The Biology and Behaviour of Ptinus Tectus Boie. (Coleoptera, Ptinidae), A Pest of Stored Products: II. The Amount of Locomotory Activity in Relation to Experimental and to Previous Temperatures

It has long been known that temperature has a great effect on the speed of biological processes. For this reason, when the speed of such a process is being measured, the experimental temperature has to be controlled. It has only recently been realized, however, that the temperature at which the biological material has been kept before the experiment may also be important, even in cases when damaging temperatures are not in question. For example, Thompson (1937) recorded the rate of heart beat of embryos of certain grasshoppers; for late embryos of Melanoplus femur-rubrum reared and tested at 20°C. this rate was considerably higher than for similar embryos tested at the same temperature but reared at 35°C. In fact, the rate for embryos reared and tested at 20°C. was about the same as that for embryos reared at 35° C. and tested at 30°C. Mellanby (1940) found that the rates of heart beat of the crested newt (Triton crista tus) were much higher at both o and 10°C. if the animals had been kept at 10° C. merely for the previous day than if they had been kept at 30°C. for the same time. Again, Hopkins (1937) found that the rate of locomotion of an amoeba, Flabelhda mira, depended on temperature in the usual way only if some hours had been allowed for temperature acclimatization ; during that period, the rate fluctuated widely before becoming steady. On the other hand, Mellanby (1939) concluded that the speed of walking of bedbugs (Ctmex lectularius) depended only on the experimental temperature and not at all on the previous temperature, provided the previous temperature had not been low enough to cause, chill coma nor high enough to be damaging, and provided the insects were allowed half a minute to acclimatize to the new temperature.

There is some other information about the effects of previous temperatures on processes such as these, but it is scanty; our work on the subject was to have been a comprehensive study of Ptinus tec tus, but it was interrupted at an early stage by injuries sustained by one of us (H.S. H.) during an air raid. The incomplete results on locomotoiy activity are presented now because they include new and useful information and because there is no immediate prospect of extending them.

Locomotory activity is important in the life of this beetle not only because it is partly responsible for spreading the species to neighbouring places, but also because the effectiveness of certain lethal agents depends on how much the insects are moving about.

In order to find the total amount of movement in a given time, it is necessary to know two things: how much of the time is spent in movement, and the speed of the movement while it is taking place. An estimate of the intensity of locomotory activity could, of course, be made by means of a suitable combination of measurements of oxygen consumption; our work, however, deals only with the amount of time spent on locomotion. It will be seen that this does depend on the previous temperature as well as on the temperature during observation.

Up to the present, three kinds of observation have been made on the effects of temperature on the locomotory activity of invertebrates. First, Crozier (1924) measured the rate of walking of a millipede, Parajulus pennsylvanicus, as a function of temperature, the object being to find the ‘critical thermal increment’. Second, Chapman et al. (1926) raised the temperature of a vessel containing various insects at a rate of 21° C. per hour, and recorded qualitatively the kind of activity—first movement of a limb, first crawling, first sign of heat paralysis, etc.—corresponding to each temperature. Third, Nicholson (1934) estimated the proportion moving in a batch of blowflies under given temperature conditions; since this is essentially the measurement we have adopted, it will be dealt with more fully.

In assessing the activity of Ptinus, the kind of activity was practically ignored. Locomotion was recorded separately from mere movements of the legs or antennae which did not result in the insect changing its location (‘virtual inactivity’), but no special attention was paid to the speed of locomotion. Ptinus tectus does not fly. When separate curves were drawn, showing locomotory activity alone and activity plus ‘virtual inactivity’, the latter were somewhat the higher, but there was no essential difference in shape. The general conclusions to be reached are not affected by discussing or including the records of ‘virtual inactivity’, so they are omitted for simplicity. We are concerned simply with the number of animals in locomotion under given conditions.

The observations of activity were made by the ‘cross-section method’ (Bentley et al. 1941). The aim is thus to count the animals which are moving at one instant; if an individual observation were to take an appreciable time, then the number of animals counted as moving would increase as the duration of this time increased, for some individuals would start to walk during the count. By this cross-section method, one might, for example, find that 60 % of the animals were moving at an instant; with a reasonably constant level of activity over a period, it can be inferred from this that on the average each animal walked about for 60 % of the period in question. If individual animals were to be considered, it would of course be found that the time spent in activity varied from one individual to another.

Some of the factors, other than temperature, which affect the activity of Ptinus were easily eliminated: thus the normal diurnal rhythm of activity was abolished by keeping the insects long enough at constant temperature and in unvarying light (Bentley et al. 1941); observations were usually not started for several hours after the animals had been placed in the observation chamber, so that the activity due to the mechanical stimulation of handling had had time to subside; as far as possible, the light intensity in the observations was the same as for the cultures. With the aid of these precautions, reasonably self-consistent results were obtained, but very smooth curves are not to be expected, because the activity of Ptinus appears to be sensitive to many conditions.

In considering the effects of temperature both before and during the observation period, there appears to be only one set of conditions in which no arbitrary choice of temperature or time relations has to be made; that is, the set of conditions in which a number of batches of animals are bred at different temperatures and the activity of each batch is assessed at its temperature of breeding. Such a procedure allows only a limited comparison to be made between different batches. If animals are to be tested at a temperature other than the temperature of breeding, then the speed at which their temperature′ is changed and the time they are left at the new temperature before testing*have to be chosen arbitrarily. Had sufficient data been accumulated, it might have been possible to reduce the qfbitrariness of the choice of conditions. The conditions chosen are mentioned under the appropriate headings.

The experimental animals were normally bred and kept at 25° C., except when the effects of a different culture temperature were to be investigated. They were bred in $212$ 1. jars (Breffitts) containing about 600 g. of wholemeal flour and 30 g. of dried yeast; the animals could drink water from a tube filled with wet cotton wool; some pads of dry cotton wool lay on the flour, and animals for experiments were usually picked off these pads. The light was on constantly, day and night, and its intensity was about 25 m.c. around the cultures. The animals used were of mixed and unknown ages: during the observations of activity they were neither fed nor allowed to drink ; none of them was used more than once.

The observations were carried out with the animals in glass dishes immersed in water baths. The type of dish was cylindrical, 16 cm. in diameter and 8·5 cm. high, with the rim ground flat to make a good joint with a plate glass lid. This lid was held on by a pair of joiner’s clamps, which also served to sink the dish, and the joint was made air tight with vaseline and, at temperatures over 30° C., an outside layer of paraffin wax. There were two holes in the lid, one for a thermometer and the other carrying a piece of 6 mm. glass tubing through which the animals were dropped into the dish. This tubing was not stoppered, but a piece of capillary′ tubing was fixed to the open end with rubber tubing, thus preventing gross exchanges of air between the dish and the outside but allowing pressure adjustments. The beetles could walk about freely on a perforated zinc platform, but were prevented from getting around the edges and below the platform by a ring of electric flex, freed from its insulating cloth and rubber, squeezed into the small gaps between the platform and the glass wall. Below the platform there were glass pots of potassium hydroxide or sulphuric acid solution of appropriate concentration to control the humidity. Observations were almost always made concurrently at three different humidities (approx. 34, 60 and 95 % R.H.); the effect of humidity on activity was small and uncertain compared with the effects of temperature and it will not be dealt with here. Five animals were used in each dish in each experiment.

One of the two water baths used for temperature control was a Refrigerated thermostat designed and constructed by Mr R. J. Whitney. There was artificial illumination of about 25 m.c. around the dishes, the tank being completely enclosed in wood. This tank was used for experiments at constant temperatures between 1·5 and 15° C. The other water bath could be used as a thermostat with the normal arrangement of a 400 W. heater, mercury-toluene regulator and Sunvic vacuum switch. For experiments between room temperature and tap-water temperature, cooling was carried out with tap water passing through metal tubing. This tank was surrounded by light-proof curtains and, once more, the light intensity was about 25 m.c. This second tank could be heated rapidly with bunsen burners and cooled with ice or a stream of water.

The first observations were made on the activity of animals bred and kept at 25° C. ; they were transferred to the experimental dishes, which had already been allowed half an hour to equilibrate in the thermostat, and then left at the constant temperature of the particular experiment for over 24 hr. The first reading was taken after 23 hr., and nine further readings were made at 10 min. intervals. For each of the three dishes with five animals in each, the number of records of activity from one experiment was thus fifty. Each point on Fig. 1 shows what percentage of these fifty records were records of animals moving from place to place. .Thus two series of experiments each with three dishes show that at 10° C. averages of between 10 and 18 % of the animals were walking about at any instant.

In Fig. I the seventy-one points represent observations on that number of different batches of five animals. Generally speaking, this graph shows a surprising selfconsistency. The only notable inconsistency was at 20°C; the three batches in the second series at this temperature were all rather inactive, and no explanation has been found for this.

In Fig. 2 the six sets of observations of Fig. 1 are shown combined in the curve marked 25° C.

A preliminary series of experiments was done with animals which had been kept cooler than 25° C. A culture of young adults (less than a month old) was simply immersed in a tank which was kept cool with tap water. Between the beginning of the experiments in May and the end in July, the temperature rose from 13·5 to 18’ C. The observations began after 3 weeks and the culture temperature was about 16° C. when most of the experiments were done. In other respects the experiments were like the previous ones.

The results are shown in the curve marked 16° in Fig. 2, and they were generally similar to those for a previous temperature of 25°C., but the activity recorded was less. Further experiments were therefore done with carefully controlled previous temperatures of 15 and 28°C. It was not then known that 28° C. is too high a temperature for the development of Ptinus (Ewer & Ewer, 1941), but it does not seem to have been too high for the purpose of these experiments.

These two sets of observations were like each other, but different from the two previous sets in tljat the cultures had to be kept in constant darkness. At each temperature and humidity two batches of five animals were studied simultaneously, one from 15 and one from 28° C. Since all the animals had originally been taken from one culture, the two sets of experiments are comparable.

It will be seen from’Fig. 2 that when observed at any temperature between 10 and 30° C. the activity of animals previously kept at 28° C. was always higher than that of animals previously kept at 15°C., the increase being about 60 % on the average.

Leaving aside for the moment the differences between the four curves of Fig: 2 and considering only their general shape, it will be seen that they are mutually confirmatory. They have been combined in the lowest curve of Fig. 4; in this case, a number of the set of experiments for a previous temperature of 25° C. (e.g. at 7·5, 12·5, 17·5° C., etc.) have been omitted, so as to avoid unduly weighting the curve with this set.

Up to an experimental temperature of 15° C. the activity after a day at constant temperature is higher at each successively higher temperature. This is in line with well-known physiological effects of temperature. But between 15 and 25° C. the activity is relatively constant. It seems very unlikely that activity′ in this broad temperature region reaches a maximum because of harmful effects, of temperature. For example, Deal (1939) has shown that Ptinus has a preferred temperature zone around 23° C. and this has been confirmed by Dr Gunn and Miss Walshe in this laboratory. Ewer & Ewer (1941) have shown that the life history of Ptinus is shortest at 23–25° C. and they did not find any harmful effects of temperature on adults at 25°C. or below. We have had generation after generation of Ptinus in cultures kept at 25° C. Indeed, Fig. 2 itself shows that after a day at 35° C. Ptinus walks about to much the same extent as it does after a day at 15°C., and this activity was not notably abnormal in character. We therefore consider that the shape of the curves between 15 and 25°C. in Fig. 2 represents the relation between temperature— previously constant for a day—and the frequency of normal non-pathological locomotory activity. This relation is quite unlike that approximately expressed by the Q₁₀ rule. Indeed, if one does not attach much value to the somewhat higher average activity at 15° C., one may say that the frequency of locomotory activity of Ptinus (not the velocity of locomotion) does not vary much between 15 and 35° C., provided the temperature has been constant for a day.

In some of the experiments just described, observations were made incidentally during the hour after transfer of the insects to the experimental dishes. Instead of the average activity being about 30 % as it was after a day, it was usually higher than 60 %, especially between 12·5 and 37’5° C. Part of this high activity was doubtless due to the recent handling which the animals had had, so that in investigating the other factor—recent change of temperature—the animals were placed ready for observation a day before observation started.

The apparatus was essentially as before, but air temperature inside the experimental dish was estimated with a thermocouple, because a mercury thermometer follows air temperatures too sluggishly. The wires of the couple were rather thick, so that the readings were affected by the temperature outside the dish, and consequently the rates of temperature change given below do not pretend to precision.

The observation chambers were prepared and, with the animals inside them, were left at about 2°C. overnight. Each chamber was then moved to the water bath for observation and the water temperature was raised at between 3 and 7·5° C. per hour. The results of four experiments with eleven different batches of animals are shown in Fig. 3. Activity rose to a maximum at between 10 and 15°C. and then fell. In a solitary experiment (three batches of animals) there was a second rise at 30°C. Thus at about 10° C. activity amounted to 40–90 %, instead of 20 % or below as in the experiments at constant temperature. Between 20 and 30°C., however, activity was much the same whether the temperature was constant or rising at the fairly slow rate of about 1° C. in 12 min.

When a faster rate of temperature rise was used—about 14° C. per hour—the resulting activity reached a high level at 15° C. and remained near too % until lethal effects began at about 40° C. Fig. 4 shows how different the activity levels were with the slower and faster rates of temperature rise.

Three experiments were carried out with the temperature falling at between 4 and 7·5° C. per hour, using eight batches of five animals each. The initial temperatures, at which the animals were kept for the day preceding observation, were 19, 27 and 29° C. The results are shown separately in Fig. 5 and combined in Fig. 6. It is perhaps surprising to find that when the temperature is falling slowly, the animals are usually as active at a given temperature as when it is rising fast; below 10° C. they seem to be more active on cooling (Fig. 6).

The results given above, incomplete though they are, are sufficient to jndicate the varied ways in which the frequency of locomotion of Ptinus depends on temperature—culture temperature, the temperature for the previous day, and the temperature, falling or rising fast or slowly, in the previous few minutes. At this stage the results are suggestive, but analysis awaits more complete observations.

It is well known that the rate of frequency of the most varied biological processes, such as oxygen consumption, mitosis, growth and even walking, increases with temperature in a way which can be approximately expressed in the Q₁₀ rule or, what comes to much the same thing, the Arrhenius formula (Bélehrádek, 1935 ; Heilbrunn, 1938). The value of is often about 2 or 3, as long as the temperature region considered is not too close to or above the optimum. Examination of our curves shows no resemblance whatever to the usual type of Q₁₀ curve, unless the optimum is considered to be at or below 15° C., for above this temperature activity rises little in any of the temperature conditions we have dealt with. It would be reducing the already difficult conception of optimum to the absurd to regard 15° C. as optimal in any way for Ptinus\ this temperature lies at or below the middle of the temperature range in which Ptinus can develop and carry on its normal activities (Ewer & Ewer, 1941). It is not particularly surprising that the frequency of locomotion of Ptinus does not follow the Q₁₀ rule; what js surprising, even with selected data, is that the velocity of walking of certain animals does roughly follow this rule.

In Nicholson’s (1934) experiments on Lucilia cuprina, activity after 12 hr. at constant temperature was highest at 30° C., which is about the preferred temperature, and fell away rapidly above and below that temperature; in temperature rising at about 7° C. per hour, activity was actually less at medium temperatures and there appeared to be a stimulating effect of changing temperature only outside the preferred zone, below 15 and above 37° C. This corresponds roughly with our results for slowly rising temperature (3–7·5° C.), as far as they go. But with slowly falling temperature the activity′ of Ptinus is much greater, and so it is with quickly rising temperature (14° C. per hour). These results demonstrate an effect of temperature change as distinct from temperature level. Such an effect has already been shown by Kennedy (1939) in experiments on the locust, Schisiocerca gregaria, in the Sudan. A sudden change of temperature eventually resulted in the locusts reaching a new activity level corresponding to the new temperature, but the inmediate result of the change was a rise above the appropriate constant level. Indeed, when the temperature fell there was an absolute rise in activity which was more prolonged than the temporary rise due to a rise in temperature. In this work of Kennedy’s, the change of temperature was very rapid, being initially i°C. in 10–15 sec.

Our work reveals a weakness of the observations made by a number of authors on the effect of rising temperature on insect activity. They raised the temperature at rates of about 7° (Nicholson, 1934), 10° (Bodenheimer et al. 1929), 11° (Hussein, 1937), 20° (Nieschulz, 1933, 1935), and zi°C. (Chapman et al. 1926) per hour. But the temperature at which most of the animals are first active in our experiments is about io°C. when the temperature is raised at 3·7′5° C. per hour and about 15° C. when it is raised at about 14° C. per hour. In fact, over the range of rates which have been used by the various authors, the rate itself is likely to be as important as the temperature in determining the point of first maximum activity; the other standard points may be affected in a similar way.

Nicholson’s work (1934) on activity in blowflies seemed to indicate a relation between activity, as measured by the method we have adopted from him, and temperature preference (Fraenkel & Gunn, 1940, pp. 206–11). In the temperature gradient apparatus in use in this laboratory (Gunn, 1934, 1935), however, Ptinus can walk along the gradient at a rate equivalent to 10° C. per minute or 600°C. per hour! It would therefore be unwise to attempt to draw from our experiments any conclusion whatever about the mechanism of preferred temperature in this particular gradient.