Gould (1957) reported that 22 of 43 _T. carolina_ moved in a homeward direction when they were released in open fields up to 5.8 miles from their original points of capture. Turtles oriented themselves by the sun; homeward headings were inaccurate or lacking on overcast days and, light reflected from a mirror caused turtles to alter their courses. Seven of ten turtles released more than 150 miles from home headed in directions that corresponded most nearly to the headings last taken (at release-points near home base) and did not necessarily correspond to the direction of home. Gould's studies point out that box turtles perhaps practice a kind of "solar navigation." His work raises the question of whether the movements of box turtles are guided by the sighting of local landmarks or whether such landmarks alter the course of movement only when acting as barriers.

In the present study two experiments were made to determine the homing ability of _T. ornata_. An adult male, taken from his normal home range in the house pond area and released 1200 feet away in the southern ravine area, traveled a generally northward course (not northeastward in the direction of home) for five days, moving a distance of approximately 1900 feet. His detached trailer was recovered several days later 740 feet southeast of the last known point in his travels (a distance that could have been covered in two days) and 150 feet from the point of original capture; he had returned to his home range by a circuitous route in a period of approximately seven days. Another adult male, captured in the southern ravine area, and released in the house pond area 1900 feet away, traveled on a course that bore approximately 25 degrees north of true homeward direction; after five days he was approximately 600 feet north of the original capture point. He then began a northeastward course that took him back to the house pond area where he remained for several days; no further data are available for this individual. It is significant that the homing males discussed above traveled greater average distances per day (based on records for nine days of trailing) than any of the other turtles studied (Fig. 27). Fitch (1958:101) released an individual one half mile from where he captured it and, one year later, recovered the turtle near the point of release.

Social Relationships

Ornate box turtles are solitary except during periods of mating.

Meetings with other individuals in the course of foraging, basking, or seeking shelter, are fortuitous and have no social significance. A broad overlapping of home ranges of both s.e.xes at the Damm Farm suggests that box turtles do not intimidate other individuals in the home range or exclude them from it. No instances of fighting were observed.

Allard (1935:336), Perm and Pottharst (1940:26), and Latham (1917) recorded instances of fights between individuals of _T. carolina_; in the latter two instances fights were between males. Stickel (1950:362) observed an incident between two males that may have been a fight; however, she was of the opinion that fights rarely occur in nature and that box turtles do not defend territories. Evans (1954:23-25) considered the behavior of _T. carolina_ reported by Perm and Pottharst (_loc. cit._) to represent "territoriality." He found "... a true hierarchy...." existing between four captive males of _T.

carolina_ and another between three captive females of the same species; young individuals in the group raised their social level in the hierarchy after receiving experimental doses of male hormone.

Evans (_op. cit._:25) pointed out that true tortoises (family Testudinidae) have a more complex pattern of social behavior than do emyid turtles.

Observations made with binoculars from the vantage point of a blind provide the only information that I have concerning the reactions of box turtles to one another under natural conditions. Turtles foraging in a bare area were not startled by the approach of other turtles, and turtles moving across the area seemed to take no notice of turtles already there, regardless of whether these turtles were moving or not.

Adults and subadults behaved in approximately the same manner.

Individuals traveling or foraging in rough terrain or in gra.s.sy areas probably are unable to see each other even when they are close to one another. Conversely, box turtles can see each other and are surely aware of each other's presence in bare, flat areas. These facts suggest that no social hierarchy exists in _T. ornata_. On one occasion an adult male and a juvenile (hatched the previous autumn) were found foraging next to one another on the same pile of cow dung.

When an individual became motionless in an att.i.tude of wariness after having detected me in my blind, its behavior evoked no response on the part of other turtles, a few feet away.

INJURIES

Fire, freezing, molestation by predators, and trampling by cattle or native ungulates are only a few natural sources of injury to which box turtles have always been exposed. Man's civilization in the Great Plains, chiefly his automobile and other machines, have compounded the total of environmental hazards. Automobiles now const.i.tute a major cause of death and serious injury to box turtles. Each year thousands are struck on Kansas highways alone, not to mention the many casualties resulting from mowing machines, combines, and other farm machinery.

Although gra.s.s fires usually occur in early spring or late fall when box turtles are underground, some turtles are surely killed by fires and many are injured. In early April of 1955 the pasture at the Damm Farm was burned. Similar burnings, I discovered, had occurred both intentionally and accidently in past years at irregular intervals. No deaths or injuries, attributable to fire were discovered in the course of intensive field work in the spring and summer of 1955, when the new gra.s.s was short and conditions for finding and marking box turtles were ideal. Badly burned individuals, if any, may have secreted themselves until their wounds had healed. In June, 1957, an adult female, that had been burned severely, was taken from a small puddle in a ravine on the Damm Farm. The soft parts of her body, excepting her head and neck, were a nearly solid ma.s.s of smooth scar tissue, the scales and rugosities of the skin being practically obliterated. The tail was reduced to a mere k.n.o.b surrounding the a.n.u.s and dead, exposed bone was visible on most of the dorsal part of the carapace. Possibly this female was burned in the fire of 1955. Lack of injury to the head and neck can probably be accounted for by the additional protection afforded these parts by the folded forelegs when the turtle was withdrawn in the sh.e.l.l.

Turtles that are smashed flat on the highway, of course, have no chance of survival. Highway fatalities are usually the result either of "direct hits," where the tire of a vehicle pa.s.ses directly over the turtle, or of repeated pummeling by subsequently pa.s.sing vehicles. The writer, while driving behind other cars that struck turtles or by sitting beside roads, has observed numerous turtle casualties. Most are struck a glancing blow by a tire and are propelled some distance through the air or on the surface of the pavement, often to the side of the road. Such a blow is usually sufficient to crack or chip the sh.e.l.l, or at least to scuff away parts of the epidermal covering.

Turtles, so injured, usually survive.

Parts of the sh.e.l.l do not break away easily, even when several deep cracks are present, and only a little bleeding occurs. A common injury inflicted on the highway is the wrenching and subsequent dislocation of the carapaco-plastral articulation. In such instances the ligamentous tissue joining the two parts is torn extensively. Under these circ.u.mstances the movable sh.e.l.l parts seem to act as a safety device, giving way under pressure that would crack the sh.e.l.l of a turtle with rigid, fixed b.u.t.tresses. Dislocations of the carapaco-plastral articulation that have healed are characterized by abnormally heavy development of ligamentous tissue, which may elaborate a h.o.r.n.y, scutelike substance on its outer surface.

The extent to which serious injury incapacitates a turtle is not known. Surely open wounds are susceptible to infection and to various kinds of secondary injury; normal activity is probably interrupted by a period of quiescence, at least in the period of initial healing.

An injured female had a hole, slightly more than one inch in diameter, in the right side of the carapace at the level of the second lateral lamina. A tight, thin membrane stretched between the broken edges of the opening; this membrane contained no bone and was covered externally by scar tissue. It was obvious that this turtle had recovered, at least in part, from a serious injury (inflicted probably by a piece of heavy farm machinery).

Minor chips, scratches, and abrasions on the sh.e.l.l result from a variety of sources, some of them mentioned above. Small rounded pits in the bony sh.e.l.l (sh.e.l.l pitting) due to causes other than mechanical injury, are found in nearly all kinds of turtles according to Carpenter (1956), Hunt (1957), and my personal observation. In _T.

ornata_, however, the condition is less common than in the specimens of _T. carolina_ described by Carpenter and in the remaining species of _Terrapene_ that I have examined.

Carpenter (1956:86) came to no conclusion as to the cause of sh.e.l.l pitting in _Terrapene carolina_ but suggested that a variety of factors including parasitic fungi, parasitic invertebrates, and simple sh.e.l.l erosion, might be responsible.

According to my own observations on turtles in the University of Kansas collections, sh.e.l.l pits range in size and shape from shallow, barely discernible depressions to deep borings; I suspect that sh.e.l.l pitting for turtles in general has many causes, some of which may be of more frequent occurrence in one species than in another.

Hunt (1957:20) presumably was referring to sh.e.l.l pitting by a more suitable name when he wrote of, "... necrosis ... of mycotic origin."

Hunt (_loc. cit._) stated that "Of those cases which have been recently examined, the author found all were due to the invasion of Mucorales beneath the plates of the epidermal laminae. This disease is of extremely common occurrence and has been found in all members of the order but is seldom found in marine species. Mycosis more frequently occurs on the plastron than on the carapace." Hunt presented no evidence to support his statement regarding invasion of the sh.e.l.l by Mucorales.

Evidence that injury to the soft parts of the body is also fairly common is seen in the many _T. ornata_ with missing feet and legs.

Stumps resulting from amputations are covered with tough, calloused skin and sometimes by h.o.r.n.y tissue similar to that of the antebrachial scales. Amputees are incapacitated only slightly in normal locomotion if a functional stump remains; probably a cripple is somewhat handicapped in other functions, such as burrowing, nest digging (females), and copulation (males). Causes of amputation are discussed in the section on predators.

Fractures of the limb bones are common. A female from Stafford County, Kansas (Pl. 29, Fig. 4), showed a typical case of fracture and subsequent repair; the right fibula had been broken and the ends dislocated; a great ma.s.s of bone joined the repaired break to the middle of the tibia, giving the entire skeleton of the leg the appearance of the letter "H." The fibula, shortened by the dislocation, no longer articulated by its proximal end with the femur; the tibia probably bore the entire load in the period of repair and the transverse connection that formed between the bones later took over the function of the fibula.

There is little doubt that ornate box turtles are stepped on or trampled by cattle, at least occasionally, but I never observed such an incident; the predilection of ornate turtles for dung insects and for moving along cattle pathways brings them to close quarters with cattle and probably did likewise with native ungulates. A steer, stepping on a box turtle, could inflict superficial damage to the sh.e.l.l or cause broken limbs but would probably not crush the turtle unless on a hard substrate.

REPAIR OF INJURIES TO THE Sh.e.l.l

Most adults and a few juveniles examined in the field and laboratory had one or more small injuries on the carapace that had healed or were undergoing repair. Such injuries almost never occurred on the plastron. In an injury that was undergoing repair, a small piece of smooth, whitened bone was exposed where a piece of epidermis was missing from the sh.e.l.l. One or more edges of the exposed bone characteristically projected over the surrounding epidermis, making the bone appear as though it had been driven forcefully, like a splinter, into the sh.e.l.l (Pl. 29, Figs. 1 and 2). Because of their curious appearance, small areas of repair were referred to in my notes as "splinter scars." The position and number of splinter scars were often recorded as supplementary means of individualizing turtles in the field.

Splinter scars result from minor abrasions that damage a few square millimeters of the sh.e.l.l. Larger areas of exposed bone were noted in only a few specimens. Two turtles at the Damm Farm had bone exposed on more than one-half the surface area of the carapace; both of these turtles were probably burned in the gra.s.s fire of 1955. Ordinarily, a break in the sh.e.l.l does not induce extensive regeneration of tissues; when sh.e.l.ls are damaged by crushing or cracking, regeneration of epidermis and bone occurs only along the lines of fracture, unless the broken parts have been dislocated. Ligamentous tissue develops in some breaks on the plastron, the broken area remaining slightly movable after healing is completed (Pl. 24).

Dissection of injured sh.e.l.ls revealed the mode of sh.e.l.l regeneration to be the same whether a large or small portion of the sh.e.l.l had been damaged. An abrasion may gouge out a small portion of the sh.e.l.l; burning, freezing, or concussion may kill a portion of the epidermis and a corresponding portion of bone beneath it without actually disfiguring the sh.e.l.l. Dead bone and epidermis become loosened at the margin of the wound. The epidermis sloughs off soon afterward but the bone adheres to the wound. New epidermis and new bone, growing from undamaged tissues at the edges of the wound, encroach on the wound beneath the layer of dead bone. The piece of dead bone is thereby gradually isolated from the rest of the sh.e.l.l and is sloughed off when healing is complete. The dead bone may come off in one piece or slough off gradually at its edges as healing proceeds toward the center of the wound. The layer of dead bone protects the wound during the process of regeneration (Pl. 30). Areas of exposed bone become white and shiny, nearly enamellike in appearance, as a result of wear on the sh.e.l.l.

The above conclusions, in regard to _T. ornata_, agree basically with the findings of Woodbury and Hardy (1948:161-162) and Miller (1955:116) on regeneration of the sh.e.l.l in desert tortoises (_Gopherus aga.s.sizi_). Danini (1946:592-4, English summary) made histological studies on regeneration of the sh.e.l.l in specimens of _Emys...o...b..cularis_; he found that new bone trabeculae formed on the surfaces of undamaged trabeculae at the edge of the wound and formed also in connective tissue at the center of the wound. Regeneration of bone was incomplete in some instances where total extirpation of a portion of the sh.e.l.l had occurred. Regenerated epidermis was usually thicker than the original scute.

Exposed bone on the sh.e.l.ls of turtles that have been injured in fires, although dead, is unmarked and shows no evidence of being burned.

Exposure to fire kills the growing portions of both the epidermis and the bone but seemingly does not actually char or disfigure the bone (although the epidermis may be so affected) (Pl. 29, Fig. 3). Injuries from fire result probably from brief encounter with the fire itself or from more prolonged contact with some surface heated by the fire. A turtle that remained in a fire long enough to have its sh.e.l.l charred would presumably have little chance of survival. Grossly disfigured sh.e.l.ls therefore do not result directly from burns but are due to the gnarled texture of the regenerated bone and epidermis remaining after the dead portions of the sh.e.l.l have been sloughed off. Information on injuries from fire was supplemented by examination of several badly burned specimens of _T. carolina_. Their sh.e.l.ls were nearly covered with exposed bone and regenerated epidermis. One specimen was so badly damaged that the entire anterior rim of its carapace was loose and could be pulled away easily to disclose a gnarled ma.s.s of regenerating bone beneath it (Pl. 29, Fig. 3). There were areas near the posterior margin of the carapace of each specimen where regenerated epidermis was evident but where the bone was seemingly uninjured; the regenerated epidermis was nearly transparent.

Areas of regenerated epidermis on specimens of _T. ornata_ were rough in texture and slightly paler than the surrounding scutes.

Color-pattern is not reproduced in the process of regeneration but irregularly shaped light blotches sometimes occur in the places where radiations or other distinct markings formerly were present. A slight depression remains on the sh.e.l.l after regeneration is completed. I suspect that small injuries may be repaired in the course of a single growing season but that injuries involving a large part of the sh.e.l.l may take several years to heal completely. Cagle (1945:45) reported that a bullet wound in the sh.e.l.l of a painted turtle (_Chrysemys picta_) healed completely in approximately 23 months. Danini (_loc.

cit._) found that regeneration of the sh.e.l.l in _Emys...o...b..cularis_ was complete in as short a time as 225 days. Woodbury and Hardy (_loc.

cit._) stated that small injuries to the sh.e.l.l of _Gopherus aga.s.sizi_ may take as long as seven years to heal.

ECTOPARASITES

Two kinds of ectoparasites were found on ornate box turtles in the course of the present study; larvae of chigger mites (_Trombicula alfreddugesi_) were abundant on specimens collected in summer and, larvae of the bot fly (_Sarcophaga cistudinis_) were found on specimens throughout the season of activity, and, in a few instances, on hibernating turtles. In general, these ectoparasites do little or no harm to ornate box turtles, although heavy infestations may cause temporary interruption of normal activity or may even cause occasional death.

Concerning the larvae of _T. alfreddugesi_, Loomis (1956:1260) wrote, "In northeastern Kansas, larvae become numerous in early June (shortly after they first appear), increase in numbers to greatest abundance throughout late June and July, decrease slightly in August, become markedly reduced in September, and only a few larvae (mostly on hosts) remain in October and early November." He considered _T. alfreddugesi_ to be the most abundant chigger mite in Kansas and stated (_op.

cit._:1265) that it is most common "... in open fields supporting good stands of gra.s.ses, weeds and shrubs, and where moderate to large populations of vertebrates are present." Loomis listed ornate box turtles (_op. cit._:1261-2) as important hosts of _Trombicula alfreddugesi_ but noted that box turtles are not so heavily infested as are certain other reptiles. The two other species of chigger mites that Loomis (_op. cit._:1368) found on _T. ornata_ in Kansas (_T.

lipovskyana_ and _T. montanensis_) were not found in the present study.

Box turtles were considered to have chigger infestations when the reddish larvae could be detected with the unaided eye. No chiggers were seen on turtles in the period from spring emergence until June 13, 1955. On the latter date a few scattered chiggers were noted on several individuals and it was on this same date that the writer received his first "chigger bites" of the year. Numbers of chiggers increased in the latter half of June and heavily infested turtles were noted throughout July. No chiggers were seen on box turtles after mid-September in 1955.

Chiggers were ordinarily found only on the soft parts of the turtles'

bodies. Early in the season infestations were chiefly on the head and neck. Favorite sites of attachment were the point where the skin of the neck joins the carapace and on the skin around the eyes. Later in the season some chiggers could be found on nearly every part of the body where soft skin was present; concealed areas of skin, such as the axillary and inguinal pockets, the a.n.a.l region, and the inner rim of the carapace (where it joins the skin of the body), harbored concentrations of chiggers. Juveniles were relatively more heavily infested than adults and, even early in the season, had chiggers attached along many of the interlaminal seams of the sh.e.l.l. Broad areas of soft, newly-formed epidermis on the sh.e.l.ls of juveniles probably afforded a better place of attachment to chiggers than did the interlaminal seams of adults. The interlaminal seams and transverse hinges of adults were not infested until the height of the season of chigger activity. Heavily infested adults, observed in early July, were literally covered with chiggers; red larvae outlined nearly all the scutes of the sh.e.l.l, the a.n.u.s, the mouth, and the eyes. When turtles were picked up for examination, chiggers could be seen moving rapidly from one interlaminal seam to another.

Box turtles kept in outdoor pens and in the laboratory did not long maintain visible infestations of chiggers, even during the time in summer when turtles found in the field were heavily infested.

A four-year-old juvenile was found nearly immersed in the shallow water of a pond on July 4, 1955; its right eye had been damaged by an especially heavy concentration of chiggers. When I released the turtle, some 50 feet from the pond, it returned to the water and spent the next four days there. The turtle was probably in a period of quiescence induced by the eye injury and the heavy infestation of chiggers; immersion in water could be expected to help free the turtle of chiggers and to relieve trauma resulting from the injured eye.