A study of shaggy mutant clones on the notum reveals that a greater number of cells are diverted into the bristle pathway of differentiation and fewer cells remain to produce the epidermis, shaggy clones differentiate supernumerary microchaetae and macrochaetae but these are found in the correct spatial locations, e.g. clusters of macrochaetae are formed round the position of the extant macrochaetae. The shaggy mutant phenotype requires the functioning of the genes of the achaete-scute (AS–C) complex but a dosage study shows that it is unlikely that the AS–C is overexpressed in shaggy cells. Data are presented that argue, also, for a correct spatial expression of the AS–C in shaggy mutants. A study of clones doubly mutant for shaggy and different achaete and scute alleles is consistent with the hypothesis that the clusters of macrochaetae formed by shaggy represent the restricted spatial domains of expression of the AS–C. The results can be reconciled with the known role for the AS–C, in determining which bristle types differentiate where, and a role for shaggy in the cell interactions, within domains of the AS–C expression, leading to the definition of only one bristle mother cell.

A rather precise pattern of sensory bristles develops on the notum of Drosophila. Exactly eleven large bristles, or macrochaetae, each of which has been named, are found at defined positions on each heminotum. They are thought to develop as a response of the cells to their position within the field of positional information. The small bristles, or microchaetae, on the other hand, are indistinguishable from one another and their number varies slightly from fly to fly. They are, however, regularly spaced out and the spacing is thought to result from a phenomenon of lateral inhibition (Wiggles-worth, 1940; Richelie & Ghysen, 1979; Held & Bryant, 1984).

The development of both macro- and microchaetae is known to be under the control of genes of the achaete-scute complex (AS–C). achaete is required mainly for the formation of the microchaetae, whereas scute is required mainly for the formation of the macrochaetae (Garcia-Bellido, 1979). The functions of these two genes, however, overlap considerably and the full pattern requires both. Different alleles of these genes result in the loss of specific bristles and there exist also hypermorphic alleles, Hairy wing, that result in overexpression and cause the production of supernumerary bristles (Campuzano et al. 1986; Garcia-Alonso & Garcia-Bellido, 1986; Balcells et al. 1988). Two other, unlinked, genes, extra macrochaetae and hairy, are thought to act as repressors of scute and achaete, and loss of function of either of these genes also results in the production of either supernumerary macro- or microchaetae (Moscoso del Prado & Garcia-Bellido, 1984). It is thus thought that the AS–C is involved in the decision to differentiate bristles in defined places (Garcia-Bellido & Santamaria, 1978). The spatial restriction of the AS–C expression would thus be the main factor that controls the basic pattern of sensory bristles [as well as other sensory organs and also some central neurones (Dambly-Chaudière & Ghysen, 1987; Garcia-Bellido & Santamaria, 1978; Cabrera et al. 1987; Romani et al. 1987)].

A molecular analysis of the AS–C (Carramolino et al. 1982) led to the description of several transcripts only two of which correspond to achaete and scute, respectively (Campuzano et al. 1985). Many achaete and scute alleles are rearrangements with break points scattered throughout the complex and mapping some distance away from the coding regions. A comparison of the position on the map of a number of terminal deficiencies and the mutant phenotype led Ruiz-Gomez & Modolell (1987) to suggest that the scute transcript is regulated by cis-acting site-specific elements that respond to topological cues. Therefore, different control sites would activate scute in each precise area where a bristle will form. Presumably this area will involve a small region of the epidermis that is nevertheless considerably bigger than simply the one cell destined to become the macrochaete in question (Ghysen & Dam-bly-Chaudière, 1988; Stern, 1954a,b) and, accordingly, Romani, Campuzano, Macagno & Modolell (in preparation) observed by in situ hybridization, expression of achaete and scute in groups of cells in the imaginal wing disc.

Here we make use of the mutation shaggy to explore the spatial domains of expression of some scute alleles on the notum. Clones of cells mutant for shaggy transform epidermal cells into bristles and thus produce vast numbers of both macro- and microchaetae with little or no spacing between them. Different types of bristles, however, are produced in different locations and it has been shown that, although shaggy cells are unable to make a correct decision between an epidermal cell pathway and that of a sensory bristle, they are nevertheless able to respond correctly to positional cues (Simpson et al. 1988). On the notum, shaggy cells produce clusters of macrochaetae around the position of the extant ones. Here we show that the AS-C is required for this mutant phenotype but that achaete and scute are probably not overexpressed and that the spatial expression of these genes remains normal. A study of double mutant clones reveals that the clusters of macrochaetae formed by shaggy clones correspond to the domains of expression of different AS–C alleles. We also show that shaggy acts synergistically with extra macrochaetae, and may indirectly influence the expression of the AS–C via extra macrochaetae.

Flies were raised on standard medium and maintained at 25 °C

Lethal mutations at the zw3 locus were described by Judd et al. (1972) and map to 3B1. We have called this gene l(1)zw3sgg(shaggy) (Simpson et al. 1988). Throughout this report we have used the amorphic allele sggD127 (isolated by P. Ripoil), which we will refer to simply as sgg. In one single experiment, another allele, sggb12 (caused by an inversion, Judd et al. 1972), was also used and this is specified in the text (Table 5). For a description of other mutations and rearrangements see Lindsley & Grell (1968) and Lindsley & Zimm (1985), DIS vols 62, 64 and 65.

Clonal analysis

Clones mutant for sgg were produced in various genetic backgrounds by X-ray-induced mitotic recombination. Unless otherwise specified in the text, 24h egg collections were made and flies were irradiated between 48 and 72 h AEL. The clones resulting from mitotic recombination were marked with yellow (y), forked (f36a), javelin (jv), multiple wing hair (mwh) or stubbly chaete (stc). Flies were irradiated with 1000R of X-rays (100kV, 4mA given for 3min 18s, 1·5mm aluminium filter, Philips MG102 constant potential X-ray system, beryllium window). Thoraces were heated in 10% KOH and mounted between coverslips in Euparal. Clones were drawn onto standard diagrams of the notum. Clones were induced in flies of the following genotypes:

  1. sgg f36a/y

  2. y sgg f36a/+

  3. Df(1)sc19, y-sgg f36a/+, mwh jv/ +

  4. sc1sgg f36a/y sc1v (5) sc1sgg f36a/y (6) y ac1sgg w f36a/ac3;stc/+

  5. y ac1sgg w/f36a

  6. y sgg f36a/ +; emc1/TM2, emc

  7. y sgg f36a/ + ;emc1/ +

  8. y sgg f36a/+;+/+

  9. y sgg f36a/+;h1/h1

  10. y sgg f36a/+;h1/+

  11. Df(1)sc19,y-sgg f36a/+;Dp(1,2)sc19,y+/+

  12. y sgg f36a/+;CyO/+

  13. y sgg f36a/+;CyO/+

  14. sggf360/+; Dp(1; 2)sc19, y+/+

  15. Dp(3; Y; 1)M2, emc+y sgg w v/ +

  16. sggf36a/+

Flies of genoptypes (8), (9) and (10) were obtained from the cross ♀ y sgg f36a/FM6; TM2, emc/+× ♂ eme1/+. TM2 carries a weak allele of emc that is viable over emc1 (Moscoso del Prado & Garcia-Bellido, 1984). eme1/TM2 flies therefore have an emc phenotype. Flies of genotype (9) can be distinguished from those of genotype (10) by virtue of the synergism between sgg and emc (see Results); they invariably display at least one extra macrochaete on the thorax. Flies of genotypes (11) and (12) were obtained from the cross ♀ FM6/y sgg f36a’, h1/ + × ♂ h1/h1. Flies of genotypes (13) and (14) were obtained from ♀ Df(1)sc19, y-sgg f36a/FM6 and y sgg/FM6 that were both crossed to Dp(1; 2)sc19, y+/CyO males and placed in the same bottles. Flies of genotypes (15) and (16) were obtained from ♀ y sgg f36a/FM6 × Dp(1; 2)sc19, y+/CyO. Flies of genotypes (17) and (18) were obtained from ♀ FM7/Dp(3; Y; 1)M2, emc+y sgg w v and ♀ sgg f36a/FM6 that were both crossed to Ore R males and grown in the same bottles.

Gene dosage studies

For the study of the phenotype of emc1/TM2, emc flies bearing different doses of sgg+ (Table 6), the following cross: ♀ sgg w/y w, TM2/ + × ♂ y w/Y; Dp(1; 2)w+70, sgg+/+; eme1/ +, yielded flies of genotypes (19) sgg w/y w; +/+; eme1/TM2, emc, (20) sgg w/y w; Dp(1; 2)w+70, sgg+; emc1/TM2, emc, and (21) y w/y w; Dp(1; 2)w+70, sgg+/+; emc1/TM2, emc. These were all distinguishable, flies of genotype (19) are w, those of genotype (20) are w+, those of genotype (21) are y. The frequency of recombination between y and sgg is of the order of only 1 %.

For the study of the phenotype of flies heterozygous for both sgg and emc and bearing different doses of achaete-scute (Table 7), two crosses were performed. From ♀ Df(1)260-1, y-/+; eme1/TM2, eme × ♂ y sgg f36a/Dp(1; Y)w+sgg+, flies of genotypes (22) and (23) were generated: (22) Df(1)260-1, y/y sgg J1, emc1/ F, (23) y sgg f36a/ +; emc1/ +. From ♀ FM6/y sgg f36a ; Dp(1; 2)sc19/CyO × ♂ eme1/TM2, flies of genotypes (24), (25) and (26) were obtained: (24) y sgg f36a/+-CyO/+-,emc1/+, (25) y sgg f36a/+, Dp(1;2)SC19/+;emc1/ +, (26) FM6/+;, Dp(1; 2)sc19, y+/+; emc1/+.

shaggy mutant clones reveal a precise pattern of bristles on the notum

Clones of cells mutant for shaggy (sgg) were produced on the dorsal thorax by mitotic recombination. These produce dense clusters of extra bristles (see Figs 1,2). A study of sgg f36a clones and y control clones issued from the same recombination event (twin spots) at 72 h AEL shows that about twice as many bristles are made by the mutant cells (Table 1). The surface area occupied by the sgg clones is not enlarged, however. Rather, the spacing between the bristles is reduced and fewer trichome-producing epidermal cells are present (Table 2). Therefore a greater number of cells are diverted into the pathway for sensory bristles, and fewer cells are left to make the epidermis. In the area surrounding the sgg clones, as on the remainder of the thorax, the pattern of bristles remains normal. Control clones produced marked bristles in normal positions and caused no disruption of the pattern.

Table 1.

Size of twin clones resulting from irradiation-induced mitotic recombination at 72 ±2h AEL in sgg f36a/y flies

Size of twin clones resulting from irradiation-induced mitotic recombination at 72 ±2h AEL in sgg f36a/y flies
Size of twin clones resulting from irradiation-induced mitotic recombination at 72 ±2h AEL in sgg f36a/y flies
Table 2.

The spacing between bristles of y sgg f36a clones induced in y sgg f36a/ + flies measured by counting the number of intervening trichomes. As a control, spacing was measured between wild-type bristles in the same position as the clone on the other hemithorax

The spacing between bristles of y sgg f36a clones induced in y sgg f36a/ + flies measured by counting the number of intervening trichomes. As a control, spacing was measured between wild-type bristles in the same position as the clone on the other hemithorax
The spacing between bristles of y sgg f36a clones induced in y sgg f36a/ + flies measured by counting the number of intervening trichomes. As a control, spacing was measured between wild-type bristles in the same position as the clone on the other hemithorax
Table 3.

Size of shaggy clones in flies bearing one, two or three doses of AS–C, irradiated between 48 and 72h AEL

Size of shaggy clones in flies bearing one, two or three doses of AS–C, irradiated between 48 and 72h AEL
Size of shaggy clones in flies bearing one, two or three doses of AS–C, irradiated between 48 and 72h AEL
Table 4.

Numbers of macrochaetae and microchaetae found in sgg clones produced in flies carrying <1, 1, 2 or 4 copies of emc+ by irradiation-induced mitotic recombination at 48–72 h AEL

Numbers of macrochaetae and microchaetae found in sgg clones produced in flies carrying <1, 1, 2 or 4 copies of emc+ by irradiation-induced mitotic recombination at 48–72 h AEL
Numbers of macrochaetae and microchaetae found in sgg clones produced in flies carrying <1, 1, 2 or 4 copies of emc+ by irradiation-induced mitotic recombination at 48–72 h AEL
Table 5.

Numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc

Numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc
Numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc
Table 6.

Numbers of macrochaetae found on the thorax of eme1 /TM2, emc flies bearing one, two or three doses of sgg+

Numbers of macrochaetae found on the thorax of eme1 /TM2, emc flies bearing one, two or three doses of sgg+
Numbers of macrochaetae found on the thorax of eme1 /TM2, emc flies bearing one, two or three doses of sgg+
Table 7.

The numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc and bearing one, two or three doses of AS–C

The numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc and bearing one, two or three doses of AS–C
The numbers of macrochaetae found on the thorax (including the humerus) of flies heterozygous for both sgg and emc and bearing one, two or three doses of AS–C
Fig. 1.

(A) Standard diagram of the wild-type heminotum of Drosophila showing the positions of macrochaetae, large circles, and microchaetae, small circles. The macrochaetae are named as follows: DC, dorsocentral; Scu, scutellar; PA, postalar; SA, supraalar; NP, notopleural; PST, presutural. (B) The distribution of macrochaetae, large dots, observed in a study of 100 y sgg f36a clones induced in y sgg f36a+ flies. The flies were irradiated between 48 and 72 h AEL, the average clone size was 16·4 ± 2·0 bristles, the largest clone comprised 69 bristles. Clones are a mixture of micro- and macrochaetae, only the macrochaetae are shown. These can be seen to be clustered around the positions of the extant macrochaetae. More bristles are found in the positions of the dorsocentral and scutellar bristles than in the positions of the other macrochaetae. It is thought that some clones in these regions may be lost, see text. Examination of a much larger sample of clones, however, shows that macrochaetae also arise in clusters around the postalar, supraalar, notopleural and presutural bristles (results not shown). (C) The distribution of microchaetae, small dots, observed from a study of the same 100 y sgg f36a clones as in B. 15 clones are shown. The bristles, which develop at a high density, can be seen to differentiate in areas normally covered by microchaetae. No microchaetae are found on the scutellum or in the vicinity of the posterior postalar bristle.

Fig. 1.

(A) Standard diagram of the wild-type heminotum of Drosophila showing the positions of macrochaetae, large circles, and microchaetae, small circles. The macrochaetae are named as follows: DC, dorsocentral; Scu, scutellar; PA, postalar; SA, supraalar; NP, notopleural; PST, presutural. (B) The distribution of macrochaetae, large dots, observed in a study of 100 y sgg f36a clones induced in y sgg f36a+ flies. The flies were irradiated between 48 and 72 h AEL, the average clone size was 16·4 ± 2·0 bristles, the largest clone comprised 69 bristles. Clones are a mixture of micro- and macrochaetae, only the macrochaetae are shown. These can be seen to be clustered around the positions of the extant macrochaetae. More bristles are found in the positions of the dorsocentral and scutellar bristles than in the positions of the other macrochaetae. It is thought that some clones in these regions may be lost, see text. Examination of a much larger sample of clones, however, shows that macrochaetae also arise in clusters around the postalar, supraalar, notopleural and presutural bristles (results not shown). (C) The distribution of microchaetae, small dots, observed from a study of the same 100 y sgg f36a clones as in B. 15 clones are shown. The bristles, which develop at a high density, can be seen to differentiate in areas normally covered by microchaetae. No microchaetae are found on the scutellum or in the vicinity of the posterior postalar bristle.

Fig. 2.

(A) A y sgg f36a clone induced in a y sgg f36a/+ fly after irradiation between 48 and 72 h AEL. Both macrochaetae and microchaetae have been formed. The macrochaetae can be seen to be clustered around the dorsocentral and the posterior scutellar bristles. No microchaetae have been formed on the scutellum. (B) A sgg f36a clone induced in a sgg f360/ y fly after irradiation between 48 and 72 h AEL. A group of microchaetae are seen in the anterior part of the thorax. In this region sgg clones never differentiate macrochaetae. (C) A y sgg f36a clone induced in a y sgg f360/ + ; h1/h1 fly after irradiation between 48 and 72 h AEL. In the mutant h1 background, sgg clones are able to differentiate microchaetae on the scutellum. (D) A sc1sgg f36a clone induced in a sc1sgg f36a/y fly after irradiation between 48 and 72h AEL. A marked posterior dorsocentral bristle and a cluster of microchaetae are found juxtaposed to a scutellum which is lacking the anterior scutellar bristle. The absence of this bristle is thought to be a consequence of cells of the mutant clone extending over the scutellum: the sc1 mutant, which lacks scutellar bristles, is epistatic in phenotype over that of sgg.

Fig. 2.

(A) A y sgg f36a clone induced in a y sgg f36a/+ fly after irradiation between 48 and 72 h AEL. Both macrochaetae and microchaetae have been formed. The macrochaetae can be seen to be clustered around the dorsocentral and the posterior scutellar bristles. No microchaetae have been formed on the scutellum. (B) A sgg f36a clone induced in a sgg f360/ y fly after irradiation between 48 and 72 h AEL. A group of microchaetae are seen in the anterior part of the thorax. In this region sgg clones never differentiate macrochaetae. (C) A y sgg f36a clone induced in a y sgg f360/ + ; h1/h1 fly after irradiation between 48 and 72 h AEL. In the mutant h1 background, sgg clones are able to differentiate microchaetae on the scutellum. (D) A sc1sgg f36a clone induced in a sc1sgg f36a/y fly after irradiation between 48 and 72h AEL. A marked posterior dorsocentral bristle and a cluster of microchaetae are found juxtaposed to a scutellum which is lacking the anterior scutellar bristle. The absence of this bristle is thought to be a consequence of cells of the mutant clone extending over the scutellum: the sc1 mutant, which lacks scutellar bristles, is epistatic in phenotype over that of sgg.

Clones of sgg cells produced earlier than 72 h AEL show a tendency to round up and contract. Sometimes vesicles of cuticle bearing marked bristles are found inside the thoraces that carry sgg clones. This is perhaps due to different cell affinities such as those described for some areas of the wing (Ripoll et al. 1988). Differential cell affinities may also be the reason why many sgg clones are found in the centre of the thorax and fewer clones are found in the region of the postalar, supraalar and notopleural bristles (see Fig. 1B). Cell proliferation is probably unaffected in sgg mutant cells: a study of twin spots on the wing blade where accurate cell measurements can be made revealed clones of expected sizes (Simpson et al. 1988).

Both extra microchaetae and macrochaetae are formed in shaggy clones. They develop, however, at fairly precise locations. The positions of micro- and macrochaetae from 100 y sgg f36a clones were drawn onto standard diagrams. On areas of the notum where microchaetae normally differentiate, sgg clones too produce microchaetae (Figs 1C, 2B). Microchaetae, therefore, although formed at a greater density, develop in those parts of the thorax that are normally covered with them. Microchaetae are not formed in ectopic positions, e.g. no microchaetae differentiate on the scutellum. Therefore, the basic spatial distribution is unaltered. In the regions surrounding the positions of the extant macrochaetae sgg clones produce both macrochaetae and microchaetae. Macrochaetae, however, are only found clustered around the positions of the extant macrochaetae (Figs 1B, 2A), they do not differentiate elsewhere. This is particularly clear for the dorsocentral and the scutellar bristles, to which we will confine the rest of our analysis. Therefore as with the microchaetae, the basic spatial distribution of macrochaetae is unaltered in the sgg mutant. These observations therefore suggest that, although extra bristles are formed by sgg cells, the basic spatial organization that dictates which type of bristle is to be placed where remains unaltered. It has in fact been shown that sgg cells on the wing blade know where they are and respond correctly to positional cues (Simpson et al. 1988).

The shaggy phenotype requires the function of achaete-scute but is not attributable to an overexpression of achaete-scute

The decisions governing which type of bristles are to be placed where on the notum are thought to fall under the control of the genes of the achaete-scute complex (AS–C). In the absence of the AS–C no bristles form (Garcia-Bellido & Santamaria, 1978). The Hairy wing (Hw) alleles of the AS–C represent a gain of function. They cause an overexpression of either achaete (ac) or scute (sc) and lead to the production of extra micro- or macrochaetae (Balcells et al. 1988). Their phenotype superficially, therefore, resembles that of sgg. (There are, however, important differences that will be discussed later). It was therefore of interest to determine whether or not the AS–C is expressed normally in sgg mutant cells or whether the sgg phenotype could be attributed to an overexpression of the AS–C. To address the question of whether the AS–C is required for the phenotype of sgg, clones doubly mutant for sgg and a deletion of the AS–C were made. Clones were induced in y Df(1)sc19sgg f36a/+; mwh jv/+; flies [Df (1)260-1 was also used and gave identical results to those with Df(1)sc19, data not shown]. In the first experiment, 50 thoraces from a late irradiation (72 ± 4h AEL) were mounted and scored for the presence of clones. A total of 29 mwh jv clones were observed together with 25 naked patches presumed to be AS–C-sgg clones. In order to establish that this result pertains to all bristles of the notum, a further 34 naked patches, from flies irradiated between 48 and 72 h AEL, were selected under the dissecting microscope. Such presumed clones covered most of the area of the notum and no y f36a bristles were ever seen. The AS–C is therefore epistatic over sgg for the thoracic phenotype and the supernumerary bristles seen in sgg clearly require the wild-type achaete and scute functions. In order to test the hypothesis that ac and sc might be overexpressed in sgg mutant cells, a gene dosage study was performed. Flies carrying only one dose of sgg and one, two, three or four copies of the AS–C+ were constructed and found to display a wild-type appearance (results not shown). Furthermore, flies heterozygous for the AS–C and carrying one, two or three copies of sgg+ are also wild-type (results not shown). Finally the phenotype of Hw1, a gain-of-function allele of ac, remains unaltered in the presence of one, two or three doses of sgg+ (results not shown). As a further test, clones mutant for sgg were made in flies bearing one, two or three copies of the AS–C and their phenotype was found to be unchanged (Table 3). From these observations, we conclude that sgg plays no regulatory role concerning the expression of the AS–C. It therefore seems unlikely that the sgg phenotype is due to a derepression of the AS–C.

Not all of the bristles found in the adult fly require the function of the AS–C. For example, many of the bristles of the wing margin, medial and dorsal triple row bristles and bristles of the double row differentiate in marginal clones that are AS–C- (Garcia-Bellido & Santamaria, 1978). shaggy, on the other hand, affects all bristle types found on the fly body including those of the wing margin, such that sgg clones on the wing blade transform epidermal hairs into marginal bristles (Simpson et al. 1988). AS–C- clones on the wing blade (away from the wing margin) differentiate trichomes. It was therefore possible to test the epistatic relationship between sgg and AS–C for this part of the body. Doubly mutant AS–C-sgg clones on the wing blade have a sgg phenotype, that is they differentiate bristles (results not shown). Therefore the AS–C appears to be epistatic over sgg only in those regions of the body in which the AS–C is required. In other words, the sgg phenotype does not result from a derepression of ac or sc such that these genes are ectopically expressed. These results argue again that the spatial expression of AS–C elements is not altered in the sgg mutant, and this would therefore explain why on the notum, although there are more bristles than is usual in sgg clones, the bristles are nevertheless formed in the correct general positions, shaggy is perhaps therefore involved in some other aspect of bristle differentiation, and plays a different role from that of proneural genes such as the AS–C for the notum [and other, perhaps related, genes (Villares & Cabrera, 1987; Alonso & Cabrera, 1988) for the wing] that govern the spatial distribution of morphologically distinct bristles.

The shaggy pattern reveals the spatial domains of expression of different scute alleles

Stem (1954a,b) made a study of clones mutant for ac which removes the dorsocentral bristles. Such clones behave autonomously. When the mosaic border line runs through the dorsocentral region, however, occasionally a bristle will form at some distance from the normal site. This locally restricted nonautonomy led Stem to suggest that the macrochaete position is first defined as a region and later narrowed down to a single cell. Our observations on sgg clones are consistent with this idea; the clusters of extra macrochaetae seen in sgg suggest the presence of a small region around each extant bristle, the limits of which specify position for that bristle. This supports the notion that determination of a bristle results from a collective decision of a group of cells in a defined area of the epidermis. Such a group of cells can be called an equivalence group (Palka, 1986). It is then thought that cell interactions between members of the group will lead to the determination of only one bristle mother cell which then inhibits its neighbours from also becoming bristles by a mechanism of lateral inhibition (Wigglesworth, 1940). A similar mechanism appears to govern the segregation of neuroblasts in the embryo (Doe & Goodman, 1985).

A number of ac and sc alleles have been described that each remove a specific bristle or a subset of bristles on the thorax. Many of these alleles are due to chromosomal rearrangements with break points at some distance from the coding regions (Campuzano et al. 1985). The sc break points are thought to affect cri-acting site-specific control elements that govern the expression of the sc transcript (Ruiz-Gomez & Modolell, 1987). This would mean that each control site specifically activates sc in the area in which the bristle under its control is destined to be produced. It is unlikely that this activation of sc occurs precisely in the one cell that will become the bristle. Rather sc would be activated in all the cells of the equivalence group that is responsible for that bristle. In other words, the size of the equivalence group would correspond to the number of cells expressing sc. It has in fact been observed that sc transcripts are found in groups of cells in the areas of the imaginal disc where each bristle will later form (Romani, Campuzano, Macagno & Modolell, in preparation) . It has also been observed in the embryo that sc transcripts are found in groups of cells from which one will segregate as a neural precursor (Cabrera et al. 1987) .

If sc is thus expressed in small regions over the thorax and if, as we have argued, this regional control is unaltered in sgg, then the clustering of macrochaetae around the position of the extant bristles seen in sgg clones would represent the spatial domains of expression of sc. In order to explore this possibility, we have made clones simultaneously mutant for sgg and specific ac or sc alleles. We chose sc1, which removes the two scutellar bristles and the anterior notopleural, and ac1 and ac3, which remove the posterior dorsocentral bristle.

shaggy clones in sc+ flies can differentiate a cluster of macrochaetae on the scutellum as in Fig. 2A. The largest clone seen on the scutellum comprised 7 macrochaetae. In the first experiment, sgg clones were induced in animals mutant for sc1, (sc1sgg f36a/y sc1v flies were irradiated.) The distribution of macrochaetae is shown in Fig. 3A. No sgg f36a macrochaetae were found on the scutellum. sggf36a macrochaetae were still found clustered in the region of the dorsocentral bristles. In the second experiment, sc1sgg clones were induced in otherwise sc+ flies (sc1sgg f36a/y flies were irradiated). Fifteen sc1sgg f36a clones were observed on the nonscutellar regions of the thorax; they had a sgg phenotype. Twelve normal control y clones were also seen. Four y clones were found on the scutellum. Six cases of naked patches on the scutellum were observed, four of which removed one scutellar bristle while the two others removed both. One of these naked patches was associated with a y twin spot and one with a sgg f360 clone on the scutum (Fig. 2D). They thus presumably correspond to the sc sgg clones, sc1 is therefore only epistatic over sgg in that region of the thorax where the mutant phenotype is seen. Elsewhere the AS–C is functioning normally and the sgg mutant phenotype is expressed. Therefore, the domain of expression of scute for the scutellar bristles, the spatial control of which is abolished in the sc1 allele, extends only over the scutellum, but covers an area larger than that occupied by each of the two normal scutellar bristles.

Fig. 3.

(A) The distribution of macrochaetae (closed circles) from 90 sgg clones marked with f36a and induced in sc1sgg f36a/ y sc1v flies. Flies were irradiated between 48 and 72 h AEL and the average clone size was 15·5 ± 2·6 (n = 54). In sc1 flies, in which the scutellar bristles are absent, the sgg clones, which are also mutant for sc1, fail to differentiate bristles anywhere on the scutellum unlike the sgg clones induced in sc+ flies in Fig. 1B. Elsewhere the clones show the typical sgg phenotype. (B) The distribution of macrochaetae from 89 sgg clones marked with y f36aand induced in y ac1sgg w f36a/ac3; stc/ + flies. Flies were irradiated between 48 and 72 h AEL and the average clone size was 15·1 ± 2·1 (n = 58). Few clones produce macrochaetae in the region of the posterior dorsocentral bristle when compared with those differentiated by sgg clones in an ac+ background (Fig. 1B). The presence of occasional sgg bristles at or close to the site of the posterior dorsocentral bristle can be attributed to the hypomorphic nature of the ac1 and ac3 mutations; such bristles are often intermediate in size, see text.

Fig. 3.

(A) The distribution of macrochaetae (closed circles) from 90 sgg clones marked with f36a and induced in sc1sgg f36a/ y sc1v flies. Flies were irradiated between 48 and 72 h AEL and the average clone size was 15·5 ± 2·6 (n = 54). In sc1 flies, in which the scutellar bristles are absent, the sgg clones, which are also mutant for sc1, fail to differentiate bristles anywhere on the scutellum unlike the sgg clones induced in sc+ flies in Fig. 1B. Elsewhere the clones show the typical sgg phenotype. (B) The distribution of macrochaetae from 89 sgg clones marked with y f36aand induced in y ac1sgg w f36a/ac3; stc/ + flies. Flies were irradiated between 48 and 72 h AEL and the average clone size was 15·1 ± 2·1 (n = 58). Few clones produce macrochaetae in the region of the posterior dorsocentral bristle when compared with those differentiated by sgg clones in an ac+ background (Fig. 1B). The presence of occasional sgg bristles at or close to the site of the posterior dorsocentral bristle can be attributed to the hypomorphic nature of the ac1 and ac3 mutations; such bristles are often intermediate in size, see text.

sgg clones were also made in flies mutant for ac1/ac3. In such flies, the posterior dorsocentral bristles are lacking, ac1 and ac3 are hypomorphic mutations, however, and occasionally a posterior dorsocentral bristle will arise; very often this bristle is considerably smaller than normal, (y ac1sgg w f36a/ac3; stc/ + flies were irradiated.) The distribution of macrochaetae found in a study of 58 sgg clones is presented in Fig. 3B. Bristles are found around the positions of all macrochaetae except the posterior dorsocentrals where only one sgg bristle was observed. Interestingly this bristle was also intermediate in size. One control stc bristle was also found in this position from observations of 26 clones. ac1sgg clones were also produced in flies that were otherwise ac+(y ac1sgg w/f36a flies were irradiated). The distribution of macrochaetae from 54 sgg clones was very similar to that of the preceding experiment (not shown). Amongst these flies eleven cases of a missing posterior dorsocentral bristle were recorded, five of which were associated with either y microchaetae or with f36a control bristles. Seven cases off36a posterior dorsocentral bristles were found. We have shown, therefore, that the cluster of macrochaetae produced by sgg cells in the region around the posterior dorsocentral bristle is removed when these cells are also mutant for ac1 (or ac1/ac3) even though in ac flies only the single posterior dorsocentral bristle is missing. Occasionally in these flies, posterior dorsocentral bristles were observed at some distance from the normal site (one case of a y bristle, two of wild-type bristles and one case of a f36a bristle; see also Stem, 1954a,b).

These results, therefore, provide further evidence for the idea that the AS–C transcripts are expressed in small domains on the thoracic epidermis, domains that cover an area considerably greater than the single cell that will finally produce a bristle. In the sgg mutant, the process that normally leads to the singling out of only one bristle mother cell is somehow defective such that a group of bristles are formed in each domain. Therefore sgg permits us to visualize the extent of the domains, within which the AS–C transcripts are effective.

Unlike shaggy, hairy and extra macrochaetae may disrupt the spatial expression of achaete and scute

The genes hairy (h) and extramacrochaetae (emc) are thought to act as repressors of ac and sc on the notum (Moscoso del Prado & Garcia-Bellido, 1984). The phenotype of emc is somewhat similar to that of sgg: mutant clones of lethal alleles produce extra macro- and microchaetae (Garcia-Alonso & Garcia-Bellido, 1988). Viable alleles of emc result in the formation of extra macrochaetae (Moscoso del Prado & Garcia-Bellido, 1984). Flies mutant for h carry extra microchaetae. Similarly the gain-of-function Hw alleles also cause additional bristles. Hw49c, which causes an overproduction of ac and sc, leads to the differentiation of extramicro- and macrochaetae, whereas Hw1, which causes an overproduction of ac, leads to the development of extra microchaetae (Garcia-Alonso & Garcia-Bellido, 1986; Campuzano et al. 1986; Balcells et al. 1988). The phenotypes of all these mutants, however, differ fundamentally from that of sgg in that the supernumerary macrochaetae or microchaetae are found not only in the regions of notum where they would normally develop but also in ectopic positions. For example, in Hw1 flies microchaetae are found on the wing blade and, in h flies, microchaetae develop on the scutellum. Similarly in the case of emc, macrochaetae are found in more anterior positions on the thorax with a new distribution, the pattern of which is reminiscent of other dipteran species (Moscoso del Prado & Garcia-Bellido, 1984; Garcia-Alonso & Garcia-Bellido, 1988). In other words, unlike sgg, the spatial distribution of the morphologically distinct bristle types is altered in these mutants. It has also been observed that, in Hw1 and HW49C wing discs, ac and sc transcripts accumulate in ectopic locations (Balcells et al. 1988).

In order to further explore the difference in phenotype between the aforementioned mutants and sgg, sgg clones were produced in flies mutant for either emc or h. 27 sgg clones resulting from irradiation of y sgg f36a/+; eme1/TM2, emc flies were studied, as well as 30 sgg clones resulting from irradiation of y sgg f36a/+; +/ + sibling flies. The average clone sizes from these two genotypes were not significantly different (Table 4). However, the clones in emc flies were found to encompass a larger number of macrochaetae (Table 4). The most striking difference, however, is found in the distribution of these macrochaetae, see Fig. 4A. Ectopic macrochaetae are found clustered in a more anterior region of the notum, an area normally covered only by microchaetae. Therefore, in flies mutant for emc, the sgg phenotype is changed such that the spatial distribution of the additional macrochaetae is altered.

Fig. 4.

Distribution of macrochaetae differentiated by sgg clones in which the cells were also carrying a variable number of copies of emc3. The flies of the genotypes used in A, B and C were siblings. All flies were irradiated between 48 and 72h AEL. Data pertaining to numbers of clones observed and clone sizes can be obtained from Table 4. (A) Flies were of the genotype y sgg f36a/+; emc1/TM2, emc and show an emc mutant phenotype; they carry less than one dose of emc+. Macrochaetae are also seen on the anterior half of the thorax in an area that only bears microchaetae in emc+ flies. (B) Flies were of the genotype y sgg f360/+; emc1/ + and therefore only carry one copy of emc+. Consistent with experiments revealing the synergism that exists between sgg and emc, in these flies, too, supernumerary macrochaetae are observed on the anterior thorax. (C) Control flies that were of the genotype y sgg f36a/+; +/ + and were therefore diploid for emc+. Macrochaetae are only found clustered around the positions of the extant bristles as in Fig. 1B. (D) Flies were of the genotype Dp(1; Y; 3)M2 y emc+sgg w v/+; +/ + and therefore the y sgg clones in this experiment carry 4 copies of emc+. Fewer macrochaetae were formed by these clones, see Table 4, they are restricted to the dorsocentral and the scutellar bristle clusters. Control flies of the genotype sgg f36a/+; +/ + and therefore diploid for emc+ were grown in the same bottles as those carrying the duplication. The distribution of macrochaetae in these sgg clones was similar to that of Figs 1B and 4C, not shown.

Fig. 4.

Distribution of macrochaetae differentiated by sgg clones in which the cells were also carrying a variable number of copies of emc3. The flies of the genotypes used in A, B and C were siblings. All flies were irradiated between 48 and 72h AEL. Data pertaining to numbers of clones observed and clone sizes can be obtained from Table 4. (A) Flies were of the genotype y sgg f36a/+; emc1/TM2, emc and show an emc mutant phenotype; they carry less than one dose of emc+. Macrochaetae are also seen on the anterior half of the thorax in an area that only bears microchaetae in emc+ flies. (B) Flies were of the genotype y sgg f360/+; emc1/ + and therefore only carry one copy of emc+. Consistent with experiments revealing the synergism that exists between sgg and emc, in these flies, too, supernumerary macrochaetae are observed on the anterior thorax. (C) Control flies that were of the genotype y sgg f36a/+; +/ + and were therefore diploid for emc+. Macrochaetae are only found clustered around the positions of the extant bristles as in Fig. 1B. (D) Flies were of the genotype Dp(1; Y; 3)M2 y emc+sgg w v/+; +/ + and therefore the y sgg clones in this experiment carry 4 copies of emc+. Fewer macrochaetae were formed by these clones, see Table 4, they are restricted to the dorsocentral and the scutellar bristle clusters. Control flies of the genotype sgg f36a/+; +/ + and therefore diploid for emc+ were grown in the same bottles as those carrying the duplication. The distribution of macrochaetae in these sgg clones was similar to that of Figs 1B and 4C, not shown.

shaggy clones were also examined after irradiation of y sgg f36a/+;h1/h1 flies and of y sgg f36a/+; h1/ + control siblings. 68 experimental and 45 control clones were studied. The distributions of microchaetae from these two genotypes are shown in Fig. 5. It can be seen that, in the h flies, sgg clones on the scutellum can include microchaetae (see also Fig. 2C). Therefore, in flies mutant for h, the sgg phenotype is altered such that the spatial distribution of the supernumerary microchaetae is changed.

Fig. 5.

The distribution of microchaetae in sgg clones produced in sibling flies homozygous or heterozygous for the mutant h1. Flies were irradiated between 48 and 72 h AEL. (A) 28 clones from y sggf36a/ +; h1/h1 flies are shown from a study totalling 68 clones. The clones were found to encompass ectopic microchaetae on the scutellum. (B) 21 clones from y sgg f36a/ +; h1/ + sibling flies are represented from a total of 45 clones studied. The distribution of microchaetae resembles that seen in flies diploid for h+ as in Fig. 1C.

Fig. 5.

The distribution of microchaetae in sgg clones produced in sibling flies homozygous or heterozygous for the mutant h1. Flies were irradiated between 48 and 72 h AEL. (A) 28 clones from y sggf36a/ +; h1/h1 flies are shown from a study totalling 68 clones. The clones were found to encompass ectopic microchaetae on the scutellum. (B) 21 clones from y sgg f36a/ +; h1/ + sibling flies are represented from a total of 45 clones studied. The distribution of microchaetae resembles that seen in flies diploid for h+ as in Fig. 1C.

These results, therefore, show that the spatial distribution of the supernumerary bristles found in sgg, both macrochaetae and microchaetae, is altered in flies mutant for emc or h. In other words, emc and h are both epistatic over sgg. We have previously argued that the regular spatial pattern normally seen in sgg reflects an accurate spatial expression of the AS–C. One explanation, therefore, for the altered distribution of macro- and microchaetae seen in sgg emc or sgg h clones would be that in these two cases the spatial distribution of the AS–C is altered. This would mean that emc and h not only cause an overexpression of sc or ac (Moscoso del Prado & Garcia-Bellido, 1984) but in addition an ectopic expression. It has already been suggested that the segmentation genes (Niisslein-Volhard & Wieschaus, 1980), of which h is one, might be responsible for the correct spatial expression of AS–C (Cabrera et al. 1987; Ghysen & Dambly-Chaudière, 1988) . A recent study reveals a normal spatial expression of ac and sc transcripts in emc and h wing discs so these genes may affect a very late expression of the AS–C (Romani, Campuzano, Macagno & Modolell, in preparation).

shaggy acts synergistically with extra macrochaetae

Because of the similarity of their phenotypes, we examined the relationship between sgg and emc and h. A dosage study revealed no relationship between sgg and h (results not shown). On the other hand, sgg and emc were found to act synergistically: the double heterozygotes display additional macrochaetae on the thorax (Table 5). The phenotype of emc was in fact found to vary with the number of doses of sgg+: the number of additional macrochaetae on the thorax of eme1/TM2, emc flies decreases with extra doses and increases with fewer doses of sgg+ (Table 6). Similarly the phenotype of clones mutant for sgg is also altered in flies carrying one, two or four doses of emc+. Although the overall clone size does not vary, a greater number of macrochaetae are found in sgg clones in which the dosage of emc+ is reduced, whereas fewer macrochaetae were formed by sgg clones that also carried four copies of emc+ (Table 4). Furthermore, the spatial distribution of macrochaetae differentiated by the sgg clones is not only altered in flies that were mutant for emc, but also in those that were heterozygous for emc (Fig. 4B). Therefore to some extent, through their respective roles on bristle differentiation, sgg and emc can compensate for one another.

The synergism between sgg and emc appears to be due to overexpression of the AS–C since it is suppressed by haploidy of the AS–C and enhanced by triploidy of the AS–C (Table 7).

We have shown that the production of supernumerary bristles caused by the sgg mutant does not appear to result from a spatial derepression of the genes of the AS–C. This is in opposition to the mutants h and emc for which a gene dosage study revealed regulatory roles on ac and sc (Moscoso del Prado & Garcia-Bellido, 1984). The results are consistent with the idea that, whereas the AS–C specifies the general position of each bristle through small regionally restricted zones of expression, sgg acts through an independent cellular phenomenon that is concerned with cell interactions leading to the sponsorship of only one bristle mother cell within each zone. Under this hypothesis, therefore, each cluster of macrochaetae produced by sgg mutant cells reflects the normal geographical extent of the zones of expression of the AS–C transcripts.

The following picture for the respective roles of the AS–C, emc, h and sgg is consistent with published data and with that presented here. The AS–C products are expressed in small discrete regions over the epidermis of the thorax as a result of the repressing effects of emc and h. These zones of expression lead to the formation of equivalence groups within which sgg and probably other genes such as those of the neurogenic class (Campos-Ortega, 1985; Dietrich & Campos-Ortega, 1984) play a role in the cell-cell interactions that lead to the definition of one bristle mother cell. This cell would then prevent other cells of the group from becoming bristles and the range of lateral inhibition would extend over the radius of each equivalence group. When the amount of emc+ is reduced, the regional control of the AS–C is no longer so effective and the sc products appear in larger areas and also in some ectopic locations. Consequently the equivalence group becomes larger than the area over which the mechanism of lateral inhibition can function and extra bristles appear. The synergism between sgg and emc can therefore be explained as follows. A single dose of emc+ leads to a relaxation of the regional control of the AS–C but in general the effect is insufficient to lead to an abnormal pattern and the mechanism of lateral inhibition is still effective. When, however, the amount of sgg+ is simultaneously reduced in these cells, the cell interactions leading to the definition of only one bristle cell are also less effective and as a result of these two effects extra bristles are formed. Lowering the number of copies of the AS–C in such flies doubly heterozygous for emc and sgg, compensates for the less effective repression by emc and restores the normal bristle pattern.

We thank Pascal Heitzler, Alain Ghysen, Pedro Ripoil and the stock centers of Bowling Green and Umea for generously providing mutant stocks, Claudine Ackerman for technical assistance, Marc Bourouis and Juan Modolell for comments on the manuscript, our colleagues at the L.G.M.E. for helpful discussions and Pierre Chambón for providing excellent working conditions.

Alonso
,
M. C.
&
Cabrera
,
C. V.
(
1988
).
The achaete-scute gene complex of Drosophila melanogaster comprises four homologous genes
.
EMBO J.
7
,
2585
2591
.
Balcells
,
L.
,
Modolell
,
J.
&
Ruiz-Gomez
,
M.
(
1988
).
A unitary basis for different Hairy-wing mutations of Drosophila melanogaster
.
EMBO J.
7
,
3899
3906
.
Cabrera
,
C. V.
,
Martinez-Arias
,
A.
&
Bate
,
M.
(
1987
).
The expression of three members of the achaete-scute complex correlates with neuroblast segregation in Drosophila
.
Cell
50
,
425
433
.
Campos-Ortega
,
J. A.
(
1985
).
Genetics of early neurogenesis in Drosophila melanogaster. T.l.N.S.
8
,
245
250
.
Campuzano
,
S.
,
Carramouno
,
L.
,
Cabrera
,
C. V.
,
Ruiz-Gomez
,
R.
,
Villares
,
R.
,
Boronat
,
A.
&
Modolell
,
J.
(
1985
).
Molecular genetics of the achaete-scute gene complex of Drosophila melanogaster
.
Cell
40
,
327
338
.
Campuzano
,
S.
,
Balcells
,
L.
,
Villares
,
R.
,
Carramouno
,
L.
,
Garcia Alonso
,
L.
&
Modolell
,
J.
(
1986
).
Excess of function Hairy-wing mutations caused by gypsy and copia transposable elements inserted within structural genes of the achaete-scute locus of Drosophia
.
Cell
44
,
303
312
.
Carramouno
,
L.
,
Ruiz-Gomez
,
M.
,
Guerrero
,
M. C.
,
Campuzano
,
S.
&
Modolell
,
J.
(
1982
).
DNA map of mutations at the scute locus of Drosophila melanogaster
.
EMBO J.
1
,
1185
1191
.
Dambly-ChaudiÈre
,
C.
&
Ghysen
,
A.
(
1987
).
Independent subpattems of sense organs require independent genes of the achaete-scute complex in Drosophila larvae
.
Genes and Dev.
1
,
1297
1306
.
Dietrich
,
U.
&
Campos-Ortega
,
J. A.
(
1984
).
The expression of neurogenic loci in imaginai epidermal cells of Drosophila melanogaster
.
J. Neurogenet.
1
,
315
337
.
Doe
,
C. Q.
&
Goodman
,
C. S.
(
1985
).
Early events in insect neurogenesis. II. The role of cell interactions and cell lineage in the determination of neuronal precursor cells
.
Devi Biol. Ill, 206-219
.
Garcia Alonso
,
L. A.
&
Garcia-Bellido
,
A.
(
1986
).
Genetic analysis of Hairy-wing mutations
.
Roux’s Arch, devl Biol.
195
,
259
264
.
Garcia-Alonso
,
L. A.
&
Garcia-Beludo
,
A.
(
1988
).
Extramacrochaetae, a trans-acting gene of the achaete-scute complex of Drosophila involved in cell communication
.
Roux’s Arch, devl Biol.
197
,
328
338
.
Garcia-Beludo
,
A.
(
1979
).
Genetic analysis of the achaete-scute system of Drosophila melanogaster
.
Genetics
91
,
491
520
.
Garcia-Beludo
,
A.
&
Santamaria
,
P.
(
1978
).
Developmental analysis of the achaete-scute system of Drosophila melanogaster
.
Genetics
88
,
469
486
.
Ghysen
,
A.
&
Dambly-ChaudiÈre
,
C.
(
1988
).
From DNA to form: the achaete-scute complex
.
Genes and Dev.
2
,
495
501
.
Held
,
L. I.
&
Bryant
,
P. J.
(
1984
).
Cell interactions controlling the formation of bristle patterns in Drosophila. In Pattern Formation: A Primer in Developmental Biology
(ed. G. Malacinski &
S. V.
Bryant
).
New York
:
Macmillan
.
Judd
,
B. H.
,
Shen
,
M. W.
&
Kaufman
,
T. C.
(
1972
).
The anatomy and function of a segment of the X chromosome of Drosophila melanogaster
.
Genetics
71
,
139
156
.
Lindsley
,
D. L.
&
Grell
,
E. H.
(
1968
).
Genetic variations of Drosophila melanogaster. Carnegie Institution of Washington. Pubheation no. 627
.
Lindsley
,
D.
&
Zimm
,
G.
(
1985
).
The genome of Drosophila melanogaster
.
Drosophila Information Service,
Vols 62, 64,
65
.
Moscoso Del Prado
,
J.
&
Garcia-Beludo
,
A.
(
1984
).
Genetic regulation of the achaete-scute complex of Drosophda melanogaster
.
Wilhelm Roux’s Arch, devl Biol.
193
,
242
245
.
NÜsslein-Volhard
,
C.
&
Wieschaus
,
E.
(
1980
).
Mutations affecting segment number and polarity in Drosophila
.
Nature, Lond.
287
,
795
801
.
Palka
,
J.
(
1986
).
Neurogenesis and axonal path-finding in invertebrates
.
T.l.N.S.
8
,
482
485
.
Richelle
,
J.
&
Ghysen
,
A.
(
1979
).
Determination of sensory bristles and pattern formation in Drosophila. I. A model
.
Devl Biol.
70
,
418
437
.
Ripoll
,
P.
,
El Messal
,
M.
,
Laran
,
E.
&
Simpson
,
P.
(
1988
).
A gradient of affinities for sensory bristles across the wing blade of Drosophila melanogaster
.
Development
103
,
757
767
.
Romani
,
S.
,
Campuzano
,
S.
&
Modolell
,
J.
(
1987
).
The achaete-scute complex is expressed in neurogenic regions of Drosophila embryos
.
EMBO J.
6
,
2085
2092
.
Ruiz-Gomez
,
M.
&
Modolell
,
J.
(
1987
).
Deletion analysis of the achaete-scute locus of D. melanogaster
.
Genes and Dev.
1
,
1238
1246
.
Simpson
,
P.
,
El Messal
,
M.
,
Moscoso Del Prado
,
J.
&
Ripoll
,
P.
(
1988
).
Stripes of positional homologies across the wing blade of Drosophila melanogaster
.
Development
103
,
391
401
.
Stern
,
C.
(
1954a
).
Two or three bristles
.
Amer. Sci.
42
,
213
214
.
Stern
,
C.
(
1954b
).
Genes and developmental patterns
.
Proc. 9th Int. Gen.
1
,
355
369
.
Villares
,
S.
&
Cabrera
,
C. V.
(
1987
).
The achaete-scute gene complex of Drosophila melanogaster: Conserved domains in a subset of genes required for neurogenesis and their homology to myc
.
Cell
50
,
415
424
.
Wigglesworth
,
V. B.
(
1940
).
Local and general factors in the development of “pattern” in Rhodnius prolixus (Hemiptera.).
J. exp. Zool.
17
,
180
220
.