Hard to measure Lilliputians: Error in the measurement of body temperature of small amphibians and a method to avoid it

Accurate measurement of body temperature (T_b) is essential for understanding the physiological ecology of ectothermic organisms. However, in small-bodied amphibians, heat exchange during handling and intrinsic limitations of measuring devices and/or protocols can introduce substantial error in T_b measurements. Here, we first estimated the magnitude of such an error for methodological approaches commonly adopted in the measurement of T_b in small ectothermic organisms in the field. We then propose a novel, low-cost restraint method, using a modified Foerster forceps, aimed at reducing heat transfer during handling. Accordingly, we compared the short-term changes (<2 min) in the T_b of frogs restrained with bare or gloved hands with those measured under the new proposed restraint method. We performed all tests using the Hylidae treefrog Dendropsophus nanus, whose small size made it perfectly suited to our goals, and we also repeated the experiments at three ambient temperatures (15 °C, 25 °C, and 35 °C), which allowed us to evaluate the influence of this variable on the measurement error. Our results clearly demonstrate that heat transfer from the researcher's hand to a small-bodied frog causes large deviations in the measured T_b and that those changes happen at an amazingly fast pace (within seconds) and are most prominent at lower ambient temperatures. The largely preconized use of gloves during measurements indeed reduced the error in T_b measurements to a moderate extent; however, the novel Foerster method herein proposed resulted in minimal change in frog's T_b and proved to be a superior and viable alternative to improve the accuracy of T_b measurement in small-bodied ectotherms.