Pet Turtles Research Articles

PET TURTLES AS A CAUSE OF HUMAN SALMONELLOSIS.

An infected pet turtle was shown to be the source of a case of human salmonellosis. The investigations that followed this finding resulted in 16 serotypes of Salmonella being designated turtle-associated types. Epidemiologic follow-up of these serotypes demonstrated, in addition to the original case, five index cases where turtles were the source of infection. Nine additional patients with salmonellosis had owned turtles but they were unavailable for sampling. Of the 22 infected persons found, 6 had been hospitalized and 13 excreted salmonellae for over 30 days. The majority of those infected were under 6 years of age; however, three mothers were also infected. The chain of transmission could be broken by the strict adherence to sanitary procedures when handling turtles or the production and marketing of turtles and turtle food free of salmonellae.

Sensory Cues in the Feeding of the Ornate Box Turtle

Although many obser+rations have been made on the behavior ofbox turtles, little is known conceraing olfaction in these animals. As far as the writer has been able to determine, Allard ( 1949: 147-148) is the only author who has actually tested olfaction in box turtles, but no definite conclusions were drawn from his experiment. Carr (1952: 17) commented that there are few experimental data to support the oft-repeated assertion that the sense of smell is acute in turtles. Pope (1955 :66) stated, sense of smell in turtles has not been carefully studied, but apparently it is good at close range. Pet turtles often appear to smell objects before eating or rejecting them. During August, September, and October, 1964, ornate box turtles (Terrapene ornat) were collected for the purpose of testing the relative importance of olfaction and vision in feeding. Each was identifiea by a number painted with fingernail polish on the carapace. The turtles were measured within two weeks after capture and again on February 7, 1965 (see Table 1 ) . The main experimental group numbered 15, and was a randolm sample as far as locality of capture, size in weight and length, and sex were concerned. Unfortunately, the larger turtles (numbers 8, 10, 12 through 18) are not well represented in the experimental results because they seldom demonstrated any eating interest. Allard (1948: 311 ) mentioned the tendency of full grown box turtles to sholw a greatly reduced interest in food in the autumn as early as September or October. Eating habits varied; many of the turtles were avid eaters, while other individuals habitually or occasionally demonstrated a marked indifference to food. These eating habits are clearly reflected by the weight gain ot loss during the winter (see Table 1). Of course, only those that displayed an interest in eating are represented in the experimental results. Many of the experiments were made during the period when these animals normally are hibernating. However, it has been assumed that the results are valid because ( 1 ) only those turtles that displayed interest in eating have been included in the experimental results, and ( 2 ) the same type of behavior was demonstrated in mid-winter (after the turtles had been warmed) as had been observed in August, September, and October. Neill (1948:114) observed,With the advent of cool weather, the [Carolina boxl turtles began to dig holes, usually in concealed spolts . . . should the weather again turn warm, they abandon the excavations and