Morphological, Behavioral and Electrophoretic Evidence Of

Morphological, Behavioral and Electrophoretic Evidence of Hybridization between the Lizards, Anolis grahami and Anolis lineatopus neckeri, on Jamaica Author(s): Thomas A. Jenssen Reviewed work(s): Source: Copeia, Vol. 1977, No. 2 (May 25, 1977), pp. 270-276 Published by: American Society of Ichthyologists and Herpetologists (ASIH) Stable URL: http://www.jstor.org/stable/1443908 . Accessed: 13/01/2012 15:20

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Society of Ichthyologists and Herpetologists (ASIH) is collaborating with JSTOR to digitize, preserve and extend access to Copeia.

http://www.jstor.org Morphological, Behavioral and Electrophoretic Evidence of Hybridization Between the Lizards, Anolis grahami and Anolis lineatopus neckeri, on Jamaica

THOMAS A. JENSSEN

An adult male hybrid resulting from a cross between Anolis grahami and Anolis lineatopus neckeri was discovered in Mandeville, Jamaica. Twenty-two diagnostic characters which differentiate the parental species were used to confirm and examine the results of hybridization. These included scalation, body proportion, body color, non-display behavior, display behavior and biochemical characters. Though many of the hybrid's features were intermediate to parental characteristics, some strongly reflected the characteristics of just one parent. This was par- ticularly true of the behavior in which the hybrid was quite grahami-like. However, when placed in social situations with a male of one or the other parental species, there was none of the agonistic behavior usually associated with male-male intraspecific interactions; the hybrid did not react as if it were a member of either parental species. Also included are the first descriptions of the signature displays for A. grahami and A. 1. neckeri.

ECENTLY there have been reports of hy- METHODS bridization between various lizard taxa Morphological, behavioral and electrophoretic 1969; Gorman and Atkins, 1968; Gor- (Gorman, comparisons were used to confirm and in most man et al., 1971; Gorman and 1975; Hall Yang, instances describe the degree of shared char- and Selander, 1973; Jackson, These in- 1973). acteristics between the hybrid (Harvard Uni- stances of hybridization occur within contact Museum of zones between or where versity, Comparative Zoology, speci- parapatric populations men MCZ R-129321) and its parental species. introduction by man has brought two previously Morphology.-Underwood and Williams (1959) separated species together. The present report, have described the diagnostic morphological however, describes a hybrid produced from par- characteristics which A. ents of On 8 distinguish grahami two sympatric species. July 1971, from A. 1. neckeri. I followed their lead when in I collected an adult Mandeville, Jamaica, the scalation, anatomical male were Anolis and choosing proportion whose parents grahami and coloration characters used for describ- Anolis be- body lineatopus neckeri. Hybridization the of the to its tween of the seven anoline on ing relationship hybrid parents. any species For meristic characters some was unknown. showing variability Jamaica previously in number, counts were made from five adult Though sympatric, A. 1. neckeri and A. gra- male specimens of each parental species (all hami do not necessarily have the same micro- Mandeville-collected); these data are given as distributions; A. I. neckeri is shade-loving (Un- mean ? standard error of the mean and range. derwood and Williams, 1959) and A. grahami and The relative dewlap size was determined by frequents sunny, open perches (Schoener motion of the three Schoener, 1971). However, the once-extensive selecting picture sequences forms in which the was deep-forest habitat of A. I. neckeri is now frag- dewlap fully extended, and in which the animals were oriented mented. As a consequence, Mandeville A. 1. laterally neckeri can be found in such diverse shaded to the camera in almost identical positions. For each the was shifted to still areas as under house foundations, along and in sequence, projector wide porous rock walls and in thick growths of projection on the frame depicting full dewlap bushes. The disruption of ancestral habitats has extension and projected onto graph paper markedly increased the area of interface between where the dewlap area was outlined. To ensure neckeri and grahami habitats (Ernest Williams, equal orientation and magnification, three pers. comm.). It is surprising more instances of points (the snout tip, eye and ear opening) of hybridization have not been found. an image had to coincide with those same land- 27( JENSSEN-ANOLIS HYBRID 271 marks of the previously projected animal. The acters and a closer relationship to A. 1. neckeri dewlap area was determined for each form with for seven characters. the set at and the other largest dewlap 100% reflected affinities for two sizes as of the Morphology.-The hybrid dewlap expressed proportions each of its in half of the 10 largest. parental species scalation characters (Table 1). Regarding an- Behavior.-The display behavior of the three atomical proportions, A. grahami has wider forms was quantitatively analyzed. Both pa- phalanges with fewer lamellae and a smaller rental species have multiple-display repertoires; dewlap than A. 1. neckeri. The hybrid resem- for the purpose of this report, however, only bled A. 1. neckeri by having long phalanges in the most often performed display type in each relationship to lamellae width, but also reflected species' repertoire is described. This display type A. grahami digits by having numerous lamellae is functionally analogous to the "signature" dis- (Table 1, Fig. 1). The hybrid had an inter- play type (Jenssen, 1977). Using displays filmed mediate sized dewlap, but was closer to the in field and enclosure, the A. grahami signature grahami-sized dewlap than to the neckeri-sized display analysis was made from 84 displays dewlap. Setting A. 1. neckeri dewlap area at performed by eight males. The A. 1. neckeri 100%, A. grahami dewlap is approximately 48% signature display was analyzed from 93 dis- and the hybrid dewlap 68% of the A. 1. neckeri plays given by four males. All subjects were dewlap area (Fig. 2). from Mandeville. The hybrid was collected Body coloration and pattern is quite different before its displays were recorded in the field. between the parental species. Of the six most In the enclosure it did not display readily. Only prominent differences, the hybrid possessed the seven complete displays of good quality have A. grahami condition for five of the features been analyzed. (Table 1). Unlike A. 1. neckeri, male A. grahami All behavior sequences were filmed with a can markedly change their body color. It varies Nizo S80 Super 8 camera set at an 18 frames/s from a darkened state during which males are filming speed. Captured lizards were placed in a deep chocolate brown to a colorful stage at a 1.3 m (L) X 0.5 m (W) X 1.0 m (H) enclosure which time they have a magenta tail, yellow fitted with simulated habitat for observation sides, and an aqua neck and head. When A. and filming. Because all three of the forms grahami and the hybrid were in the darkened were wary and A. I. neckeri extremely shy, a state, hourglass-shaped markings were evident blind was used in conjunction with the en- along the dorsal midline of the trunk (Table 1). closure. Analysis of the resulting films was A. 1. neckeri lacks this pattern. with a Kodak accomplished frame-by-frame the Ektagraphic MFS-8 projector using a technique Non-display behavior.-Behaviorally, hybrid described in detail and Hover was similar to A. grahami (Table 1). Anolis gra- by Jenssen (1976). hami has a of vocalizations which it Each form's signature display was divided into repertoire artificial units. Unit divisions were based on emits during social contexts and when handled In contrast, A. 1. neckeri does easily detectable features of the display. The (Milton, 1974). not sounds other than hisses. The duration of each unit was statistically described produce hybrid was found to to A. (mean, standard error of the mean and 95% squeak similarly grahami when handled confidence limits of the mean) to reflect rela- upon capture. When the two tive degree of stereotypy. Absolute amplitude initially captured, parental have but differ- of display head bobs was not measured; how- species characteristic, markedly ent behavior. Anolis shows no ever, the relative relationship of amplitude be- escape grahami defensive with a tween the various head bobs within a display is special postures; interspersed attitude are and brief reflected in the summary graphs. quiescent unpredictable episodes of vigorous struggling accompanied by of the Proteins.-Electrophoretic analysis hybrid squeaking sounds. Anolis 1. neckeri, on the was done Preston Webster by using techniques other hand, assumes a head out and described Webster and Burns up, dewlap by (1973). mouth open posture when held. Though both species will attempt to bite if the opportunity RESULTS AND DISCUSSION arises, A. I. neckeri does not exhibit the vigorous Twenty-two characters were examined on the thrashing nor does it vocalize except to hiss. three forms; of these, the hybrid showed a The hybrid possessed the A. grahami escape be- closer relationship to A. grahami for 15 char- havior and vocalizing ability. 272 COPEIA, 1977, NO. 2

TABLE1. COMPARISONSBETWEEN TWO PARENTAL SPECIES, Anolis grahami AND Anolis lineatopus neckeri, AND A HYBRID FOR SCALE, ANATOMICAL PROPORTION, BODY COLORATION AND BEHAVIOR CHARACTERISTICS.

Diagnostic Character A. grahami Hybrid A. 1. neckeri supraorbital prominence terminates as in grahami prominence continues semicircle scales just anterior to the orbits almost to nares anterior subocular 1 scale 1 scale 2 scales scale separated 0-1 2-3 - from canthral (0.8 ? 0.2) (2.2 0.2) ridge by: interparietal scale 4 scales 4 scales 6 scales separated from 3-4 5-6 supraorbital semi- (3.8 ? 0.2) (5.4 ? 0.4) circles by: size of interparietal equal in size approximately approximately scale in comparison '2 as large 1/2 as large with supraorbital disc description of granular, only anterior imbricating, elongate; all scales markedly superciliary scales scales just posterior to posterior scales elongate and canthral scales elongate smaller than in imbricating neckeri; but larger than in grahami number of loreal 5 rows 6 rows 6 rows scale rows 5-6 6 (5.2 ? 0.2) (6 - 0.0) number of subdigital 22 scales 22 scales 17 scales toe scales and lamellae 21-25 17-19 along length of 2nd (22.4 ? 0.7) (17.4 ? 0.4) & 3rd phalanges of 3rd toe on right forelimb description of scales slightly swollen, imbricate slightly swollen, swollen, imbricate of anterior margin and smooth imbricate, mostly and keeled of dewlap smooth, and a few weakly keeled description of dorsal swollen, granular, and scales small, moderately swollen, and lateral scales without keels slightly swollen small, and keeled of the trunk and keeled description of imbricate, broad, smooth, imbricate, broad, imbricate, broad, belly scales and larger than scales on distinctly keeled, distinctly keeled, and dorsum and sides and larger than larger than scales on scales on dorsum dorsum and sides and sides ratio of length of 2.5-2.7 4.0 4.1-5.1 2nd & 3rd phalanges (2.6 ? 0.03) (4.5 ? 0.18) of 3rd toe on right forelimb divided by widest subdigital lamellae on 3rd toe JENSSEN-ANOLIS HYBRID 273

TABLE 1. (Continued)

Diagnostic Character A. grahami Hybrid A. 1. neckeri relative area of 48% 68% 100% smaller dewlaps compared with largest dewlap dewlap color orange with narrow as grahami very pale yellow with border of yellow a varying sized patch of faint orange patch of orange colora- absent absent present tion under chin finely patterned spots present present absent on dorsum of head and neck dark brown hourglass- present, evident as grahami absent shaped marking along in darkened dorsal midline coloration state of trunk magenta coloration on present, vivid present, moderate absent base of tail greenish line over absent present present shoulder and along flank ability to vocalize yes yes no description of does not mouth gape or as grahami gapes mouth widely characteristic extend dewlap, intermittently with dewlap usually behavior upon thrashes vigorously from extended, remains initial capture side to side, and squeaks passive except for occasional directed effort to bite collector habitat type most open to semi-open habitat; found living in same never in direct sun- frequented appears to have a relatively microhabitat with light except when high preferred body grahami-in a direct ambient temperature temperature sun-exposed area is cool; shows a strong association with shady microhabitats descripition of a series of five "plateaued" as grahami,but a series of three signature display bobs separated by short cadence shows more continuous "sine wave" waiting periods; cadence variability bobs; cadence is is very stereotyped very stereotyped

The two parental species are separated in cooler temperatures. On the morning when I Mandeville by distinctly different climatic collected the hybrid, I was filming a male A. niches. Anolis grahami is found in hot, open grahami displaying as he made his territorial to semi-open microhabitats, while A. 1. neckeri "rounds." He was on a small water tower which is restricted to heavy shade and seems to prefer had no surrounding shade. At places on the 274 COPEIA, 1977, NO. 2

grahami HYBRID neckeri ZJlllll grahami 11,W'' ' U I I I';i HYBRID r~/, ...... n I I , , ---' neckeri , , I.... L,

Fig. 1. Subdigital lamellae of the fourth digit on the right forefoot of Anolis grahami, the hybrid and A. lineatopus neckeri. water tower the substrate was so hot that when the A. grahami stopped to display he curled his toes off the substrate and moved .?r4 ?? ;'; . up on as he I~~~~~~~~~~~~~~~~~ was finishing the display. It was on the top of - - - .1 ., this water tower that I first noticed the hybrid. On the basis of this observation, the hybrid did Fig. 2. Overlay of the heads with extended dew- not share the A. 1. neckeri avoidance of bright, laps of Anolis grahami, the hybrid and A. line- hot habitats. atoputsneckeri. The conditions under which I first noted the hybrid are noteworthy. The male A. grahami were those seen during interspecific encounters; which I was filming on the water tower began the strongest agonistic behavior observed was to move slowly and deliberately, postured and lateral presentation, slight side flattening and pulsed his dewlap. This indicated another con- dewlap pulsing. The hybrid did not react as if specific male had been seen. I then saw the it recognized males of either parental species as intruder about a meter away from the territorial a conspecific. owner on an upper crossbrace of the tower. The behavior.-The Mandeville A. animals dewlapped at each other, but besides Display grahami is a series of five several signature displays, the males never pro- signature display "plateaued" head bobs with the gressed into more agonistic display types. In waiting periods separating bobs The duration of each of the dis- addition, the A. grahami male's initial approach (Fig. 3). did not lead to a close distance "face-off." The play's nine units was very consistent as evi- hybrid neither approached nor showed marked retreat. This being extremely atypical territorial l I I I I I I Il behavior between conspecific males, I collected grahami the UNITS hybrid and only then discovered its mixed I I X 2 , 3 , 4 5 , 6 , 7 ,8 9 t origins. Though the A. grahami showed initial L- signs of reacting to the hybrid as a conspecific Al` male, his behavior waned closer HYBRID agonistic upon UNITS I , 2 , 3 , 4 t 5 6, 7 ,9i approach. - - -- Some social interactions were staged in an AM I to determine if the would de- attempt hybrid neckeri fend a territory against members of its parental UNITS ,, 2 4, ,6, species. Would the hybrid consider itself an A. grahami, A. 1. neckeri, both, or neither? After the hybrid had been placed in a 1.3 m (L) X 0.5 m X 0.7 m enclosure for a , I I I , (W) (H) day, 0 I 2 3 4 singly introduced male A. grahami and A. I. SECONDS neckeri were placed in the enclosure at different times over a two week At no time was Fig. 3. Display-action-patterngraphs of which the period. upper edge of the blackened areas represents head there extensive agonistic behavior which marks amplitude through time. Unit durations are mean intraspecific male encounters. The hybrid was values taken from 84 A. grahami, 7 hybrid and 93 A. 1. a large animal (65 mm S-V length, 6.25 g body neckeri signature displays. The stippled area and the represents a different type of head bob pattern wt.), introduced males were of about sometimes associated with A. 1. neckeri the signature same size. The only interactions evoked display. JENSSEN-ANOLIS HYBRID 275

1.0

0.8 C) 0 zZ 0 0.6 0 W) 0.4 g E *H n P H-

0.2 nn

2 3 4 5 6 7 8 9 I 2 3 4 5 6 7 8 9

UNITS Fig. 4. Descriptive statistics of unit means for 84 A. grahami (g), 7 hybrid (H) and 93 A. 1. neckeri (n) signature displays. Horizontal lines give the unit means, ends of white bars give the standard error of the mean and ends of black bars give 95% confidence limits of the means. denced by the units' narrow confidence limits but its temporal relationship to the signature for the means (Fig. 4). Amplitude, however, was display is in no way stereotyped. Qualitatively, quite variable. Head bobbing could be very the head bob pattern is quite different from subtle, with the head elevated solely by the that of A. grahami. Where A. grahami has box- neck muscles. The vertical movements of the like bobs with waiting periods between bobs, head could be exaggerated by amplitude com- A. 1. neckeri head bobs are like a sine wave with ponents produced by the forelimbs, or even at no intervening pauses (Fig. 3). times involving four-legged push-ups. The dura- As was true of A. grahami displays, the three- tions of the units, however, were not altered bob signature display pattern of A. 1. neckeri regardless of the means by which a lizard ef- was very stereotyped; this is reflected by narrow fects his signature display pattern. confidence limits of the unit means (Fig. 4). Fifty-two (62%) of the signature displays were The particular narrowness of the confidence followed by dewlap extension. The number of limits for this species' display units, however, is dewlap pulses varied from a single extension accentuated by the small values of the unit dura- and retraction to a series of five pulses; a single tions. Actually, the coefficients of variability dewlap pulse following the head bobs was most for mean unit durations of both parental spe- common (40% of displays with pulses) and a cies were of a similar magnitude, generally five pulse series the least common (5% of dis- ranging between 14 and 24%. plays with pulses). Duration of dewlap pulses The A. I. neckeri signature display frequently was extremely variable; however, the last pulse was performed in pairs. About a third of the of a series was of longer duration than the pre- displays were repeated after an interdisplay ceeding ones. If dewlapping followed the head pause of 0.32 ? 0.02 sec (range 0.28 - 0.44 sec). bobbing pattern, initiation of dewlap appear- When the displays were given in pairs, the ance was stereotyped, averaging 0.38 + 0.05 sec second display was shorter in duration than after the ninth unit; the actual dewlapping the first (Wilcoxon matched-pairs signed-ranks sequence, however, showed no consistency. test, P = 0.0003); they averaged 1.08 + 0.03 and + The A. 1. neckeri signature display does not 0.93 0.03 sec, respectively. incorporate the use of the dewlap; dewlapping The three-bob signature display was often may infrequently follow the head bob sequence, accompanied by fast, double bobs of low am- 276 COPEIA, 1977, NO. 2

plitude (Fig. 3). Observed casually, this kind of trophoretically the animal is indisputably a bob pattern very much resembled courtship grahami-neckeri hybrid." bobs seen in other iguanid species. However, no females were present in the enclosure or in ACKNOWLEDGMENTS close proximity in the field when the filmed A NSF to E. E. Williams behavior was performed. These rapid head bobs grant (B 019801X) to which usually appeared before the signature display, supported my trip Jamaica during time the was found. Dr. and Mrs. but they could also follow the signature display. hybrid J. Howard Frank afforded me their and To study the relationship between the rapid hospitality use of their Mandeville as a head bobs and the signature display, 36 filmed property study area in signature display sequences were analyzed which during my stay Jamaica. Supplementary data were under a NIMH recorded the lizard relocating itself before stop- display gathered and NSF ping to display. All antecedent behavior to the grant (6F02M4 33435-01A1) grant to T. A. signature display was thus recorded. Half of (BMS 74-20143) Jenssen. Under other Preston Webster these displays were not preceded by the rapid circumstances, would have been a collaborator on this double head bobs, 22% were preceded by the paper. The absence of his work here affirms our loss. one double bob pattern, 17% were preceded by two double bob patterns and 11% were preceded by three quick successions of the double bob LITERATURE CITED pattern. The social significance of this rapid head bob was not determined. GORMAN,G. C. 1969. Intermediate territorial dis- pattern of a Anolis lizard The did not play hybrid (Sauria: Iguanidae). hybrid signature display pattern Zeit. Tierpsychol. 26:390-393. show any compromise between the parental sig- , AND L. ATKINS.1968. Natural hybridization nature displays (Fig. 3). This is in contrast to between two sibling species of Anolis lizards: Gorman's observation that the chromosome cytology. Science 159:1358-1360. (1969) display ,P. H. C. DESSAUERAND 0. Boos. of an Anolis aeneus X Anolis trinitatis LICHT, J. hybrid 1971. Reproductive failure among the hybridizing was intermediate to its parental species' displays. Anolis lizards of Trinidad. Syst. Zool. 20:1-18. The A. grahami X A. 1. neckeri hybrid's head , ANDS. Y. YANG. 1975. A low level of back- bob characteristics definitely resembled those of crossing between the hybridizing Anolis lizards A. of the unit of Trinidad. Herpetologica 31:196-198. grahami. Comparisons hybrid's W. AND R. K. SELANDER. 1973. durations with those of the HALL, P., Hybridiza- grahami display, tion of karyotypically differentiated populations however, were significantly different (P < 0.05) in the Sceloporusgrammicus complex (Iguanidae). for all but one of the units (Fig. 4). The mean Evolution 27:226-242. duration of the total JACKSON,J. F. 1973. The phenetics and ecology of hybrid's display (3.28 sec) a narrow zone. Evolution 27:58-68. was also different from that of the hybrid considerably JENSSEN, T. A. 1977. Display diversity in anoline A. grahami signature display (4.66 sec). The lizards and problems of interpretation. In: N. wide confidence limits for the hybrid's mean Greenberg and P. MacLean [ed.]. Behavior and unit durations show a weaker of stereo- neurology of lizards: An interdisciplinary con- degree ference. In Press. than is characteristic of lizards N.I.M.H., Rockville, Maryland. typy displays by ,AND E. L. HOVER. 1976. Display analysis of of the parental species. This is reflected in the the signature display of Anolis limifrons (Sauria: large coefficients of variability values for some Iguanidae). Behaviour 57:227-240. of the unit durations; MILTON,T. H. 1974. Structure and social signifi- hybrid's they ranged cance of vocalization Anolis from 13-60%. by grahami grahami (Sauria: Iguanidae). M.S. Thesis, Virginia, Poly- Proteins.-Preston Webster's death in the technic Institute and State University. tragic SCHOENER,T. W., AND A. SCHOENER.1971. Struc- fall of 1975 cut short the contributions of a tural habitats of West Indian Anolis lizards. I. generous and competent person. Webster was Jamaican lowlands. Breviora 368:1-53. conducting an extensive electrophoretic anal- UNDERWOOD,G., AND E. E. WILLIAMS. 1959. The of the anoles when he died. anoline lizards of Jamaica. Bull. Inst. Jamaica ysis Jamaican Sci. Ser. However, he had the No. 9:1-48. analyzed hybrid specimen WEBSTER, T. P., AND M. BURNS. 1973. and of Anolis J. Dewlap representative samples Jamaican color variation and electrophoretically detected to confirm the hybrid's parental species. The sibling species in a Haitian lizard, Anolis brevi- following is quoted from a letter to me from rostris. Evolution 27:368-377. Preston Webster. "Of the grahami-neckeri hybrid: The chromosomes suggest a hybrid origin DEPARTMENT OF BIOLOGY, VIRGINIA POLYTECHNIC by their diversity in morphology, and there is INSTITUTE AND STATE UNIVERSITY, BLACKSBURG, considerable non-disjunction in meiosis. Elec- VIRGINIA 24061. Accepted 30 March 1976.