southw.Nat., vol. 2, nos. 2, 3, pp. 105-121.1961a. Competitive relations in lizard populations, pp. 460-489. In W. F.Blau- (ed.), Vertebrate speciation. Austin: Univ. of Texas Press.1961b. Observations of the activities of small animals (Reptilia and Mam-malia) on a quadrat in Southwest Texas. American Midi. Nat..voL 65, no. 1, pp. 127-138.
78 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 122Neil, E. IT.1964. An analysis of color changes and social behavior of Tilapia mossam-bica. Univ. California Publ. Zool., vol. 75, no. 1, pp. 1-58.Noble, G. K.1939. The role of dominance in the life of birds. Auk, vol. 56, pp. 253-273.Noble, G. K., and Bradley, H. T.1933. The mating behavior of lizards: Its bearing on the theory of sexualselection. Ann. New York Acad. Sci., vol. 35, pp. 25-100.O'Brien, G. P.; Smith, H. K.; and Meyer, J. R.1965. An activity study of a radioisotope-tagged lizard, Sceloporus undulatushyacinihinus (Sauria; Iguanidae). Southw. Nat., vol. 10, no. 3,pp. 179-1S7.Oliver, J. A.1948. The anoline lizards of Bimini, Bahamas. American Mus. Nov., no.1383, pp. 1-36.Pearson, P. G.1955. Population ecology of the spadefoot toad, Scaphiopus h. holbrooki(Harlan). Ecol. Monogr., vol. 25, no. 3, pp. 233-267.Rand, A. L.1952. Secondary sexual characters and ecological competition. Fieldiana,Zool., vol. 34, no. 6, pp. 65-70.Rand, A. S.1962. Notes on Hispaniolan herpetology: 5, The natural history of threesympatric species of Anolis. Breviora, no. 154, pp. 1-15.1964a. Ecological distribution in anoline lizards of Puerto Rico. Ecology,vol. 45, no. 4, pp. 745-752.1964b. Inverse relationship between temperature and shyness in the lizardAnolis lineatopus. Ecology, vol. 45, no. 4, pp. 863-864.In press. The adaptive significance of territoriality in Iguanid lizards.Symposium on Lizard Ecology, Univ. Missouri Press.In press. Ecological distribution of Anolis at Kingston, Jamaica. Bull.Mus. Comp. Zool., Harvard.RUIBAL, R.1961. Thermal relations of five species of tropical lizards. Evolution, vol.15, pp. 98-111.In press. Evolution and behavior in West Indian anoles. Symposium onLizard Ecology, Univ. Missouri Press.Schmidt, K. P.1928. Ampliibians and land reptiles of Porto Rico, with a list of thosereported from the Virgin Islands. Sci. Surv. Porto Rico and theVirgin Islands, vol. 10, pp. 1-160.Stebbins, R. C, and Eakin, R. M.1958. The role of the "Third Eye" in reptilian behavior. American Mus.Nov., no. 1870, pp. 1-40.Storer, R. W.1952. Variation in the resident sharp-shinned hawks of Mexico. Condor,vol. 54, pp. 283-289.Tinbergen, N.1939. The behavior of the snow bunting in spring. Trans. Linn. Soc. NewYork, vol. 5, pp. 1-94.1951. The study of instinct. London: Oxford Univ. Press, xii + 228 pp.1957. The functions of territory. Bird Study, vol. 4, pp. 14-27.
NO. 3535 ANOLIS LINEATOPUS?RAND 79Tinkle, D. W.; McGregor, D.; and Dana, S.1962. Home range ecology of Uta stansburiana stejnegeri. Ecology, vol. 43,no. 2, pp. 223-229.Underwood, G.1951. Reptilian retinas. Nature, vol. 167, pp. 1S3-1S5.Underwood, G., and Williams, E. E.1959. The anoline lizards of Jamaica. Bull. Inst. Jamaica Sci. Ser., vol. 9,pp. 1-48.Wetmore, a.1916. Birds of Porto Rico. United States Dept. Agric. Bull. 326, pp. 1-140.Williams, E. E., and Rand, A. S.1961. Notes on Hispaniolan herpetology: 2, A review of the Anolis semili-neatus group with tiie description of Anolis cochranae, new species.Breviora, no. 135, pp. 1-11.o