Heredity 54 (1985) 209-217 The Genetical Society of Great Britain Received 3 September 1984 Enzymatic and quantitative variation in European and African populations of simulans

P. Hyytia, Laboratoire de Biologie et Génétique Evolutives CNRS, 91 190 Gif sur Yvette, France; P. Capy, * Departmentof Biology, McMaster University, J. R. David and Hamilton, Ontario Canada L8S 4K1. *R. S. Singh

Allozyme polymorphism at 15 loci of D. simulans was studied in 7 natural populations from Europe, North and tropical Africa. Morphological traits were studied in nine European and eleven Afrotropical strains. Within a population, the biochemical polymorphisms of Drosophila simulans and its sibling are not very different, although D. simulans has a lower heterozygosity. Between-populations genetic differentiation is however much lower in D. simulans than in D. melanogaster. Several loci of D. simulans do exhibit latitudinal trends but these are relatively weak. For morphological traits, both species show an increase of size with latitude, but geographic variation is again less pronounced in D. simulans. Both species are native to tropical Africa and have colonised the rest of the world. During this process, D. simulans has undergone much less geographic differentiation than has D. melanogaster, so that the ecological success of the two species is not correlated with similarities in their genetic properties.

INTflODUCTION 1978; Stalker, 1980; Knibb et a!., 1981), allozyme variation (Kojima eta!., 1970; Band, 1975; Mettler Amongabout 2500 species known in the et a!., 1977 Triantaphyllidis et a!., 1980; 1982; drosophilid family (Wheeler, 1981), only 21 have Oakeshott et a!., 1981a, b; David, 1982; Singh et a subcosmopolitan status, in that they exist in at a!., 1982; Capy et a!., 1983; Anderson and Oake- least three non-adjacent biogeographic realms shott, 1984),.biometrical traits (Teissier, 1956; Tan- (David and Tsacas, 1981). All these species occupy tawy and Mallah, 1960; David et a!., 1976; 1977; their present geographic range as a result of Capy et a!., 1983) and physiological traits (David involuntary transport by man. Among the 21 and Bocquet, 1975; Allemand and David, 1976; species, the sibling D. melanogaster and D. David et a!., 1977; Cohet et a!., 1980; Parsons, simulans are especially interesting since they are l980b, C;Bouletreau-Merleet a!., 1982). Several the only two which possess large populations in authors argue that this extensive genetic differenti- both tropical and temperate conditions (David and ation is one cause of the demographic success of Tsacas, 1981). Many studies of the ecology, D. melanogaster under a variety of ecological con- ecophysiology and genetics of these domestic ditions. species have attempted to understand their popula- D. simulans is closely related -toD. tion dynamics, ecology and other reasons of which melanogaster and shows comparable ecological may underlie their demographic success (e.g., Par- success. We might hence expect a comparable sons, 1975; l980a; 1983; Parsons and Stanley, genetic differentiation among its populations. 1981). However, although D. simulans is far less studied Populations of D. melanogaster, are genetically it already appears that, for a variety of traits, it is highly differentiated, for chromosomal poly- less variable than D. melanogaster. For example, morphism (Ashburner and Lemeunier, 1976; natural populations are monomorphic for their Watanabe and Watanabe, 1977; Voelker et a!., chromosome structures (Ashburner and 210 P. HYYTIA, P. CAPY, J. R. DAVID AND R. S. SINGH Lemeunier, 1976); the species does not exhibit a systems in eich species. Moreover, we have also latitudinal dine for ethanol tolerance (David and indicated when alleles with identical mobilities Bocquet, 1975); its protein polymorphism, studied were observed in the two species. with 2-dimensional electrophoresis, is lower than Biometrical analysis was performed on 9 in D. melanogaster (Onishi et a!., 1982). Surpris- French and 11 Afrotropical strains. Each strain ingly few studies have been devoted to allozyme was founded by several (in most cases more than polymorphism in D. simulans (O'Brien and 10) females and kept in the laboratory as a mass Maclntyre, 1969; Berger, 1970; Triantaphyllidis, culture in bottles. Larvae were grown at 25°C at 1973; Steiner et a!., 1976; Triantaphyllidis et a!., low density on a killed yeast medium and for each 1980;Salam eta!., 1981; 1982; Cabrera eta!., 1982; strain, six different traits were measured on 30 Onishi eta!., 1982; Anderson and Oakeshott, 1984) different adults. These traits were fresh adult and most of these concern temperate popula- weight (taken a few hours after emergence), wing tions. and thorax lengths, sternopleural and abdominal D. melanogaster and D. simu!ans are now chaetae numbers and ovariole number in females known to be native to tropical Africa (Tsacas and (see David, 1979 for more details). Viability and Lachaise, 1974; David and Tsacas, 1981) and in duration of development were also measured for D. melanogaster, there is extensive genetic diver- each strain, starting from a sample of 300 eggs (see gence between the ancestral, Afrotropical popula- David et a!., 1976, for techniques). tions and the European ones. Here we compare the allozyme polymorphism of seven natural popu- lations of D. simulans from Europe, North and RESULTS tropical Africa (mainland and islands). Differences in metric characters are often easier to detect than Al/ozyme frequencies are differences in gene frequencies (e.g., Lewontin, Five of the 15 loci studied: ADH (alcohol dehy- 1984) and are generally thought to be more directly drogenase), FU (fumarase), HK-3 (hexokinase 3), subject to natural selection than are allozymes ME (malic enzyme) and 6-PGD (6 phosphoglu- (Kimura, 1983). Our analysis has hence been conate dehydrogenase) were monomorphic in all extended to include morphological comparisons populations. between French and Afrotropical populations of The most polymorphic loci (table 1) as shown D. simulans and D. me!anogaster. by average expected heterozygosities (table 2) are EST-C (esterase-C), EST-6 (esterase 6), ACPH (acidphosphatase), PGM (phospho-gluco- MATERIALSAND METHODS mutase), XDH (xanthinedehydrogenase), ALDOX (aldehydeoxidase) and HK-l Wildliving females were used to initiate isofemale (hexokinase 1). These results generally confirm lines. Two adult were electrophoresed from what was already found by previous authors with each line after a few generations (from I to 10) of the exception of HK- I (Kojima ci a!., 1970; Onishi laboratory culture. The populations sampled were cia!.,1982 and Cabrera ci a!., 1982) which was from Villeurbanne, France (20 lines), Antibes, previously found to be polymorphic (2 alleles) only France (23 lines), Athens, Greece (25 lines), Nas- at low heterozygosity (0.019) in a single Spanish rallah, Tunisia (42 lines), Brazzaville, Congo (39 population. ADH is generally monomorphic in D. lines), Mahé, Seychelles islands (23 lines) and simu!ans and we failed to find any rare alleles, Tananarive, Madagascar (26 lines). although such alleles have been reported from Methods of starch gel electrophoresis were other parts of the world: such as Texas (Kojima adapted from Shaw and Prasad (1970) and Ayala eta!., 1970); Hawaii (Steiner et al., 1976) and Spain cia!.,(1972), and 15 different gene-enzyme loci (Cabrera et a! 1982). a-GPDH is monomorphic were studied in each population. Alleles were in most Drosophila species (Lakovaara and characterised by their migration distance on the Keranen, 1980), although rare alleles in D. gel, the most common one receiving the arbitrary simulans, are known from Texas (Kojima et a!., value of 1 00. Expected heterozygosity was calcu- 1970), Hawaii (Steiner et a!., 1976) and Greece lated (assuming Hardy—Weinberg equilibrium) for (Triantaphyllidis ci a!., 1980). each locus and populations. Few enzyme loci are Table 3 shows that, in most populations, about mapped in D. simu!ans, However, in all cases, an half of the 15 loci sampled were polymorphic with homology could be established with an already a mean of 2 alleles per locus and an average identified locus of D. melanogasier and we have heterozygosity of 16 per cent. Differences between used the same nomenclature for describing enzyme populations are generally small, although more GEOGRAPHIC VARIATION IN D. SIMULANS 211

Table 1Frequency of alleles in populations of Drosophila simulans. Alleles marked *exhibitthe same mobility as alleles found in D. melanogaster

France Villeurb. Antib. GreeceTunisiaCongo Seychelles Madagascar Average Locus and Allele 457°N 436°N 38°N 35°6N 4°2S 38°S l87S ACPH: acid phosphatase 095 0300 0200 0250 0131 0087 0130 0039 0l50 100 0700 0800 0729 0851 0873 0870 0913 0829 106 . . • 0021 0018 0040 • 0048 0021 Het. 0420 0320 0406 0258 0229 0226 0163 0288±0096 ALDOX: aldehyde oxidase l•00 0744 0767 0910 0792 0967 l•000 0962 0880 103 0256 0233 0090 0208 0033 • 0038 0l20 Het. 0381 0357 0164 0329 0064 0 0073 0195±0158 EST-C: Esterase C 087 ...... • 0010 0001 089 ...... • 0010 0001 093 0138 0267 02l7 0274 0058 0239 0049 0175 096 0338 0289 0l52 0360 0026 0l41 0l86 0207 100 0475 0256 0522 0183 0377 0500 0363 0364 102 0025 O'l67 0087 017l 0507 0109 0324 0226 104 0025 0022 0022 0012 0032 0011 0059 0026 Het. 0640 0751 0649 0732 0596 0661 0723 0679±0057 EST-6 esterase 6 087 ...... • 0182 0207 0060 093 • 0067 • • 0020 • 0020 0014 0.96* 0350 0244 0292 0429 0442 0329 0340 0362 1.00* 0650 0689 0667 0544 0357 0427 0530 0537 103 .. . • 0042 0018 • 0037 0090 0024 104 ...... • 0020 0003 Het. 0455 0462 0468 0509 0644 0665 0595 0542±0090 a-GPDH: o-Glycerophos- phate deydrogenase 095 ...... 0 0 0013 . . . . . 0003 1.00* 1000 1000 1000 1000 0987 1000 1000 0997 Het. 0 0 0 0 0025 0 0 0004±0009 HK-l: hexokinase I 097 .. . . 0020 . 0036 0100 0173 0045 l.00* 0971 0964 0970 0963 0869 0837 0808 0915 103 0029 0036 0010 0037 0095 0063 0019 0040 Het. 0066 0069 0059 0071 0235 0285 0317 0157±0Il6 MDH-C: malate dehy- drogenase, cytoplasmic 1.00* 1000 1000 1000 1000 1000 1000 0990 0998 104 . 0010 0002 Het. 0 0 0 0 0 0 0020 0003±0007 ODH: octanol dehydrogenase 0.96* 0088 0011 0021 . . . . . • 0020 0015 1.00* 0912 0989 0979 1000 0981 1000 0970 0980 104 .. . .. 0 00 0019 . 0010 0005 Het. 0161 0022 0042 0 0037 0 0059 0046±0055 PGM: phospho-gluco-mutase 093 0038 0067 0135 0024 . 0109 . 0047 094 0125 0089 0010 0095 0039 . 0019 0054 1.00* 0838 0789 0854 0881 0890 0739 0952 0857 l06 0055 .. . . 0071 0152 0029 0042 Het. 0281 0362 0252 02l4 0201 0419 0092 0260±0108 XDH: xanthine dehydrogenase 098 0044 0214 0210 0106 0143 0175 0067 0135 100 0956 0786 0790 0894 0857 0825 0933 0865 Het. 0084 0336 0332 0l90 0245 0289 0125 0228±0099 212 P. HYYTIA, P. CAPY, J. R. DAVID AND R. S. SINGH

Table 2Comparison of polymorphism within and between populations for polymorphic loci of I). sirnulans. Fixation index is HT-HS/ HT for 8 highly polymorphic loci. Chi square values and numbers of degrees of freedom (in brackets) are given (all values are statistically significant, p <001)

Within popul. Total Fixation heteroz. heteroz. index Chi-square Loci (H5) (H!) (H! ''5 (df) H!

ACPI-I 0-288 0302 0-043 389 (6) ALDOX 0196 0-215 0-091 68-1 (6) EST-C 0-684 0-754 0094 2 14-3 (20) EST-6 0542 0-569 0048 I 164 (13) a-GPDH 0-004 0-004 0 — HKl* 0157 0-165 0053 4433 (3) MDH-c 0003 0003 0 — ODH 0-046 0047 0035 34-7 (6) PGM 0-260 ((271 0041 235 (6) XDH 0-229 0-236 0-033 205 (6) mean (10 polym. loci) 0241 0-253 0046 mean (all IS loci) 0161 0168 ((031 * Chi-squaretest done between the 4 temperate and the 3 tropical populations. numerous alleles were found in two of the tropical Chi-square test on the 8 highly polymorphic loci populations from Congo and Madagascar. shows that there is a highly significant geographic Previously published estimates for heterozy- heterogeneity (table 2). A more general method of gosity of D.simulansrangefrom 0059(Onishiet comparing the total heterozygosity to within popu- aL, 1982)to 0173 (Triantaphyllidis eta!., 1980). lation heterozygosity is given by Wright's fixation However, these variations largely reflect differen- index Fyi (table 2). The observed values are gen- ces in the number of loci sampled and the choice erally low, with an average of 0046 for the 10 of loci and figures for the same locus studied by polymorphic loci. As in D. me!anogaster(Singhet different authors, do not show great differences a!., 1982),F.ST is not correlated with average between populations. Although the proportion of heterozygosity. polymorphic loci in one Japanese population Although there is significant heterogeneity (Onishi eta!.,1982), is 029, all other observed among populations, there are no significant corre- values fall between 04l and 046, not very different lations between allele frequencies and latitude. from 052 obtained by ourselves. The average num- Table 1 does suggest that there is a weak latitudinal ber of alleles per locus varies from I 46 (Cabrera trend for four loci. For ACPH°95 and EST-6'°°, c/ a!., 1982) to 207 (Triantaphyllidis eta!., 1980), we note a decrease of allele toward the Equator. thus including the value of 200 found here. For ALDOX, the frequency of the 1 .00 allele How much genetic divergence is there between increases in tropical populations, and the HK-1 geographically distant populations of the cos- polymorphism of the 4 temperate populations is mopolitan specie D.simulans? Anhomogeneity significantly lower than is that of those in the

Table 3 Summary of allozyme polymorphism in various populations of Drosophilaciniulans

Villeurhanne Antihes (France) (France) Greece Tunisia Congo Seychelles Madagascar Average proportion polym. loci* 053 053 0-53 0-47 06(1 0-40 060 052 alleles per locus 1-80 1-93 2-00 187 213 187 240 2-00 average heterozygosity: 0-249 0269 0237 023(1 0228 0255 (>217 0-241 10 polymorphic loci average heterozygosity: all ((-166 0 179 0-158 0254 0152 0- 170 0- 145 0161 IS loci * Calculatedat I per cent level. GEOGRAPHIC VARIATION IN D.SIMULANS 213

Table 4 Genetic identities (above diagonal) and genetic distances (below diagonal) between geographical populations of D. simulans

Populations Anti. Vill. Gree. Tuni. Congo Seych. Madag. Antibes 0992 0992 0991 0971 0944 0982 Villeurbanne 0008 0993 0996 0977 0948 0985 Greece 0008 0007 0988 0980 0955 0987 Tunisia 0009 0004 0-012 0981 0947 0988 Congo 0029 0023 0019 0019 0954 0992 Seychelles 0057 0054 0046 0054 0048 0963 Madagascar 0018 0016 0013 0012 0008 0037 tropics. Finally, for PGM, an increase of heterozy- 0024 (n =21).Distances between the 4 temperate gosity with latitude is observed, if we exclude the and the 3 tropical populations are significantly very isolated Seychellian population. Similar (p

Table 5 Comparison of French and Afrotropical strains in Drosophila simulansandDrosophila melanogasferforvarious quantitative traits. Statistical significance: *(p <0-05), **(p < 0-01)

D. sitnulans D. tnelanogasfer

France Trop. Africa France Trop. Africa Traits Sex (n = 9) (n = 11) Difference (n = 30) (n 22) Difference

Fresh weight 9 10924±2-17 93-97±1-16 l527±2.33** 123-22±0-81 l03-37±l01 19.85±l28** (mg-102) 87-32± 1-83 73-63± 1-67 1369±947** 91-91 78-03±0-94 13-88± l17** Thorax length 9 105-72±0•46 102-04±0-62 368±080** 10761 10214±058 214±066** (mm-l02) d 9429±063 9039±064 3.90±0.90** 9523±039 88-74±0-46 6.49±0.60** Wing length 9 19729± 164 19123±254 6.06±3.19* 227-25±0-77 206-29±1-14 2096± l32** (mm-I02) c 17321 16871±211 4.50±2.60* 19806±079 17880±085 1926± l.l7** Sternopleural chaetae 9 21-06±0-24 1926±037 1.80±0.46* 1957±022 1741±020 216±0315* (number) 19-34±0-25 17-73±037 1.61±0.46* 18-44±022 16-24±0-21 220±0.32** Abdominal chaetae 9 41-37±0-68 39l0±0•68 2.27±0.97* 43•23±0-29 39-32±046 3-91 (number) d 32-45±0-60 3l77±062 068±087 35-47±030 3142±031 405±0.44** Ovarioles 9 3796±094 33-13±079 4-83± I.2l** 47-57±0-45 38-59±0-66 898±0-74 (number) Development 9 199-66±3-93 205-69±405 —603±672 20816±206 20746± 1-75 0-70±2-84 (hours) d 206-26±3-64 210-72±3-91 —446±543 213-22±2-13 21327± 1-57 005±728 Viability 9+d 73-97±4-80 70-97±2-85 3-00±5-35 73-43±2-24 67-60±2-61 5-83±3-44 214 P. HYYTIA, P. CAPY, J. R. DAVID AND R. S. SINGH than in European strains and, among II calculated ourselves, we obtain a mean of 012; a propor- differences, 8 are significant at the 005 level. These tion of polymorphic loci of 044; and an estimate results are similar to those already obtained in D. of mean number of alleles per locus of 18. These melanogaster (Table 5 and David et a!. 1977). estimates, albeit imprecise are comparable to those Viability and duration of development show found in other organisms (Powell, 1975; Nevo, only small and non significant differences between 1978). D. simulans is less variable than are some French and African strains; in D. melanogaster tropical Drosophila species but is not very different also these physiological traits do not discriminate in this respect from the average Drosophila. An between tropical and temperate populations. average of these parameters in D. rnelanogaster The contrast between temperate and tropical using the studies of Kojima et al., (1970), Langley strains of D. simulans can be illustrated by a prin- et a!., (1974), Band (1975), Triantaphyllidis et a!., cipal components analysis (fig. 1) on 5 biometrical (1980), Cabrera et al., (1982), Onishi eta!., (1982), traits available for both sexes (weight, wing and Singh et a!., (1982), provides an estimate of 015 for H, of 0'65 for proportion of polymorphic loci, and of I 8 for the number of alleles per locus. Although these comparisons should be extended Ma/es Females to more loci studied with identical techniques in the same laboratory, it does seem that while D. France melanogaster may have a slightly elevated hetero- France zygosity and a greater proportion of polymorphic loci, neither of the two sibling species are excep- tional in this respect within the genus Drosophila. If we extend the comparison to single loci, we find that while D. simu!ans is monomorphic or almost Africa monomorphic for loci which are highly polymor- phic in D. melanogaster (such as ADH, a-GPDH First axis and G6PD) several loci (such as EST-C and ACPH) are more variable in D. simulans. Figure1 Resultof a principal components analysis applied to 5 biometrical traits of French and Afrotropical strains ACPH, ALDOX,EST-6and HK-l in D. of Drosophilaimulans. Eachellipseiscalculated for simulans do apparently show some tendency to including 75perdent of strains. The two lirst axes restitute vary with latitude (although this work needs to 92 per cent of the total information. be extended by the study of more populations on different continents) and only the EST-6 thorax lengths and chaetae numbers). The two first (Anderson and Oakeshott, 1984) data are at all axes constitute 92 per cent of the total variation. satisfactory. Genetic distances are smaller be- The first is mainly correlated to length characters tween populations living in the same geographic (thorax and wing) and the second to chaetae num- area with the extension of the isolated Seychelles bers. Males are clearly separated from females population. because of their small size and geographic dis- There is a clear contrast between D. crimination is better for females than for males. melanogaster and D.si,nulans: D. sirnu!ans is far less geographically diversified than is its sibling. Although it is difficult to compare "D" in 2 species DISCUSSION with different sets of loci and different geo- graphical patterns of sampling, the average value Howmuch polymorphism and geographic differ- of D in D. simu!ans is about 4 times lower than entiation is there in D. simulans and how do these in D. melanogaster (Singh el a!., 1982) (see fig. 2). data compare to those already obtained from its The genes studied in D. simu!ans are a subset of sibling cosmopolitan species D. melanogaster'? those studied in D. me!anogasier and its popula- Publishedresults on allozyme polymorphism tions were sampled from much less diverse places. for D.simulans are quite diverse. The level of However, the lower variability of D. simu!ans is heterozygosity is strongly influenced by the choice confirmed even if we consider only the case of of loci and by the number of loci studied. However, single loci and examine the values of their fixation if we average the values of Kojima ci a!., (1970), index. In D. me!anogaster, 17 loci out of 25 had Steiner eta!., (1976), Triantaphyllidis eta!., (1980), values superior to 01 and the average was 0l64 Cabrera et a!., (1982), Onishi et a!., (1982) and (Singh ci a!., 1982). Tn I). simulans no fixation GEOGRAPHIC VARIATION IN 0. SIMULANS 215

A direct comparison of phenotypic divergence Simulans D=0024 in the two species is not possible since D. simulans () is smaller than D. melanogaster. We use a relative 0a) C measure: for each trait, each sex and each species the ratio of the difference to the average value of the trait in French and African populations is 0 calculated. This index is superior for D. .0a) Melanogaster D = 0 105 melanogaster in 10 out of 11 cases. The only excep- E5 2 tion concerns male weight for which the changes are almost identical in the two species. If we com- pare these proportions by a simple statistical test, Hi1TrHTh1fl-kfl for example Chi square, we get a value of 736 005 01 015 02 with 1 degree of freedom (p <0.01). We can con- Genetic distance clude that for morphology, D. simulans is less Figure 2 Distribution of genetic distances calculated between geographically variable than is D. melanogaster, geographic populations of D. simulans (this work) and D. although the contrast is less pronounced than for melanogaster (after Singh et a!., 1982). D: average value aflozymes. of genetic distance. For D. simulans, dotted values corre- D. simulans is monomorphic for chromosome spond to distances including the Seychelles population. structure, does not exhibit a latitudinal dine for ethanol tolerance and is less variable than D. index has a value superior to 0• 1 and the average melanogaster for itsprotein polymorphism over 10 polymorphic loci gives a value of 0046. measured by 2-dimensional gel electrophoresis. In D. melanogaster, the greatest genetic contrast is However, there do exist parallel latitudinal dines between Afrotropical and .temperate populations in the two species in Australia for tolerance to (David, 1982; Singh et al., 1982) and nothing desiccation (Parsons, 1980b, c), and D. simulans analogous exists in D. simulans, in spite of the fact from Africa appears to differ from European popu- that Afrotropical populations are probably ances- lations in the composition of cuticular hydrocar- tral in both species. bons (Luyten, 1982). In addition pattern of change There is an increase in average values of in mitochondrial DNA suggests that D. simulans morphological traits in European populations of is divided into three geographic groups, one found D. simulans. The changes concern fresh weight, in Madagascar, a second in Seychelles, New thorax length, sternopleural and abdominal Caledonia and Hawaii, and a third in the rest of chaetae numbers, and ovarioles in females, and the world (Baba-Aissa and Solignac, 1984). are similar to those of D. melanogaster (table 5). It was expected 20 years ago (Baker and Steb- This evolutionary parallelism is illustrated by a bins, 1965) that there may be some common genetic discriminant analysis of 6 morphological traits of properties of colonising species. Our results show females (fig. 3). that this suggestion is far from true. D. sirnulans and D. melanogaster are closely related and are the most efficient colonisers in the genus Drosophila. However, the genetic structures of geo- Africa graphic populations are different; for most traits, Africa (including protein polymorphism, chromosome structure, alcohol tolerance and morphology) D. simulans is much less differentiated than is D. melanogaster. It is not yet known whether the two species do show geographical parallelism in other 0. Me/anogaster traits which might result from the action of natural France selection. France

Ffr5t dXS Figure 3 Result of a discriminant analysis applied to 6 bio- Acknowledgments We are grateful to Drs D. Lachaise, B. metrical traits of females of D. melanogaster and D. Pintureau, S. Tsakas and J. Vouidibio for providing some simulans. Each ellipse includes 75 per cent of strains. The D. simulans populations as well as to Mrs De Scheemacker- two first axes restitute 99 per cent of the total information. Louis and E. Pla for doing the biometrical measurements. 216 P. HYYTIA, P. CAPY, J. ft DAVID AND R. S. SINGH

REFERENCES within the Drosophilid family (Diptera). CR.Soc. biogeog., 5711—26. KIMURA, M. 1983.TheNeutral Theory of Molecular Evolution. ALL.F3MAND, R. ANI) DAVID.J. R. 1976. The circadian rhythm CambridgeUniv.Press, Cambr., 367 pp. ofovipositionin Drosophila melanogaster: A genetic KNIBB, w. p.. 1982.Chromosomeinversion polymorphisms in latitudinal dine in wild populations. Experientia, 32, 1403— Drosophila melanogaster. II. Geographic dines and climatic 1404. associations in Australasia, North America and Asia. ANDERSON, I'. R. AND OAKFSHOTT, .i. c., 1984. Parallel geo- Genetica, 58, 213-221. graphical pattern of allozyme variation intwo sibling KNIBB,W.R., OAKESHOTr, 5.I.. AND GIBSON, J. R. 1981. Drosophilaspecies. Nature, 308, 729—731. Chromosomeinversionpolymorphism in Drosophila ASI-IBURNER,M. ANI) LEMEUNIER,i-. 1976. Relationship melanogaster.I.Latitudinal dines and associations withinthemelanogaster speciessubgroup of the genus between inversions in Australian populations. Genetics, 98, Drosophila().I. Inversion polymorphisms in 833-847. Drosophilamelanogaster andD.simulans. Proc. R. Soc. KOJIMA,K. I., c;ILLI-.spIE, J. AND TOI3ARI, Y. N. 1970. A profile Lond. B, 193, 137-159. of Drosophila species enzymes assayed by electrophoresis. AYAI..A,I. J., POWELl.,R., IRACEY. M. L., MOURA, C.A. AND I. Number of alleles, heterozygosities, and linkage dis- PERLZ-SALAS. s. 1972. Enzyme variability on the D. willis- equilibrium in Glucose-metabolizing systems and some toni group.IV. Genievariation in natural populations of other enzymes. Biochem. Genet., 4, 627-637. D. willistoni. Genetics, 71, 113—139. I,AKOVAARA.S. ANt) KERANIN, L. 1980. Variation at the BABA.AISSA,F. AND SOLIGNAC. M.1984. La plupart des popu- a-gpdhlocusof Drosophilids. Hereditas, 92, 251-258. lations actulles de Drosophilasimulans ontprobablement LANC;LLY,C. H., TDBARI, Y. N. ANI) KOJIMA, K. I. 1974. Link- pour ancêtre une femelle unique dans un passé recent. CR. age disequilibrium in natural populations of Drosophila Acad. Sci. Paris. 299, 289-292. melanogaster. Genetics, 78, 921-936. BAKER,H. G. AND STEISBINS,G. L. 1965. The genetics of LEWONTIN,IS. C.1984. Detecting population differences in colonizing species. New York, Acad. Press, 588pp. quantitative characters as opposed to gene frequencies. BANI), H. T. 1975.A surveyof isozyme polymorphism in a Amer. Nat., 123, 115-124. Drosophila melanogaster naturalpopulations.Genetics, 80, LUYTI:N,I. 1982.Variations intraspCcifique et interspécifique 761—771. deshydrocarbures cuticulaires chez Drosophila simulans ISERGER,E. M. 1970. Acomparisonof gene-enzyme variation et des espèces affines. CR. Acad. Sci. Paris, 295, 733-736. between Drosophilarnelanogaster andD.simulans. MUITLER, I. F., VOELKER, R. A. ANT) MUKAI, T. 1977. Inversion Genetics, 66, 677—683. dines in populations of Drosophila melanogaser. Genetics, BOULETREAU.MERLE,5., ALI.EMAND. R., COHET Y. ANI) 87 169-176. DAVID, J. R. 1982. Reproductive strategy in Drosophila NEI, M. 1972. Genetic distance between populations. Am. Nat., melanogaster: significance of a genetic divergence between 106, 283-290. temperateand tropical populations. Oecologia,53,323-324. NEVO,E. 1978. Genetic variation in natural populations: pat- CABRERA,V.M., GONZALEZ,A.M., LARRAGA, J. M. AND terns and theory. Theoret. Pop. Biol., 13, 121-177. GULLON,A.1982. Electrophoretic variability innatural OAKESHOTE, J. 0., (:HAMBERS,C',. K., GIBSON, J. B. AND WIll.,. populations ofDrosophila inelanogaster and D. simulans. COK5, D. A. 1981a. Latitudinal relationshipof esterase-6 Genetica, 59, 191—201. andphosphoglucomutase gene frequencies in Drosophila CAPY. I'., DAVID, j.R.,ALLEMANI), R., HYYTIA, P. AND melanogaster. Heredity, 47(3),385-396. ROuAuLT, i. 1983. Genetic properties of North African OAKESHOTT, 3. 0., GIBSON, J. B.,ANDERSON, P.N., KNIHB, W. Drosophila melanogaster and comparison with European R.,ANDERSON,D. G. ANI) CHAMBERS, C. K. 1981h. andAfrotropical populations. Genet.Sel. Evol., 15 (2), Alcohol dehydrogenase and Glycerol-3 phosphate dehy- 185-200. drogenase dines in Drosophila melanogaster on different COHET,Y., VOUIDIBIO,J. AND DAVID, J. R. 1980. Thermal continents. Evolution, 36(I), 86-96. toleranceand geographic distribution: a comparison of OORIEN,S. 3. AND MACINTYRE, R. .t. 1969. An analysis of cosmopolitanandtropical endemic Drosophilaspecies.J. gene-enzyme variabilityinnatural populations of Therm. Biol., 5, 69-74. Drosophila melanogaster and f). simulans. Am. Nat., /03, DAVID, .1. IC 1974. Utilization ofmorphologicaltraitsfor the 97—113. analysisof genetic variability in wild populations. Aquilo, ()NISHI,S., LEIGH BROWN, A. J., VOULKER, R. A. AND Ser. Zool., 20, 49-61. LANGLEY, C. F!. 1982. Estimation of genetic variability in DAVID,5. N. 1982.Latitudinal variabilityof Drosophila natural populations of I). simulans bytwo dimensional melanogaster: Allozyme frequencies divergence between andstarch gel electrophoresis. Genetics, /00, 127-136. Europeanand Afrotropicalpopulations.Biochemical PARSONS,P. A. 1975. The comparative evolutionary biology of Genetics, 20, 7/8,747—761. the sibling species. Drosophila melanogaster and D. DAVID.5. R. ANI) BOCQUF.T, c. 1975. Similarities and differen- simulan.s. Qua!. Rev. BioL, 50, 151-169. ces in latitudinal adaptation of two Drosophila sibling PARSONS, P. A. 1980a. lsofemale Strains and evolutionary species.Nature, 257 588—590. strategies in natural populations. Evolutionary Biology, /3, DAVID,J. N., BOCQUET, C. AND P.A, E. 1976. New results on 175—217. thegeneticcharacteristics oftheEar East race of Drosophila PARSONS,p A. 1980b. Parallel climatic races fortolerances to melanogaster. Genet. Res. Camb., 28,253-260. high temperature desiccationstressin two Drosophila DAVID,J. R., BOCQUEt, C. ANT) DL SCHEEMAFKUR.LOUIS. NI. species.J.Biogeog., 7 97-101. 1977.Geneticlatitudinal adaptation of Drosophila PARSONS,I'.A. l980c. Adaptativestrategies innatural popula- melanogaster: New discriminative biometrical traits tions of Drosophila: Ethanol tolerance, desiccationresist- between European and Equatorial African populations. ance and development times in climatically optimal and Genet. Res. Camb., 30, 247-255. extreme environments. Theor. AppI. Genet., 57, 257-266. DAVID,3.R.AND TSACAS, L. 1981. Cosmopolitan, subcos- PARSONS,I'. A. 1983. The Evolutionary Biology of Colonizing mopolitan and widespread species: Different strategies Species. Cambridge Univ. Press, Cambridge, 262pp. GEOGRAPHIC VARIATION IN D.SIMULANS 217

PARSONS. P. A. AND STANLEY. S. M. 1981. Domesticated and TRANTAPHYLL!D1S, C. D. 1973. Allozyme variation in popula- widespread species. Ashburner, M., Carson, H. L. and tions of Drosophilamelanogaster andD.simulans from Thompson, J. N. Jr., (eds.) TheGenetics and Biology of Northern Greece. J.Heredity, 64,69—72. Drosophila, Vol.3a, Acad. Press, New York, pp. 349-393. TRIANTAPFIYLLIDIS, C. D., SCOURAS. Z. 0., PANOURGIAS, J. POWELL, .J. R. 1975. Protein variation in natural populations of N. AN!) !OANN!D!S, 0. C. 1982. Allozyme variation in . Evolut.Biol., 8, 79-119. Greek wild populations of Drosophilanielanogoster and SALAM, F. A., EL-ADL, A. M. AND KOSBA. Z. A. 1981. Isozyme D.simulans alonga North-South gradient. Genetica,58, polymorphism in Drosophila.IV.Interspecific variation 129—136. and population dynamics. Proc.7th Europ. Drosoph. Conf, TRIANTAPHYLLIDIS, C. t)., PANOURGIAS, J. N., SCOURAS, J. Oulu-Finland,pp. 227-235. G. AND IOANN!D!S, 0. C. 1980. A comparison of gent- SHOW, C. R. AND PRASAD, R. 1970. Starch gel-electrophoresis enzyme variation between Drosophilamelanogaster and of enzymes. A compilation of recipes. Biochem.Gene!., 4, Drosophilasimulan.s. Genetica, 51, 227—231. 297—320. TSACAS. L. AND LACHAISE, D. 1974. Quatre nouvelles espèces SINGH, R. A., HICKEY, D. A. AND DAVID, J. R. 1982. Genetic de Ia Chte-d'Ivoire du genre Drosophila,groupe differentiation between geographically distant populations Melanogasteretdiscussion de l'origine du sous-groupe of Drosophilamelanogaster. Genetics,101,235—256. melanogaster(Diptera;Drosophilidae). Ann.Univ. Abid- STALKER, H. 1). 1980. Chromosome studies on wild populations jan, Ser. E. (Ecol.), 7, 193—211. of Drosophilamelanogaster. II.Relationship of inversion VOELKER, R. A., COCKERHAM, C. C., JOHNSON. F. M., SCHAF- frequencies to latitude, season, wing-loading and flight FFR, H. F., MUKAI, T. AN!) METTLER, L. E. 1978. Inversions activity. Genetics,95, 211—223. fail to account for allozyme dines. Genetics,88, 515—527. STE! NER, W. W., SANY, K. L. AND PAIK, 'c. 1976. Electrophoretic VOELKER, R. A., MUKA!, T. ANt) JOHNSON, F.M. 1977. Genetic variability in island population of Drosophilasimulans and variation in populations of Drosophilatnelanogaster from D.inimigrans. Biochem. Genet., 14, 495-506. the Western United States. Genetica,47,143—148. TANTAWY, A. 0. AN!) MALLAU, o. s. 1961. Studies on natural WATANABE, T. K. AND WATANABE, T. 1977. Enzyme and populations of Drosophila.I.Heat resistance and geo- chromosome polymorphism in Japanese natural popula- graphic variation Drosophilamelanogaster andD.simulans. tions of Drosophila melanogaster.Genetics, 85, 319-329. Evolution, 15, 1—14. WHEELER, M. R. 1981. The drosophilidae: a taxonomic over- TEISSIER,- o. 1957. Discriminative biometrical characters in view. Ashburner, M., Carson, H. L. and Thompson, J. N. French and Japanese Drosophilamelanogaster. Proc. In!. Jr. (eds.) TheGenetics and Biology of Drosophila, Vol.3a, Gene!. Synip. Tokyo-Kyoto. 1956.Cytologia suppl., 502- Academic Press, New York, pp. 1-121. 505.