Preening Behaviors of Flies (1984)

©1984, 2013 by Dallas Denny
Source: Dallas Denny. (1984). Preening behaviors of flies, with an emphasis on the common housefly, Musca domestica. Paper for Dr. Dick Porter, Psychology 357P, George Peabody College of Vanderbilt University.
 

 

 

 

Preening Behaviors of Flies

With an Emphasis on the Common Housefly, Musca domestica

A Review

Fall, 1984

By Dallas Denny 

For Dr. Dick Porter

Psychology 357P

Peabody College of Vanderbilt University

 

The preening (grooming) behavior of flies was relatively little studied before 1968. Since then, ethograms of a number of species have appeared in the literature. An unpublished ethogram of house fly grooming was made by Sustare and Burtt in 1976. While the patterns of grooming behavior have been defined, replications have not been done, and the functions of preening behavior and the manner in which the patterns are grouped into clusters (bouts) are not well known. The behavior of two captured house flies and a third fly of an unknown species are discussed. An analysis of unpublished data collected by Burtt, Denny, Molina, and Howell is also presented.

Casual observation of the common housefly, Musca domestica or of related species will reveal that these insects spend a good deal of their time in preening movements. Although there is some evidence that preening has functions other than cleaning, the terms preening, grooming, and cleaning have been used interchangeably in the literature, and will be so in this paper.

Considering the ready availability of these flies in great numbers during much of the year, coupled with ease of capture and the high percentage of time spent in preening, it is not surprising that a number of researchers have studied this group of behaviors. The purpose of the present paper is to review findings of previous researchers and to replicate the findings of Sustare and Burtt (1976).

As late as 1969, Szebenyi remarked that very little was known about the grooming behavior of flies. He noted that both Lorenz (1956) and Tinbergen (1965) had made (apparently somewhat casual) assertions about the grooming of Drosophila (fruit flies). Lorenz mentioned that wingless mutants possessed the normal movements of cleaning the wings (p. 55), and Tinbergen asserted that flies will always preen their wings as soon as dust particles stick to them” (p. 85). Szebenyi identified only two previous studies, one by Heinz (1949), and the second by Connolly (1968).

The paper by Heinz is in German with an English abstract. Heinz described the cleaning movements of lab-reared Drosophila mutants, and found them to be identical with those of wild flies. He also attempted to determine the conditions which elicited cleaning behavior. Heinz’ paper was apparently somewhat lacking in quantitative data (Szebenyi, 1969).

Connolly (1968) reported that Weidmann (1950), discovered that male fruit flies, Drosophila melanogaster, showed increased rates of preening when their courtship sequences were interrupted. Connolly (1968) found that, in Drosophila, flies engaged in preening were unlikely to be physically touched by other flies. Increasing the density of flies resulted in higher rates of preening (not, however, as a result of increased contact between flies). Connolly also observed a mutant strain with very poor vision, and found that the rate of grooming of the vision-impaired flies did not change when other flies were introduced. He concluded that preening serves as a signaling device influencing the spacing of flies.

Szebenyi (1969) prepared an ethogram of the grooming behavior of Drosophila melanogaster. He observed that preening movements were generally performed by the legs and were of two types: (1) sweeping movements of the legs over the body surface; and (2) rubbing together of legs along the tarsal joints. He observed on several occasions that a small particle of yeast which adhered to the body surface of the flies were picked up by the limbs, traveled along the tarsal joints, and dropped off of the end of the legs. He noted that the direction of movement of the particle was influenced by the bristles on the legs (which all point away from the body). Szebenyi remarked that grooming may have other functions, for instance, distributing wax over the body surface. In addition to 20 body cleaning (sweeping) and 7 leg cleaning (rubbing) movements, he noted that the movements could be subdivided as: (1) head cleaning movements; (2) thorax cleaning movements; (3) abdomen cleaning movements; and (4) wing cleaning movements. Recently Diliwith & Blomquist (1982) identified abdominal sites of sex pheromone synthesis in female house flies. The pheromone is presumably transferred to the legs (and hence to other parts of the body) by grooming movements.

Szebenyi (1969) also attempted to collect quantitative data on grooming behaviors. He managed this by assigning a simple code number to each behavior pattern. Patterns were spoken into a tape recorder, and then the tape was replayed and the data transferred to paper. Szebenyi was well aware of sources of error in this type of data collection, and addressed the problem thoroughly. By comparing data with cine, he determined the error rate to be 16 percent, with half of the error accruing in the transformation of the event into verbal sounds, and the other half in transferring the speech to paper. Szebenyi also considered the mechanical error introduced by the tape recording machine, but concluded that this error was minimized because of its constant error rate. Szebenyi concluded that grooming behavior was built up of complexes of the simple patterns he observed. Grooming occurs in indeterminate bouts; it starts suddenly, and is intruded upon by other behaviors.

Uta Seibt (1972) investigated whether, due to the unusual shape of diopsid flies, grooming patterns would be correspondingly modified, and whether their cleaning patterns were less stereotyped than other flies. She was also interested in the reaction of these flies to external irritants and amputations. She described the grooming of the stalk-eyed flies Diopsis sulcifrons and Diopsis somaliensis in detail. She found the cleaning behavior of her flies were similar to other flies; new behavior patterns appeared only due to the extreme development of their eye-stalks. Covering the flies with flour resulted in no changes in form nor sequence of cleaning behaviors, although there was a striking difference in the frequency of the patterns. Flies with amputated eye-stalks continued to groom the nonexistent eye. Flies with amputated limbs continued to move the stumps as if in cleaning patterns. These flies, however, exhibited a behavior pattern not observed in intact flies (both the middle legs cleaned the intact foreleg); there was likewise a new pattern involving grooming of the eye-stalk. Seibt concluded that grooming patterns were under strong central nervous system control.

Dawkins & Dawkins (1976) observed the grooming behavior of blowflies. They found eight categories of grooming behaviors: (1) front leg grooming; (2) tongue grooming; (3) head grooming; (4) middle leg groomed by front legs; (5) middle leg groomed by back legs; (6) back leg grooming; (7) abdomen grooming; and (8) wing grooming. These researchers undertook a lengthy and somewhat confusing analysis of the probability of certain behaviors following other behaviors, using a Markov chain analysis. They concluded: (1) that there were high probabilities that certain behaviors would be alternated with other behaviors; (2) that grooming was organized into bouts, involving either the first or third pairs of legs, but not both; (3) that most bouts contained an odd number of behaviors; and (4) that postural facilitation is involved in grooming (that is, flies shift their postures to compensate for the grooming movements.

Sustare and Burtt (1976) prepared an ethogram of grooming behaviors of the common house fly, Musca domestica. They divided grooming behaviors into fourteen patterns which were similar to Seibt’s (1972) and Dawkins & Dawkins (1976) (Table 1). They recorded 1033 sequences, including 22,838 individual grooming patterns. The longest bout observed consisted of 565 patterns. Their sequential analysis revealed that behaviors tended to occur in two-pattern cycles, and three-cycle patterns at a smaller frequency. They divided grooming patterns into “front-end” and “back-end” behaviors, as did Dawkins & Dawkins (1976).

Burtt, Denny, Molina, & Howell (1977) covered house flies with flour and left them in closed containers. The flies were observed under a microscope, and notes made of the routes of transfer of flour from their bodies. The flow of the flour followed a model proposed by Sustare and Burtt (1976).

Other research has centered on other types of flies: Drosophila (Hay, 1973, 1976; and Angus, 1974); and others (Sutcliffe & Mclver, 1974); and Schroeder, Mitchell, & Miyabara, 1974. There seem to have been no other reports of house fly grooming behavior.

 

Method

 

Subjects

Subjects were two house flies (Musca domestica) captured in Nashville, Tennessee during November and December, 1984. The flies were individually housed in clear glass bottles and fed daily with a paper tissue which had been soaked in skim milk. A third fly of a related species was captured, housed, and observed, but escaped before species identification could be made. This fly was similar in appearance to, but about twice the length of, house flies.

Because of the scarcity of subject flies, unanalyzed data (2000 patterns) collected by Burtt, Denny, Molina, & Howell (1977) were analyzed.

 

Data collection and analysis

An observer (the author) practiced until fluent the coded descriptions used by Sustare and Burtt (1976) in their ethogram. The codes were spoken into a tape recorder as they occurred, and then later marked on paper and tabulated. This procedure was essentially identical to that used by Szebenyi (1969). Reliability data were not taken, but it is reasonable to assume that the error rate was similar to that found by Szebenyi (16%).

 

Results

The data are presented in tabular form for the previously collected data (Tables 2 and 3). All behaviors observed in the two currently observed flies are not presented, but fell within one of the patterns described by Sustare and Burtt (1976). However, all of the patterns noted by Sustare and Burtt were not observed. This is not surprising, in that the unobserved patterns were all unilateral patterns, which occur relatively infrequently, and the total number of behaviors observed was small. All of the bilateral patterns observed by Sustare and Burtt were observed. The previously collected data showed all of the patterns observed by Sustare and Burtt, and no unidentified behaviors. The fly of unknown species escaped before more than 200 patterns had been observed and recording, but all of Sustare and Burtt’s (1976) bilateral patterns were observed. A hereto undocumented pattern was observed: the fly used both back legs to groom one of the wings; one leg was placed on the ventral surface, and one on the dorsal surface of the wing.

Discussion

With large numbers of captive flies, usually a large number are grooming. It takes a relatively short period of time to collect a large number of grooming patterns. Considering that the presence of other flies seems to affect grooming, future comparisons should take into account whether flies were grouped or solitary. It is probably that there are differences in grooming at different times in a fly’s activity cycle; for instance, there seems to be considerable grooming after feeding. There were not a sufficient number of flies available to verify Sustare and Burtt’s (1976) ethogram, although all behaviors observed in house flies were documented by Sustare and Burtt.

The previously undocumented pattern of the large fly of unknown species suggests that a comparative study of house flies and their relatives might prove fruitful.