Effects of Ambient Temperature on Warm-Up in the Moth Hyalophora Cecropia

The capacity of lepidoterans to maintain high and relatively uniform body temperatures during flight has attracted attention from a number of investigators, particularly during the last few years. However, the phenomenon remains incompletely understood and some of the published data are contradictory, notably with regard to the influence of ambient temperature on rate of pre-flight warm-up.

Heath & Adams (1967) reported that in the sphingid, Celerio (Hyles) lineata, and the saturniid, Rothschildia jacobae, rate of warm-up was independent of ambient temperature. McCrea & Heath (1971) have reported a similar situation in Manduca sexta (Sphingidae), and Hanegan & Heath (1970) found the same in Hyalophora cecropia (Satumiidae). Other workers, however, who have measured this phenomenon in Lepidoptera have reported that rates of warm-up increase directly with ambient temperature just as is the case in heterothermic birds and mammals. In sphingids the dependence of warm-up rate on ambient temperature has most recently been reported for M. sexta by Heinrich & Bartholomew (1971). However, the same situation had been shown previously by Dotterweich (1928, p. 414) and Dorsett (1962). In addition, it has been reported that the rate of warm-up in the monarch butterfly, Danaus plexippus, increases directly with ambient temperature (Kammer, 1970).

The factors responsible for these conflicting results are not clear - differences in taxa, populations, age of individuals, and methods of measurements all have been suggested. The present study re-examines warm-up in Hyalophora cecropia in the hope of resolving this confusion.

The moths were obtained as pupae from a commercial dealer in New York. Seven males (mean weight, 1·84 ±0·45 g) and 11 females (mean weight, 3·90 + 0·88 g) were measured. Forty-gauge copper-constantan thermocouples were implanted on the morning of eclosion as described by Heinrich & Bartholomew (1971). The moths, with thermocouples attached, were placed in a transparent plastic box and left undisturbed overnight in a constant-temperature cabinet at 20 °C and a photo-period of 12 h (lights on at 8.00 a.m.). The following morning at about 10.00 a.m. the temperature was set at the desired level and the system was allowed to come to equilibrium. The moths were then stimulated to arouse by turning off the lights in the chamber and sometimes by touching them gently. Over a period of about 6 h each moth was stimulated to warm-up and then allowed to cool-down at 15, 20, 25 and 30 °C. Not all moths would warm-up at all four temperatures.

Usually, the lowest temperature was employed first and the highest last, but occasionally the sequence was reversed. The only air movement was that produced by the wing quivering of the moths. All the animals were measured on the first day after eclosion. Environmental and body temperatures were recorded on a multichannel potentiometer at a chart speed of 1 cm/min with the body temperature being printed at intervals of either 5 or 10 sec. Temperature measurements were accurate to o-i °C.

Under the conditions of measurement the rate of thoracic warm-up increased linearly with increasing ambient temperatures (Figs. 1, 2). The warm-up rates of males and females did not differ significantly and were not affected by the sequence of ambient temperatures employed. When their thoracic temperatures reached 34−41 °C, the moths attempted to fly. Thoracic temperatures at time of take-off were slightly lower at ambient temperatures of 15 and 20 °C than at 25 and 30 °C (Table 1).

The animals usually attempted to fly for only a brief period and often stopped fluttering after a few seconds and cooled down. During warm-up most of the heat was confined to the thorax. Abdominal temperature rose slowly and by only a few degrees, usually not peaking until after the thorax had begun to cool (Fig. 3). Abdominal temperatures were recorded from five moths. No correlation was apparent between ambient temperature and the rate or extent of increase in abdominal temperature. The mean increase of abdominal temperature was 3·2 °C (range 1·1 to 6·5).

The rates of cooling of the moths were essentially linear when plotted semi-logarithmically. Cooling was most rapid, of course, when the difference between thoracic and ambient temperatures (ΔT) was greatest. However, the rate of cooling at a given ΔT was virtually unaffected by either ambient or thoracic temperature per se (Fig. 4).

We were unable to confirm the conclusion of Hanegan & Heath (1970) that rate of warm-up in H. cecropia is independent of ambient temperature. We found that the rate of warm-up in this species on the first day after eclosion was a linear function of ambient temperature between 15° and 30 °C.

The differences between our findings and those of Hanegan & Heath are probably due to differences in experimental procedure. Our animals were all freshly emerged. Those of Hanegan & Heath were measured continuously for a week. It is known in H. cecropia that ‘the ability of the moth to raise its body temperature decreases with age’ (Oosthuizen, 1939 p. 72). Consequently, averages over a 7-day period may be misleading.

Hanegan & Heath usually recorded thoracic temperatures at intervals of about a minute, while we employed intervals of 5 and 10 sec. At an ambient temperature of 30 °C warm-up is often completed in 1 min or less. Thus, with a sampling interval of i min the rate of warm-up is almost certain to be underestimated at high ambient temperatures as compared with low ambient temperatures.

Our data and those of Heinrich & Bartholomew (1971) indicate that the rate of warm-up in saturniids and sphingids increases with ambient temperature. If this is true, the theory proposed by McCrea & Heath (1971, p. 426), that heat production during warm-up depends on the difference between thoracic and ambient temperature (ΔT), encounters difficulties. It seems more reasonable to suggest that the rate of heat production during warm-up is primarily determined by the thoracic temperature, while the rate of heat loss is primarily determined by ΔT. For a given thoracic temperature, ΔT is inversely related to ambient temperature. Hence, the lower the ambient temperature, the greater the heat loss, and consequently, the slower the rate of warm-up. Such a simple interpretation does not require the postulation of special physiological mechanisms or elaborate control systems.