VECTOR CONTROL
Inheritance of resistance to organophosphorus insecticides in Culex tarsalis Coquillet* CHARLES S. APPERSON 1 & GEORGE P. GEORGHIOU 2
The mode of inheritance of resistance to parathion, fenitrothion, fenthion, chlorpyr fosmethyl, and parathion-methyl was studied in a strain of C. tarsalis. Resistance in the F1 hybrid was partially dominant. Backcross experiments indicated that resistance is derived from more than one gene. This conclusion was confirmed by repeated backcrosses of the F1 to the susceptible strain and selection of the backcross progeny with a discriminating dosage ofparathion-methyl.
Resistance to organophosphorus insecticides in a multiresistant strain of C. tarsalis has been shown to be due to a lower rate of insecticide absorption and increased insecticide detoxification (1). The present study concerns the mode of inheritance of this resistance.
Genetic techniques
MATERIALS AND METHODS
Insect strains The resistant strain (R) was collected from the Coachella Valley of California and selected in the laboratory by parathion-methyl pressure to a resistance level of 93.5 x. In this strain high levels of cross resistance at LC95 were shown towards fenitrothion (49.4 x ), parathion (55.6 x ), fenthion (76.8 x ), and chlorpyrifos-methyla (253.8 x) (2). The susceptible strain (S) was collected from the San Joaquin Valley of California about 20 years ago and has been maintained in the laboratory under insecticide-free conditions. The rearing procedures used were as described by Georghiou et al. (3) with certain modifications (4, 5). Bioassay procedures All the insecticides used were of analytical grade. Solutions of the desired concentrations were prepared * This investigation was supported in part by NIH Training Grant No. ES 47 from the National Institute of Environmental Health Sciences to the senior author. 1 Present address: Lake County Mosquito Abatement District, 410 Esplanade, Lakeport, CA 95453, USA. 'Professor of Entomology, University of California, Riverside, CA 92502, USA. a 0,0-dimethyl 0-(3,5,6-trichloro-2-pyridinyl) phosphorothioate.
3329
in acetone on a w/v basis. The testing procedures for larvae have been described by Georghiou et al. (3). Tests were replicated on at least 7 different days. The data for the R, S, and F1 populations were subjected to probit analysis (6).
Crosses were made using 150-200 adults of each sex. Reciprocal F1 crosses were made to expose maternal effects. The mode of inheritance of resistance to organophosphorus insecticides was investigated by means of parathion, fenitrothion, fenthion, chlorpyrifosmethyl, and parathion-methyl-i.e., those compounds towards which the strain showed the highest levels of resistance. The degree of dominance or recessiveness of the F1 hybrid was determined with reference to the log dose-probit mortality lines of the parental strains. To determine the number of genes involved in resistance the F1 hybrid was backcrossed to the S strain. An expected curve for the backcross progeny was calculated assuming monofactorial inheritance as already described (7, 8). Observed and expected mortalities were tested for significance using X2 analysis. Further experiments to test for monofactorial inheritance were carried out by repeated backcrossing and selection of the backcross progeny according to the method of Wright (9). These selections were carried out with 0.02 mg/litre parathion-methyl, which discriminates between S and F1 phenotypes.
97-
BULL. WORLD HEALTH ORGAN., Vol. 52, 1975
98
C. S. APPERSON & G. P. GEORGHIOU
Table 1. LC5o values for various insecticides tested on Fi hybrids of reciprocal crosses of C. tarsalis Insecticide
parathion
fenitrothion fenthion chlorpyrifosmethyl parathion-methyl
RcTxS?
R?xST
LC5o (mg/litre)
LC50 (mg/litre)
(fiducial limits)
(fiducial limits)
0.051 (0.049-0.053) 0.084 (0.075-0.095) 0.033 (0.032-0.034)
0.07 (0.068-0.073) 0.093 (0.082-0.11) 0.039 (0.038-0.041)
0.098 (0.084-0.11) 0.043 (0.041-0.045)
0.12 (0.11-0.13) 0.051 (0.048-0.054)
SF
95 90
Z
7
o
FENTHION6
80 60 50
F1
*backcross 70 -expected ,
-
.0
CLjJ4 J.
40' 30
-530 4
~~~~~~backcross
20
observed
10 5 0.001
3 0.1
0.01
dosage (mg/litre) WHO 75304
Fig. 2. Dosage-mortality relationships of larvae treated with fenthion. RESULTS AND DISCUSSION
The data in Table 1 reveal a slight matroclinous effect in resistance. This is evident from the higher resistance to insecticides shown by the F1 hybrid from the R ? x Sc3 cross. The log dose-probit mortality lines for the reciprocal crosses are significantly different with the exception of that for fenitrothion. Fig. 1-4 reveal that the F1 hybrid (R5 x ST) is partially dominant in resistance to parathion, fenitrothion, fenthion, and chlorpyrifos-methyl. The F1 hybrid is more intermediate in resistance to parathion-methyl, as shown in Fig. 5. In other studies on the inheritance of organophosphorus resistance in mosquitos the F1 hybrids have exhibited partial dominance towards malathion in C. tarsalis (10), C. pipiens fatigans (11), and C. p. pallens (12). Partial dominance has been reported for diazinon resistance
5
FIO
80
*70
R
backcross 1
6
expected
7
>V 60
a
40 30 20
2
Fig. 3.Dosage
dostaget (mglitre)ip
flavetrae
with fenitrothion.
.~~~~~~~~~~~~~~~~ 95 90
S
95
-
/
90 6
80
70
-. >60
backcross 0 /expected
-
6
-
6
/ 11
80 t 70 60
CHLORPYRIFOS-bMETHYL backcross
/~
*0.001
.
.,
6
~~~~~PARATHION
-
20*o backcross 10 ^observed 5
004
expected
-
50~~~~~~~~~~~~~~ "40-T 30
Fe
4
10
/
,backcross observed
l
0.01 0.1 dosage7 (mg/litre)
1.0
.I.I. ..
0.001
0.01 dosage
0.1
3 , 1.0
. ,.
(mg/litre)
WHO 75303
WHO 75306
Fig. 1. Dosage-mortality relationships of larvae treated with parathion.
Fig. 4. Dosage-mortality relationships of larvae treated with chlorpyrifos-methyl.
ORGANOPHOSPHORUS INSECTICIDE RESISTANCE
95 -PARATHION
t
PMETHYL
1
90
kcross
:50 30 20
i
expcted
-
*
A
O
backcross
:expected /
/
4
X
(3 abackcross moberved C.5 p. ~ -S0 ~ ~ ~~~~~~~H 030 obsv red 3 1.0 .0. 1 0.01 0.001 dosage (m/litre) WHO 75307 10
Fig. 5. Dosage-mortality relationships of larvae treated with parathion-methyl. in C. p. molestus (13), and for fenthion (14), temephos,a3 and fenitrothion in C. p. pallens (15). However, resistance to fenthion in C. p. fatigans was found to be slightly recessive (16). Fig. 1-5 reveal that the observed and expected mortalities in the backcrosses (F1 [R ? x SS3] 9 x SS)
were significantly different, by x2 analysis (P = 0.05), at the majority of dosage levels tested. Observed mortalities at the critical points corresponding to the 50% level of expected mortality are plotted with their calculated 95 % confidence intervals (Fig. 1-5). These computations confirm the significance of the differences between the observed and expected curves of the backcrosses. Since there was no overlap between the parathion-methyl log dose-probit mortality lines of the F1 hybrid and R strain, a backcross (Fl [R Y x SS] ? x R &) was made between the two populations to test further the results of the other
backcrosses. Significant deviation of the observed from the expected mortality was again found upon chi square analysis (Fig. 5). Because the results of each backcross differed significantly from the expected for monofactorial inheritance, it is concluded that organophosphorus resistance in the R strain is derived from more than one gene. For parathion, fenitrothion, and fenthion a depressed plateau is evident in the curve of the observed backcross. This may indicate that a major gene plus additional modifiers are involved in resistance. At higher dosages, observed mortalities exceeded the calculated ones in most cases. This was to be expected a
O,O'-(thiodi-4,1-phenylene) O,O,O',O'-tetramethyl phos-
phorothioate (" Abate ").
99
because of the independent assortment of modifiers and the major gene in the backcross. Thus, mortality greater than expected would be manifested when only the major gene is present-i.e., in the absence of modifiers. Where higher resistance existed in the R strain, as with chlorpyrifos-methyl and parathionmethyl, the plateau in the observed backcross was not as distinctly evident (Fig. 4 and 5). In this case, levels of mortality at the higher dosages were closer to the expected levels. This may indicate that a greater number of modifier genes are involved, some of which may be closely linked with the major gene. In other studies, tolerance to malathion (17) and parathion (18) in Aedes aegypti was atributed to polygenes. Resistance to fenthion in C. p. fatigans was attributed to a single major gene plus modifiers (16). Wright (9) has proposed a method for confirming polyfactorial inheritance. Repeated backcrosses are made to the S strain, and the progeny are selected at the midpoint of their distribution at a dosage that eliminates all susceptible phenotypes but is not lethal to the F1 hybrid. If monofactorial inheritance is involved the backcross progeny should consistently fall into two numerically equal classes-i.e., there should be 50 % mortality with each backcross selection. However, if more than one gene is associated with resistance then the percentage of mortality should increase with each selected backcross generation, since minor factors for resistance are progressively eliminated through independent assortment. This procedure was used to confirm the finding of polyfactorial inheritance in the R strain. Table 2 presents the effect of repeated backcrossing and selection on the stability of parathion-methyl resistance. An increase in mortality from 51.4 % to 75 % was observed over three backcross generations with a dosage of 0.02 mg/litre of parathion-methyl. These data and the results of the above-mentioned backcross experiments are consistent with a polygenic mode of inheritance of organophosphorus resistance in C. tarsalis. Table 2. Effect of repeated backcrossing and selection by parathion-methyl (0.02 mg/litre) on stability of resistance in C. tarsalis larvae Backcross generation
No. larvae
selected
Percentage mortality
1
1500
51.4
2
1500
60.7
3
1500
75.0
100
C. S. APPERSON & G. P. GEORGHIOU
ACKNOWLEDGEMENTS The authors thank Professor D. M. Yermanos, Department of Plant Sciences, University of California, Riverside, USA for suggestions in connection with this study.
RtSUME TRANSMISSION HERiDITAIRE DE LA RJSISTANCE AUX INSECTICIDES CHEZ CULEX TARSALIS COQUILLET
Le mode de transmission hereditaire de la resistance aux insecticides organophosphores a ete etudie sur une souche de Culex tarsalis Coquillet possedant les degres de resistance suivants (par rapport a la souche sensible): parathion-methyl, 93,5 x; fenitrothion, 49,4 x; parathion, 55,6 x; fenthion, 76,8 x et chlorpyrifos-methyl, 253,8 x . Chez l'hybride F1 la resistance etait partiellement dominante pour le parathion, le fenitrothion, le fenthion et le chlorpyrifos-methyl, et pratiquement intermediaire pour le parathion-methyl. Des retrocroisements avec les parents S et R ont montre que la r6sistance etait liee a plus d'un gene, ce qui a ete confirme par retro-
ORGANOPHOSPHORLS
croisements repetes entre l'hydride F1 et la souche sensible et s6lection des descendants au moyen d'une dose constante de parathion-m6thyl tuant tous les phenotypes sensibles sans etre letale pour les hybrides F1. Dans ces conditions, la mortalite des descendants obtenus par retrocroisement passait de 51,4% a 75% en trois generations. La presence d'un plateau en creux sur la courbe de regression dose-mortalite correspondant aux descendants obtenus par retrocroisement est interpretee comme montrant que la resistance est li6e a un gene majeur mais fortement influence par des genes modificateurs.
REFERENCES 1. APPERSON, C. S. & GEORGHIOU, G. P. Journal of economic entomology, 68: 153-157 (1975). 2. APPERSON, C. S. & GEORGH1OU, G. P. American journal of tropical medicine and hygiene (in press) (1975). 3. GEORGHIOU, G. P. ET AL. Bulletin of the World Health Organization, 35: 691-708 (1966). 4. APPERSON, C. S. & GEORGHIOU, G. P. Mosquito news, 32: 457 (1972). 5. APPERSON, C. S. & GEORGHIOU, G. P. Journal of medical entomology, 11: 573-576 (1974). 6. FINNEY, D. J. Probit analysis, London, Cambridge University Press, 1952. 7. GEORGHIOU, G. P. Advances in pest control research, 6: 171-230 (1965). 8. GEORGHIOU, G. P. Experimental parasitology, 26: 224-255 (1969). 9. WRIGHT, S. The genetics of quantitative variability, London, HMSO, 1952.
10. PLAPP, F. W. JR ET AL. Mosquito news, 21: 315-319 (1961). 11. UmNo, T. & SuzuKi, T. Japanese journal of sanitary zoology, 17: 191-195 (1966). 12. TADANO, T. Japanese journal of experimental medicine, 39: 13-16 (1969). 13. SuzUKI, T. & UMINO, T. Japanese journal of sanitary zoology, 20: 205-208 (1969). 14. TADANo, T. Japanese journal of sanitary zoology, 20: 158-160 (1969). 15. TADANo, T. Japanese journal of experimental medicine, 40: 59-66 (1970). 16. DORVAL, C. & BROWN, A. W. A. Bulletin of the World Health Organization, 43: 727-734 (1970). 17. PILLAI, M. K. K. & BROWN, A. W. A. Journal of economic entomology, 58: 255-266 (1965). 18. MATSUMURA, F. & BROWN, A. W. A. Mosquito news, 23: 26-31 (1963).
Inheritance of resistance to organophosphorus insecticides in Culex tarsalis Coquillet* CHARLES S. APPERSON 1 & GEORGE P. GEORGHIOU 2
The mode of inheritance of resistance to parathion, fenitrothion, fenthion, chlorpyr fosmethyl, and parathion-methyl was studied in a strain of C. tarsalis. Resistance in the F1 hybrid was partially dominant. Backcross experiments indicated that resistance is derived from more than one gene. This conclusion was confirmed by repeated backcrosses of the F1 to the susceptible strain and selection of the backcross progeny with a discriminating dosage ofparathion-methyl.
Resistance to organophosphorus insecticides in a multiresistant strain of C. tarsalis has been shown to be due to a lower rate of insecticide absorption and increased insecticide detoxification (1). The present study concerns the mode of inheritance of this resistance.
Genetic techniques
MATERIALS AND METHODS
Insect strains The resistant strain (R) was collected from the Coachella Valley of California and selected in the laboratory by parathion-methyl pressure to a resistance level of 93.5 x. In this strain high levels of cross resistance at LC95 were shown towards fenitrothion (49.4 x ), parathion (55.6 x ), fenthion (76.8 x ), and chlorpyrifos-methyla (253.8 x) (2). The susceptible strain (S) was collected from the San Joaquin Valley of California about 20 years ago and has been maintained in the laboratory under insecticide-free conditions. The rearing procedures used were as described by Georghiou et al. (3) with certain modifications (4, 5). Bioassay procedures All the insecticides used were of analytical grade. Solutions of the desired concentrations were prepared * This investigation was supported in part by NIH Training Grant No. ES 47 from the National Institute of Environmental Health Sciences to the senior author. 1 Present address: Lake County Mosquito Abatement District, 410 Esplanade, Lakeport, CA 95453, USA. 'Professor of Entomology, University of California, Riverside, CA 92502, USA. a 0,0-dimethyl 0-(3,5,6-trichloro-2-pyridinyl) phosphorothioate.
3329
in acetone on a w/v basis. The testing procedures for larvae have been described by Georghiou et al. (3). Tests were replicated on at least 7 different days. The data for the R, S, and F1 populations were subjected to probit analysis (6).
Crosses were made using 150-200 adults of each sex. Reciprocal F1 crosses were made to expose maternal effects. The mode of inheritance of resistance to organophosphorus insecticides was investigated by means of parathion, fenitrothion, fenthion, chlorpyrifosmethyl, and parathion-methyl-i.e., those compounds towards which the strain showed the highest levels of resistance. The degree of dominance or recessiveness of the F1 hybrid was determined with reference to the log dose-probit mortality lines of the parental strains. To determine the number of genes involved in resistance the F1 hybrid was backcrossed to the S strain. An expected curve for the backcross progeny was calculated assuming monofactorial inheritance as already described (7, 8). Observed and expected mortalities were tested for significance using X2 analysis. Further experiments to test for monofactorial inheritance were carried out by repeated backcrossing and selection of the backcross progeny according to the method of Wright (9). These selections were carried out with 0.02 mg/litre parathion-methyl, which discriminates between S and F1 phenotypes.
97-
BULL. WORLD HEALTH ORGAN., Vol. 52, 1975
98
C. S. APPERSON & G. P. GEORGHIOU
Table 1. LC5o values for various insecticides tested on Fi hybrids of reciprocal crosses of C. tarsalis Insecticide
parathion
fenitrothion fenthion chlorpyrifosmethyl parathion-methyl
RcTxS?
R?xST
LC5o (mg/litre)
LC50 (mg/litre)
(fiducial limits)
(fiducial limits)
0.051 (0.049-0.053) 0.084 (0.075-0.095) 0.033 (0.032-0.034)
0.07 (0.068-0.073) 0.093 (0.082-0.11) 0.039 (0.038-0.041)
0.098 (0.084-0.11) 0.043 (0.041-0.045)
0.12 (0.11-0.13) 0.051 (0.048-0.054)
SF
95 90
Z
7
o
FENTHION6
80 60 50
F1
*backcross 70 -expected ,
-
.0
CLjJ4 J.
40' 30
-530 4
~~~~~~backcross
20
observed
10 5 0.001
3 0.1
0.01
dosage (mg/litre) WHO 75304
Fig. 2. Dosage-mortality relationships of larvae treated with fenthion. RESULTS AND DISCUSSION
The data in Table 1 reveal a slight matroclinous effect in resistance. This is evident from the higher resistance to insecticides shown by the F1 hybrid from the R ? x Sc3 cross. The log dose-probit mortality lines for the reciprocal crosses are significantly different with the exception of that for fenitrothion. Fig. 1-4 reveal that the F1 hybrid (R5 x ST) is partially dominant in resistance to parathion, fenitrothion, fenthion, and chlorpyrifos-methyl. The F1 hybrid is more intermediate in resistance to parathion-methyl, as shown in Fig. 5. In other studies on the inheritance of organophosphorus resistance in mosquitos the F1 hybrids have exhibited partial dominance towards malathion in C. tarsalis (10), C. pipiens fatigans (11), and C. p. pallens (12). Partial dominance has been reported for diazinon resistance
5
FIO
80
*70
R
backcross 1
6
expected
7
>V 60
a
40 30 20
2
Fig. 3.Dosage
dostaget (mglitre)ip
flavetrae
with fenitrothion.
.~~~~~~~~~~~~~~~~ 95 90
S
95
-
/
90 6
80
70
-. >60
backcross 0 /expected
-
6
-
6
/ 11
80 t 70 60
CHLORPYRIFOS-bMETHYL backcross
/~
*0.001
.
.,
6
~~~~~PARATHION
-
20*o backcross 10 ^observed 5
004
expected
-
50~~~~~~~~~~~~~~ "40-T 30
Fe
4
10
/
,backcross observed
l
0.01 0.1 dosage7 (mg/litre)
1.0
.I.I. ..
0.001
0.01 dosage
0.1
3 , 1.0
. ,.
(mg/litre)
WHO 75303
WHO 75306
Fig. 1. Dosage-mortality relationships of larvae treated with parathion.
Fig. 4. Dosage-mortality relationships of larvae treated with chlorpyrifos-methyl.
ORGANOPHOSPHORUS INSECTICIDE RESISTANCE
95 -PARATHION
t
PMETHYL
1
90
kcross
:50 30 20
i
expcted
-
*
A
O
backcross
:expected /
/
4
X
(3 abackcross moberved C.5 p. ~ -S0 ~ ~ ~~~~~~~H 030 obsv red 3 1.0 .0. 1 0.01 0.001 dosage (m/litre) WHO 75307 10
Fig. 5. Dosage-mortality relationships of larvae treated with parathion-methyl. in C. p. molestus (13), and for fenthion (14), temephos,a3 and fenitrothion in C. p. pallens (15). However, resistance to fenthion in C. p. fatigans was found to be slightly recessive (16). Fig. 1-5 reveal that the observed and expected mortalities in the backcrosses (F1 [R ? x SS3] 9 x SS)
were significantly different, by x2 analysis (P = 0.05), at the majority of dosage levels tested. Observed mortalities at the critical points corresponding to the 50% level of expected mortality are plotted with their calculated 95 % confidence intervals (Fig. 1-5). These computations confirm the significance of the differences between the observed and expected curves of the backcrosses. Since there was no overlap between the parathion-methyl log dose-probit mortality lines of the F1 hybrid and R strain, a backcross (Fl [R Y x SS] ? x R &) was made between the two populations to test further the results of the other
backcrosses. Significant deviation of the observed from the expected mortality was again found upon chi square analysis (Fig. 5). Because the results of each backcross differed significantly from the expected for monofactorial inheritance, it is concluded that organophosphorus resistance in the R strain is derived from more than one gene. For parathion, fenitrothion, and fenthion a depressed plateau is evident in the curve of the observed backcross. This may indicate that a major gene plus additional modifiers are involved in resistance. At higher dosages, observed mortalities exceeded the calculated ones in most cases. This was to be expected a
O,O'-(thiodi-4,1-phenylene) O,O,O',O'-tetramethyl phos-
phorothioate (" Abate ").
99
because of the independent assortment of modifiers and the major gene in the backcross. Thus, mortality greater than expected would be manifested when only the major gene is present-i.e., in the absence of modifiers. Where higher resistance existed in the R strain, as with chlorpyrifos-methyl and parathionmethyl, the plateau in the observed backcross was not as distinctly evident (Fig. 4 and 5). In this case, levels of mortality at the higher dosages were closer to the expected levels. This may indicate that a greater number of modifier genes are involved, some of which may be closely linked with the major gene. In other studies, tolerance to malathion (17) and parathion (18) in Aedes aegypti was atributed to polygenes. Resistance to fenthion in C. p. fatigans was attributed to a single major gene plus modifiers (16). Wright (9) has proposed a method for confirming polyfactorial inheritance. Repeated backcrosses are made to the S strain, and the progeny are selected at the midpoint of their distribution at a dosage that eliminates all susceptible phenotypes but is not lethal to the F1 hybrid. If monofactorial inheritance is involved the backcross progeny should consistently fall into two numerically equal classes-i.e., there should be 50 % mortality with each backcross selection. However, if more than one gene is associated with resistance then the percentage of mortality should increase with each selected backcross generation, since minor factors for resistance are progressively eliminated through independent assortment. This procedure was used to confirm the finding of polyfactorial inheritance in the R strain. Table 2 presents the effect of repeated backcrossing and selection on the stability of parathion-methyl resistance. An increase in mortality from 51.4 % to 75 % was observed over three backcross generations with a dosage of 0.02 mg/litre of parathion-methyl. These data and the results of the above-mentioned backcross experiments are consistent with a polygenic mode of inheritance of organophosphorus resistance in C. tarsalis. Table 2. Effect of repeated backcrossing and selection by parathion-methyl (0.02 mg/litre) on stability of resistance in C. tarsalis larvae Backcross generation
No. larvae
selected
Percentage mortality
1
1500
51.4
2
1500
60.7
3
1500
75.0
100
C. S. APPERSON & G. P. GEORGHIOU
ACKNOWLEDGEMENTS The authors thank Professor D. M. Yermanos, Department of Plant Sciences, University of California, Riverside, USA for suggestions in connection with this study.
RtSUME TRANSMISSION HERiDITAIRE DE LA RJSISTANCE AUX INSECTICIDES CHEZ CULEX TARSALIS COQUILLET
Le mode de transmission hereditaire de la resistance aux insecticides organophosphores a ete etudie sur une souche de Culex tarsalis Coquillet possedant les degres de resistance suivants (par rapport a la souche sensible): parathion-methyl, 93,5 x; fenitrothion, 49,4 x; parathion, 55,6 x; fenthion, 76,8 x et chlorpyrifos-methyl, 253,8 x . Chez l'hybride F1 la resistance etait partiellement dominante pour le parathion, le fenitrothion, le fenthion et le chlorpyrifos-methyl, et pratiquement intermediaire pour le parathion-methyl. Des retrocroisements avec les parents S et R ont montre que la r6sistance etait liee a plus d'un gene, ce qui a ete confirme par retro-
ORGANOPHOSPHORLS
croisements repetes entre l'hydride F1 et la souche sensible et s6lection des descendants au moyen d'une dose constante de parathion-m6thyl tuant tous les phenotypes sensibles sans etre letale pour les hybrides F1. Dans ces conditions, la mortalite des descendants obtenus par retrocroisement passait de 51,4% a 75% en trois generations. La presence d'un plateau en creux sur la courbe de regression dose-mortalite correspondant aux descendants obtenus par retrocroisement est interpretee comme montrant que la resistance est li6e a un gene majeur mais fortement influence par des genes modificateurs.
REFERENCES 1. APPERSON, C. S. & GEORGHIOU, G. P. Journal of economic entomology, 68: 153-157 (1975). 2. APPERSON, C. S. & GEORGH1OU, G. P. American journal of tropical medicine and hygiene (in press) (1975). 3. GEORGHIOU, G. P. ET AL. Bulletin of the World Health Organization, 35: 691-708 (1966). 4. APPERSON, C. S. & GEORGHIOU, G. P. Mosquito news, 32: 457 (1972). 5. APPERSON, C. S. & GEORGHIOU, G. P. Journal of medical entomology, 11: 573-576 (1974). 6. FINNEY, D. J. Probit analysis, London, Cambridge University Press, 1952. 7. GEORGHIOU, G. P. Advances in pest control research, 6: 171-230 (1965). 8. GEORGHIOU, G. P. Experimental parasitology, 26: 224-255 (1969). 9. WRIGHT, S. The genetics of quantitative variability, London, HMSO, 1952.
10. PLAPP, F. W. JR ET AL. Mosquito news, 21: 315-319 (1961). 11. UmNo, T. & SuzuKi, T. Japanese journal of sanitary zoology, 17: 191-195 (1966). 12. TADANO, T. Japanese journal of experimental medicine, 39: 13-16 (1969). 13. SuzUKI, T. & UMINO, T. Japanese journal of sanitary zoology, 20: 205-208 (1969). 14. TADANo, T. Japanese journal of sanitary zoology, 20: 158-160 (1969). 15. TADANo, T. Japanese journal of experimental medicine, 40: 59-66 (1970). 16. DORVAL, C. & BROWN, A. W. A. Bulletin of the World Health Organization, 43: 727-734 (1970). 17. PILLAI, M. K. K. & BROWN, A. W. A. Journal of economic entomology, 58: 255-266 (1965). 18. MATSUMURA, F. & BROWN, A. W. A. Mosquito news, 23: 26-31 (1963).
