Volatile repellent, benzaldehyde, see Figures 2A and 2B. The experiment was carried out in a dark room 15 cm away from a turned-on parallel visible-light source, which served for visibility throughout the duration of the experiment. (A lighted room also works but the light must be uniform, which may be difficult to achieve, because non-uniform light is itself a stimulus.) For Method 1 (Figure 2A) about 20 flies were placed into a vial, it was plugged with cotton, then it was placed into the dark room for 30 minutes parallel to the light source to adapt the flies to this level of light. At 15 minutes into incubation 1 ml of 100 mM benzaldehyde dissolved in 1.5% agar was pipetted into another vial, which was then covered with mesh cloth (we used tulle); a connector served to keep the cloth in place. This vial of benzaldehyde was covered with Parafilm and then placed into the dark room. At 30 minutes the flies were gently pounded into a 2.5×20 cm test tube, which was then connected to the benzaldehyde vial by means of the connector, all in the dark room. Then the flies were gently pounded to the cloth. Starting at that time, they were counted every minute in each third of the test tube for 20 minutes. A control without benzaldehyde was also run.

For Method 2 (Figure 2B) the experiment was again carried out in a dark room 15 cm away from a turned-on parallel visible-light source. About 20 flies were placed into a 2.5×20 cm test tube with a screw cap (Fisher 14-930F) and kept in the dark room for 30 minutes parallel to the light source. At that time the flies were randomly distributed. Then the screw cap was gently removed and replaced with a screw cap containing 1 ml of 100 mM benzaldehyde dissolved in 1.5% agar. The number of flies in each third of the test tube was then counted every minute for 10 minutes.

Non-volatile attractant, sucrose, see Figures 3A and 3B. Either the flies were used without starvation, or they were first starved in presence of 1 ml of 1.5% agar for 18 hours to make them hungry. Method 1 (Figure 3A): In one 2.5×20 cm test tube that was microwaved to make warm the entire interior was evenly coated with 3 ml of 1.5% agar, and in another 2.5×20 cm test tube, it was also warmed and coated with 3 ml of 1.5% agar containing 100 mM sucrose. This was done by placing melted agar into the tube and then rotating the tube horizontally, taking care to ensure even coating. The exterior of the tube was then run under cool water until the agar hardened. A vial was coated in the same way with 1 ml of 1.5% agar, about 20 flies were put into it, and it was plugged with cotton. All three were then placed in a dark room for 30 minutes 15 cm away from a turned-on parallel visible-light source. The cotton was then removed, the flies were put into the non-sucrose test tube and pounded down, and the two test tubes were put together with a connector. A timer was then started and the number of flies was scored every minute in each quarter of the test tubes. In a control, both tubes contained only agar, no sucrose.

Method 2 (Figure 3B): In an alternative assay that did not involve coating with agar, 100 mM sucrose in 1 ml of 1.5% agar was placed at the end of a 2.5×20 cm test tube, then the starved flies in a vial were connected to the test tube (after 30 min in the dark as above). A timer was started and the flies in each third of the test tube were scored every minute. In a control the agar contained no sucrose.

Attraction to water, see Figure 4. Flies were deprived of water for 18 hours by placing them in a vial containing sucrose crystals for food. Then about 20 of these flies were put into an empty vial plugged with cotton for 30 minutes in a dark room 15 cm away from a parallel always turned-on visible-light source. A 2.5×20 cm test tube with a connector attached (covered with parafilm) and another 2.5×20 cm test tube with Kimwipe and 0.5 ml distilled water (covered with parafilm) were also placed parallel to the light source. After 30 minutes, the flies were then transferred into the test tube without Kimwipe and pounded down to bring them to its end, then this tube was put together with the test tube that had Kimwipe soaked in distilled water. The number of flies was counted every minute in each quarter. In a control, the same was done except that no water was added to the Kimwipe. (Water of course is volatile but see Results and Discussion.)

Non-volatile repellent, quinine HCl, see Figure 5. The procedure was the same as for sucrose in Figure 3A, including agar throughout, except that the flies were not starved, they started at the rounded end of the 100 mM quinine HCl half, and the number of flies in each half was scored. In a control, no quinine was used.

Male flies were mutagenized overnight with 25 mM ethyl methanesulfonate in 1% sucrose on Kimwipe paper. Then they were mated with virgin attached-X females, then at six days the flies were removed, and at around two weeks the new flies were studied (approximately 1 to 7 days old). About 5,300 of these F1 males were screened at 34°C, around 500 at a time, by placing them near the repellents benzaldehyde and heat at one end of an 11×70 cm tube while the other end had the attractants light and a favored temperature. Parental flies quickly migrated to the attractant end, while mutant flies were distributed throughout. In this way five independent male mutants were isolated (see RESULTS, Part II). Full details of these methods will be presented by Vang et al. in “Drosophila mutants that are motile but fail to respond to stimuli” (manuscript in preparation, 2012).

Flies were placed in a tube near one end that had the repellents benzaldehyde and heat, while the other end had the attractants light and a favored temperature (see MATERIALS AND METHODS, Part II). By one minute the parent had already moved away from the end that had repellents to the end that had attractants, while the mutants failed to do that or did it slowly. Figure 19A shows the response of parent and Figure 19B of a mutant (four other mutants have also been studied). When only a single stimulus was presented, the parent by one minute had already responded, while the mutant did not respond; such results were obtained as well for the following single stimuli (nearly all those described above): light, benzaldehyde, quinine, temperature, and gravity (data not shown). In the absence of any stimuli the mutants moved slightly slower than the parent. Motility was also judged by a walking assay which continuously records freely walking single flies at 22°C in a tube that has an infrared light gate situated in the middle of its long axis; interruption of the light gate triggers a circuit ^[19]. Again the mutants moved slightly slower than the parent (data not shown). All the above comes from experiments at 34°C, but at room temperature the same results were obtained as far as carried out. Full details of all these results will be presented by Vang et al. in “Drosophila mutants that are motile but fail to respond to stimuli” (manuscript in preparation, 2012). A genetic study of the mutants is underway.

In this report we demonstrate the responses of Drosophila to a variety of stimuli by use of methods that are easy to carry out. Earlier work by others has often used different approaches that are sometimes more complex than these. Here we summarize some of that earlier work and relate it to the methods described in this report.

The response to light by fruit flies was measured already in 1905 ^[20] and then in 1918 ^[21] by methods basically similar to the one we employed although ours is perhaps “simpler” (see Figure 10A here). In another assay similar to the one here, fruit flies in a tube were separated from an empty tube by means of a barrier that could be removed when the light was turned on ^[22], Figure 3 in ref. ^[23]. Similarly the effect of age of the flies on phototaxis was measured in tubes that had the flies in a distant dark portion while light shone in at the non-dark portion ^[24]. The response of fruit flies to light has been measured in a countercurrent apparatus ^[25].

“Fast” phototaxis is movement towards light by flies that have been shaken, while movement towards light in absence of shaking is “slow” phototaxis ^[26]. To prepare the flies for study in Figure 10A they were first gently pounded down, so the initial response could be “fast” phototaxis; after some minutes the response shown in Figure 10A slows down, which may be due to “slow” phototaxis, since the flies were then no longer being shaken. Alternatively, the slow-down could be due to slow adaptation, which operates on a time scale of many minutes, instead of fast adaptation, which occurs within a second ^[3], ^[27].

Methods for measuring the behavioral response of fruit flies to volatile chemicals – smell chemotaxis, olfaction – have been reviewed ^[28]. In an avoidance assay flies in a test tube run away from repellent placed at the open end on a Q-tip coming out of a cotton swab ^[29]. The assay used in the present report is fundamentally similar to that (Figures 11A and 11C). In the jump assay a single fly jumps when a pulse of odorant is passed over it ^[30]. Another test for olfaction is based on a single tethered fly turning toward a higher concentration of chemical ^[31]. In the Y-shaped assay ^[32] and the T-shaped assay ^[33]–^[35] the flies avoid or are attracted into the branch that contains odorant. Response to volatile attractants has been measured in a 1- to 2-day assay by relying on diffusion of odor coming out of a micropipette ^[36]. We have tried, but did not succeed, to find a faster assay for volatile attractants.

For assays of the behavioral response to non-volatile chemicals – taste chemotaxis – see a review of available methods ^[28]. Flies have been fed on different chemicals or food sources each colored with a different dye, then color ingested was measured to determine the amount of tastant eaten ^[37]–^[39]. The proboscis extension reflex has been used to measure a variety of repulsive or attractive tastants ^[40]–^[43]. The methods used to assay for response to sucrose (Figures 12A and 12C) and water (Figure 13A) and response to quinine (Figure 14A) are different from those methods.

In the case of responding to water, the proboscis extension reflex has been used ^[42], ^[43] and the amount of water consumed by the fly has been measured ^[43]. The present assay for response to water (Figure 13A) is different from these: flies had to find the water. How did they do that? In a control that had no water at its end, some flies did get to the end just as fast as when there was water at the end, so it is not necessary to invoke the idea that the flies had to smell the water in order to get to the water, although that remains a possibility.

To assay response to humidity, flies have previously been placed in the middle between moist air in one tube and dry air in another, then the flies made a choice ^[44]. Our assay for response to high humidity is basically similar although “simpler” (Figure 15A).

The response of flies to temperature has previously been measured by placing them in the middle between high temperature in one tube and low temperature in another, as above for humidity, and also by providing them with a linear thermal gradient ^[44]. A fly was shown to avoid heat by jumping when its abdomen was heated or when it was put on a hot plate ^[45]. Our assay demonstrates avoidance of heat (Figure 16A) and of cold (Figure 17A) by measuring the movement of the flies in a thermal gradient.

The response to gravity has previously been measured by use of a vertical countercurrent apparatus ^[17], ^[46] and a vertical maze ^[14], ^[16], ^[47]–^[50]. In addition, flies have been observed as they climbed up a vertical tube ^[15], ^[21], ^[24], ^[51], ^[52]; our method (Figure 18) is basically similar, yet “simple” and quantitative.

The methods described in this report have proved suitable for our present research on the isolation and study of mutants that fail to be attracted or repelled although they are motile. Two repellents - benzaldehyde and heat - were placed at one end of a tube and two attractants - light and a favored temperature - were placed at the other end. Then the flies were put near the repellent end. The parent quickly went to the attractant end (Figure 19A) while the mutant did not (Figure 19B). The mutant failed also to respond to any individual attractant or repellent. Similar results were obtained for four other mutants we isolated [Vang et al., “Drosophila mutants that are motile but fail to respond to stimuli” (manuscript in preparation, 2012)]; see elsewhere ^[53] for a preliminary account of this.

What is the defect in the mutants reported here? We suppose that sensed information from all the different sensory receptors goes finally to central processing, which is the place that directs movement towards or away from stimuli. We consider that central processing is defective in these mutants so that they can no longer respond to stimuli, but motility still occurs. Central processing may well turn out to be located in the central complex.

In previous work by others ^[54]–^[61], mutants defective in attraction to light were isolated, then those that failed were tested anatomically for structural defects in their central complex. Those with such structural defects were found to be defective in motility. These authors have studied how motility is regulated by the central complex, while we have studied attraction and repulsions in largely motile mutants. It seems likely that both approaches deal with properties of the central complex.