@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Zoology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Sharp, Cecil Bert"@en ; dcterms:issued "2010-02-19T04:47:31Z"@en, "1977"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """There has recently been renewed interest concerning the location of the major loci responsible for the Segregation Distortion phenomenon in Drosophila melanogaster. Hartl (1974) has shown that two major sites are involved: Sd and Rsp. Rsp confers insensitivity to SD chromosomes, while Sd is considered to be the major locus that initiates distortion, Sd is located to the left of Rsp and both are located between Tft and cn. Ganetzky (1977) has extended these findings by showing that just distal to pr there is a locus that, if deleted on a SD chromosome, eliminates distortion and he argues that this is the Sd site. Ganetzky (1977) also uncovered another important locus, in or near the heterochromatin of 2L, that, if deleted from a SD chromosome, greatly reduces the ability of that chromosome to distort and he argued that this site is an enhancer of SD, E(SD). Ganetzky (1977) , also suggests that Rsp might be located very close to the centromere in the proximal heterochromatin of 2R. The results presented here demonstrate the presence of an important component of SD located within the proximal heterochromatin of 2L. These results also show that there is another important site located just distal to pr. However, when this site is removed by recombination from a SD chromosome, a certain level of residual distortion remains. It is argued that the site that Ganetzky (1977) called E(SD) is likely responsible for this residual distortion in the absence of the site just distal to pr. Thus the site near pr is called Sd₁ and the site near 1t is called Sd₂,. Loss of either site results in a large reduction, but not complete elimination, of the distorting ability of a SD chromosome. Other data are presented that, on the whole, agree with Ganetzky's (1977) proposal that Rsp is located in the centromeric heterochromatin of 2R, very close to the centromere. Miklos and Smith-White (1971) have suggested that k (the segregation ratio observed from a given mating) is a deceptive measure of the degree of distortion and they have proposed another method of measuring distortion based on their model of sperm dysfunction. Some of the weak assumptions of this model are discussed and a simpler alternative is presented. The alternative model assumes that the potential segregation ratios of a population of SD males follow a truncated normal distribution. Data are presented that are not necessarily inconsistent with this assumption. The same data show that it is likely that certain SJJ chromosomes differ in their susceptibility to modifiers of It is concluded that at present k provides the clearest measure of distortion."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/20482?expand=metadata"@en ; skos:note "THE LOCATION AND PROPERTIES OF SOME OF THE MAJOR LOCI AFFECTING THE SEGREGATION DISTORTION PHENOMENON IN DROSOPHILA MELANOGASTER by CECIL BERT SHARP B.Sc, Uni v e r s i t y of B r i t i s h Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Department of Zoology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1977 © C e c i l Bert Sharp In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced d e g r e e at t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the Head o f my Depar tment o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Zoology Depar tment o f The U n i v e r s i t y o f B r i t i s h C o l u m b i a 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date Oct. 4, 1977 i i ABSTRACT There has recently been renewed i n t e r e s t concerning the l o c a t i o n of the major l o c i responsible for the Segregation D i s t o r t i o n phenomenon i n Drosophila melanogaster. H a r t l (1974) has shown that two major s i t e s are involved: Sci and Rsp. Rsp confers i n s e n s i t i v i t y to S>D chromosomes, while S<1 i s considered to be the major locus that i n i t i a t e s d i s t o r t i o n , ^d i s located to the l e f t of Rsp and both are located between T f t and cn. Ganetzky (1977) has extended these findings by showing that j u s t d i s t a l to ;p_r there i s a locus that, i f deleted on a SD chromosome, eliminates d i s t o r t i o n and he argues that t h i s i s the Sd^ s i t e . Ganetzky (1977) also uncovered another important locus, i n or near the heterochromatin of 2L, that, i f deleted from a SD chromosome, greatly reduces the a b i l i t y of that chromosome to d i s t o r t and he argued that t h i s s i t e i s an enhancer of J3D, E(SD). Ganetzky (1977) , also suggests that Rsp might be located very close to the centromere i n the proximal heterochromatin of 2R. The r e s u l t s presented here demonstrate the presence of an important component of J5D located within the proximal heterochromatin of 2L. These r e s u l t s also show that there i s another important s i t e located j u s t d i s t a l to _p_r_. However, when t h i s s i t e i s removed by recombination from a JSD chromosome, a c e r t a i n l e v e l of r e s i d u a l d i s t o r t i o n remains. I t i s argued that the s i t e that Ganetzky (1977) c a l l e d E(SD) i s l i k e l y responsible for t h i s r e s i d u a l d i s t o r t i o n i n the absence of the s i t e j u s t d i s t a l to J D T . Thus the s i t e near _p_r i s c a l l e d Sd^ and the s i t e near JLt i s c a l l e d Sd^,. Loss of e i t h e r s i t e r e s u l t s i n a large reduction, but not complete elimination, of the d i s t o r t i n g a b i l i t y of a SD chromosome. Other data are presented that, on the whole, agree with Ganetzky's (1977) proposal that Rsp i s located i n the centromeric heterochromatin of 2R, very close to the centromere. Miklos and Smith-White (1971) have suggested that k (the segregation r a t i o observed from a given mating) i s a deceptive measure of the degree of d i s t o r t i o n and they have proposed another method of measuring d i s t o r t i o n based on t h e i r model of sperm dysfunction. Some of the weak assumptions of th i s model are discussed and a simpler a l t e r n a t i v e i s presented. The a l t e r n a t i v e model assumes that the p o t e n t i a l segregation r a t i o s of a population of SD males follow a truncated normal d i s t r i b u t i o n . Data are presented that are not nec e s s a r i l y inconsistent with t h i s assumption. The same data show that i t i s l i k e l y that c e r t a i n SJJ chromosomes d i f f e r i n t h e i r s u s c e p t i b i l i t y to modifiers of I t i s concluded that at present k provides the clearest measure of d i s t o r t i o n . XV TABLE OF CONTENTS PAGE ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v LIST OF FIGURES v i ACKNOWLEDGEMENT v i i i GENERAL INTRODUCTION 1 CHAPTER I. ON MEASURING DISTORTED SEGREGATION RATIOS Introduction . . . . . 8 Materials and Methods • 28 Results and Discussion 30 Conclusion 46 CHAPTER I I . ON THE LOCATION AND PROPERTIES OF SOME OF THE LOCI AFFECTING SEGREGATION DISTORTION Introduction • 50 Materials and Methods 54 Results and Discussion 55 Conclusion 86 LITERATURE CITED 87 LIST OF TABLES TABLE PAGE I. The distorting abilities of the b SD-5 recombinants. 60 II. The distorting abilities of the b pr SD-5 recombinants. 62 III. The distorting abilities of miscellaneous SD-5 recombin-ants. 64 IV. The distorting abilities of R(SD-5)-x/SMl males mated to homozygous cn bw females. 67 V. The distorting abilities of R(SD-5) b pr-5 under various circumstances. 68 VI. Tests to determine if Sd_+, on b pr It pk cn is semi-active. 71 VII. The distorting abilities of R(SD-72) b pr-x/cn bw males mated to cn bw/cn bw females. 75 VIII. The sensitivites of b-bearing recombinants, from Or-R/b pr rl cn females, when heterozygous with R(SD-5) pk cn. 79 IX. The sensitivities of cn-bearing recombinants, from Or-R/b pr rl cn, when heterozygous with R(SD-5)b-8. 80 X. The sensitivities of cn bw chromosomes with JT-ray induced lethals on them. 82 V I LIST OF FIGURES FIGURE PAGE 1. The relationships between kc and km at various values of k f . 12 2. The relationships between k c and at various values of k . 14 m 3. The relationships between the distribution of make values within a male (m), the threshold of dysfunction (P), the proportion of Sp_+ sperm that dysfunction (e), and the make level of a male (m1). 21 4. The relationship between dk/dm1 and mean k. 23 • 5. A model similar to Miklos and Smith-White's, except P is allowed to vary and 0^ e q u a l s ^ . 26 6. The observed k distributions of the 6 \"SD\" chromosomes tested. 3 2 7. A diagrammatic test for normality of the k distributions of the 6 \"SD\" chromosomes tested. 3 5 8. A diagrammatic test for normality of the distributions df arc sin 7k\" of the 6 \"SD\" chromosomes tested. 37 9. The observed and predicted relationships between the variance of k (without the binomial component) and mean k. 43 v i i FIGURE PAGE 10. The complementation relationships between the putative SD-5 deficiencies and Hill iker's (1976) groups of lethal mutations in the proximal heterochromatin of 2L. 57 11. The complementation relationships between the cn bw chromosomes bearing putative deficiencies and Hill iker's (1976) groups of lethal mutations in the proximal hetero-chromatin of 2R. 84 ACKNOWLEDGEMENT I would like to thank Dr. David G. Holm for his support of and interest in the present study. Thanks are also due to Mr. Steve Borden for analysing some of the data with a computer and to Dr. Conrad F. Wehrhahn for providing useful statistical advice. The technical assist-ance of Loren Caira was most helpful, as were the many suggestions of Dr. Arthur J . Hilliker. 1 GENERAL INTRODUCTION Segregation d i s t o r t e r (SD) was f i r s t reported by Sandler, Hiraizumi, and Sandler i n 1959. They noted that some second chromosomes taken from a natural population near Madison, Wisconsin, showed abnormal segregation r a t i o s . In p a r t i c u l a r , these chromosomes were recovered far i n excess of the 1:1 r a t i o expected from the cross of SD/cn bw males mated to cn bw/cn bw females. (cn, cinnabar eyes, and bw, brown eyes, together produce a white-eyed phenotype). The recovery r a t i o , defined as the number of ^D progeny divided by the t o t a l number of progeny and given the symbol k, was often i n excess of 0.9. However, the value of k for the r e c i p r o c a l cross, SD/cn bw females mated to cn bw/cn bw males, was 0.5. These r e s u l t s suggested that the mechanism of jSD resided i n some abnorm-a l i t y of e i t h e r male spermatogenesis or spermiogenesis. Sandler, Hiraizumi, and Sandler (1959) proposed that SD induced a break i n i t s homologue during prophase I and that t h i s break fused to form a d i c e n t r i c bridge and acentric fragments during anaphase I I . Preliminary c y t o l o g i c a l evidence indicated that d i c e n t r i c bridges and acentric fragments could be observed during male meiosis (Sandler, Hiraizumi, and Sandler 1959); however, these observations could not be v e r i f i e d (Peacock and Erickson 1965). Although t h i s hypothesis does not have d i r e c t c y t o l o g i c a l support, i t does have i n d i r e c t genetic evidence i n i t s favour. Hiraizumi (1961) observed that rare cn bw chromosomes recovered from SD/cn bw males had lower homozygous v i a b i l i t i e s than did cn bw chromosomes recovered from SD\"*\"/cn bw males. In addition, Crow, Thomas, and Sandler (1962) found that the response of male r e -combination to r a d i a t i o n was much greater i n SD heterozyous males than i n SD + heterozyous males. Both of these experiments were consistent with the hypo-thesis that SD induces chromosome breaks i n i t s homologue. Peacock and Erickson (1965) were unable to detect any c y t o l o g i c a l abnorm-3 a l i t i e s i n the meiotic d i v i s i o n s of STJ males. Because of t h i s and the i n d i r e c t nature of the evidence i n support of the breakage hypothesis, they modified the fu n c t i o n a l pole hypothesis of Novitski and Sandler (1957) to explain SD. The functional pole hypothesis holds that i n male meiosis there are two poles and that the products of one pole function i n f e r t i l i z a t i o n , while the products of the other pole are r e g u l a r l y nonfunctional, although motile and transferred to the female. Peacock and Erickson (1965) proposed that SD operates by prefer-e n t i a l l y o r i e n t i n g the SD chromosome toward the functional pole. In support of t h i s model, Peacock and Erickson (1965) showed that with both SD/cn bw and wild type males the number of progeny a female yielded was equal to approxi-mately one-half the number of sperm transferred to the female. However, Zimmering and Fowler (1968) have shown that the e f f i c i e n c y of sperm u t i l i z a t i o n v a ries between females, depending upon t h e i r genotype. The (yellow body colour) females used by Peacock and Erickson normally u t i l i z e only about one-h a l f of t h e i r stored sperm, whereas Oregon-R (wild type) females use up to eighty per cent of the sperm transferred by Oregon-R males. Thus i t would appear that because of the type of female used to obtain progeny to sperm r a t i o s , Peacock and Erickson's (1965) r e s u l t s were a t y p i c a l . The f u n c t i o n a l pole hypothesis does not appear to apply g e n e r a l l y to Drosophila males and consequently i t i s a poor s t a r t i n g point from which to develop an hypothesis f o r explaining mechanisms f o r the action of SD. A much more reasonable s t a r t i n g point i s provided by the r e s u l t s of Hartl,. Hiraizumi, and Crow (1967). They demonstrated that i f both SD/cn bw and cn bw/ cn bw males were provided with an excess of v i r g i n females u n t i l the males be-came s t e r i l e , and i f the females were re-brooded u n t i l they ceased to lay f e r t i l i z e d eggs, then the cn bw/cn bw males produced twice as many progeny as did the SD/cn bw males. This f i n d i n g suggested that, compared to cn bw/cn bw males, a SD/cn bw male can produce only one-half the number of functional 4 sperm. This i n turn suggested that i n SD/cn bw males the sperm which carry cn bw chromosomes are somehow rendered dysfunctional. Some of the steps involved i n t h i s dysfunction process have recently been elucidated. Tokuyasu, Peacock, and Hardy (1977) have reported a d e t a i l e d study of spermiogenesis i n SD/cn bw males and they have compared those r e s u l t s with e a r l i e r observations they made on spermiogenesis i n normal males (Peacock, Tokuyasu, and Hardy 1972; Tokuyasu 1974). The e a r l i e s t differences between normal males and SD/cn bw males are observed during the transformation of the early s p h e r i c a l spermatid nucleus into the highly condensed and compact nucleus of the mature spermatid. In a SD/cn bw male the chromatin of cn bw bearing spermatids f a i l s to condense to the same degree as ei t h e r the chroma-t i n of S>D bearing spermatids or the chromatin of a l l the spermatids i n a normal male. Later i n spermiogenesis these spermatids with improperly condensed heads w i l l frequently f a i l to become i n d i v i d u a l i z e d by the i n d i v i d u a l i z a t i o n mem-brane that d e l i m i t s the other spermatids from the syncytium i n which they were formerly located. In normal males, a f t e r i n d i v i d u a l i z a t i o n the sperm undergo a c o i l i n g process p r i o r to release into the t e s t i c u l a r lumen. In SD/cn bw males, those sperm which are improperly i n d i v i d u a l i z e d are frequently not c o i l e d with the rest of the bundle and subsequently degenerate. However, even i n SD/cn bw males that have k values near 1.0, one frequently observes among the c o i l e d sperm of a s i n g l e cyst a few sperm that have improperly condensed chromatin. Thus, although the majority of the cn bw-bearing sperm are l o s t during spermiogenesis, a few are l i k e l y transferred to females, but t h e i r a b i l i t y to function i n f e r t i l i z a t i o n i s severely l i m i t e d , probably owing to t h e i r abnormally condensed n u c l e i . Improper condensation of chromatin suggests that the t r a n s i t i o n from l y s i n e - r i c h to a r g i n i n e - r i c h histones observed during spermiogenesis i n normal 5 Drosophila males (Das, Kaufmann, and Gay 1964) i s somehow impaired i n the cn bw-bearing spermatids of SD/cn bw males. In f a c t , t h i s i s l i k e l y the case, since Kettanah and H a r t l (1976) have found that the l y s i n e - r i c h to arginine-r i c h t r a n s i t i o n cannot be detected cytochemically i n _SD homozygotes. Although the f i r s t cytochemical and morphological manifestations of seg-regation d i s t o r t i o n occur during the nuclear condensation phase of spermiogene-s i s , Mange (1968) has shown that the segregation r a t i o s of SD/cn bw males are s e n s i t i v e to temperature shocks applied during the early stages of meiosis. However, t h i s observation i s not n e c e s s a r i l y inconsistent with the hypothesis that the mechanism of segregation d i s t o r t i o n involves modification of the t r a n s i t i o n from l y s i n e - r i c h to a r g i n i n e - r i c h sperm histones, since Gould-Somero and Holland (1974) have demonstrated by means of autoradiography that RNA syn-thesis ceases pr e - m e i o t i c a l l y i n Drosophila males. Because RNA synthesis ceases pre-meiotically i t i s not unreasonable to assume that the factors which mediate the improper condensation of chromatin i n SD + bearing spermatids are also present pre-meiotically. Furthermore, i t i s not unreasonable to assume that temperature shocks can a f f e c t these factors i n such a way that they w i l l a l t e r the frequency of SD + sperm with improperly condensed chromatin. In s p i t e of the recent advances e l u c i d a t i n g the general mechanism by which segregation d i s t o r t i o n operates, there i s s t i l l a lack of s o l i d informa-t i o n concerning the l o c i involved i n the process. Numerous studies have been undertaken to solve t h i s problem; however, the r e s u l t s of these studies have been amazingly ambivalent. The purpose of the present study was to approach t h i s mapping problem by using methods that d i f f e r e d somewhat from those used by previous workers, i n the hope that a firmer conclusion could be made con-cerning the l o c i involved. This i n turn should c e r t a i n l y help to complement biochemical studies determine a more d e t a i l e d understanding of the mechanisms involved i n segregation d i s t o r t i o n . Previous studies have examined the properties of SD recombinants recover-ed between pr (purple eyes) and cn (cinnabar eyes) (Sandler and Hiraizumi 1960b, Hiraizumi and Nakazima 1967), Cy_ (curly wings) and cn (Crow, Thomas and Sandler 1962), and T f t (tufted wings) and cn (Hartl 1974). These pai r s of markers provide one locus i n the euchromatin of 2L and the other i n the euchromatin of 2R. Since the centromeric heterochromatin had not been marked i n the e a r l i e r studies, one could not p o s i t i o n the SD l o c i r e l a t i v e to t h i s block of heterochromatin. In the present study the marker chromosome used to obtain SD recombinants was b pr It pk cn. This chromosome c a r r i e d l j : ( l i g h t eyes), a locus located i n the centromeric heterochromatin of 2L ( H i l l i k e r and Holm 1975). In addition to recombination, I also employed deletion mapping. However, before examining the r e s u l t s of these mapping experiments, I w i l l present some considerations pertaining to the measurement of d i s t o r t e d segre-gation r a t i o s . 7 CHAPTER I ON MEASURING DISTORTED SEGREGATION RATIOS INTRODUCTION One o f t h e most i m p o r t a n t f a c t o r s w h i c h must be t a k e n i n t o c o n s i d e r a t i o n when e s t i m a t i n g a g e n e t i c s e g r e g a t i o n r a t i o i s t he r e l a t i v e v i a b i l i t y o f t h e d i f f e r e n t g e n o t y p e s i n v o l v e d . F o r e x a m p l e , a l t h o u g h t h e chromosomes on w h i c h two d i f f e r e n t a l l e l e s o f a s i n g l e gene a r e l o c a t e d may be s e g r e g a t i n g r a n d o m l y , t he f r e q u e n c i e s o f t h e two a l l e l e s i n t h e r e c o v e r e d p r o g e n y may n o t be e q u a l b e c a u s e p r o g e n y b e a r i n g one o f t h e a l l e l e s may be l e s s v i a b l e t h a n p r o g e n y b e a r i n g the o t h e r . I n t h e c a s e o f S e g r e g a t i o n D i s t o r t e r i n D r o s o p h i l a m e l a n o g a s t e r , one i s i n t e r e s t e d i n d e t e r m i n i n g the v i a b i l i t y o f p r o g e n y b e a r i n g the ^ D chromosome r e l a t i v e t o t h e v i a b i l i t y o f p r o g e n y b e a r i n g t h e SP\"*\" chromosome. One method o f d o i n g t h i s makes use o f t h e p r o p e r t y t h a t J5D o p e r a t e s i n m a l e s , b u t n o t i n f e m a l e s . I f s e g r e g a t i o n i s random i n , f o r e x a m p l e , S D / c n bw f e m a l e s , t h e n t h e r e l a t i v e r e c o v e r y o f SD and cn b w - b e a r i n g p r o g e n y s h o u l d a p p r o x i m a t e l y r e f l e c t t h e r e l a t i v e v i a b i l i t y o f p r o g e n y b e a r i n g t h e s e chromosomes. One p r o b l e m w i t h t h i s method i s t h a t r e c o m b i n a t i o n does n o t o c c u r i n m a l e s , b u t does o c c u r i n f e m a l e s . Thus SD and cn bw p r o g e n y r e c o v e r e d f r o m male S D / c n bw p a r e n t s w i l l n o t be g e n o t y p i c a l l y i d e n t i c a l t o p r o g e n y o f t h e same v i s i b l e p h e n o t y p e s r e c o v e r e d f r o m f ema le S D / c n bw p a r e n t s . H o w e v e r , t h i s p r o b l e m i s a l l e v i a t e d t o a c e r t a i n e x t e n t b e c a u s e most JSD chromosomes have one o r more i n v e r s i o n s , i n t h e r i g h t arm o f chromosome 2 , w h i c h g r e a t l y r e d u c e the f r e q u e n c y o f r e c o m b i n a t i o n be tween cn. and bw. A n o t h e r p r o b l e m w i t h t h i s method i s t h a t i t i s p o s s i b l e t h a t t h e r e l a t i v e v i a b i l i t y o f SD and c n bw b e a r i n g p r o g e n y may d i f f e r d e p e n d i n g upon w h e t h e r t he ma le o r f e m a l e p a r e n t was S D / c n bw. T h a t i s , i t i s c o n c e i v a b l e t h a t r e l a -t i v e v i a b i l i t y may be a s s o c i a t e d w i t h a m a t e r n a l e f f e c t . 9 However, I do not f e e l that the possible e f f e c t s of eit h e r of the above problems would be s u f f i c i e n t l y large to n u l l i f y the advantages of attempting to measure the r e l a t i v e v i a b i l i t y of SD and SD + bearing progeny by performing the r e c i p r o c a l cross, i . e . with SD/SD+ as the female parent. I f one assumes that the method i s approximately v a l i d , then one can use the segregation r a t i o determined from a cross where the female parent i s SD/SD+, given the symbol k^, to correct the segregation r a t i o determined from a cross where the male parent i s SD/SD+, given the symbol k ^ This corrected segregation r a t i o , given the symbol k^, i s obtained as follows: k k c / k f k m + K ) I T This formula, however, does not take into account v a r i a t i o n i n k and k_. m r One can use the means of k and k^ i n order to obtain an estimate of k , but m f c t h i s estimate w i l l not be the mean of k , i t w i l l be the mode of k . I t i s c c generally preferable to use the mean of k c- Mood, G r a y b i l l , and Boes (1974) give the following method for approximating the mean of a derived v a r i a b l e such as k : c a2 k c d2 \\ E(k ) - k + h Var(k ) + % V a r ( k J c c m ->> 2 m 3 k 2 a k 2 It i s also of i n t e r e s t to know the variance of k , i n order that confidence c l i m i t s can be estimated for E(k ). Mood, G r a y b i l l , and Boes (1974) give the c following as an estimate of the variance of k : c | w c Var(k ) =^ \\ Var(k ) + k / m \\ < 5 k m Var(k f) 10 This method of estimating the mean and variance of a derived v a r i a b l e , f ( x ) , depends upon expanding f(x) i n a Taylor s e r i e s about E(x). The accuracy of the approximation depends upon the magnitude of the remainder term i n the Taylor series (Johnson and Kotz 1969). As can be seen from Figure 1, the curves of k versus k could be e a s i l y estimated by a Taylor series at values m c of k^ between 0.3 and 0.7, since the curves are close to being l i n e a r . Figure 2 shows that the same holds for curves of k,. versus k at values of k between f c m 0.3 and 0.7. Thus, I would conclude that i t would.be quite reasonable to use these approximations i f and k^ are both between 0.3 and 0.7. The required p a r t i a l d e r i v a t i v e s are given below: FIGURE 1 The relationships between kc and km at various values of k f. 12 FIGURE 2 The relationships between k and k f at various values of k 15 The difference between and E(k c) I s usually quite small ( i n the t h i r d d e c i -mal place) and consequently for most purposes i t would be s a t i s f a c t o r y to use k i n place of E(k ). c c The phenomenon c a l l e d i n s t a b i l i t y by Sandler and Hiraizumi (1960a) has suggested to some (Miklos and Smith-White 1971) that possibly other problems are associated with using k as a measure of the degree of d i s t o r t i o n . Sandler and Hiraizumi (1960a) noted that unrecombined chromosomes, such as SD-5 and SD-72, i n v a r i a b l y had very high mean k values (approximately 0.99) as w e l l as very small variances of k. Because these unrecombined chromosomes had small variances, they were c a l l e d stable. However, whenever bw-bearing SD recombinants, derived from SD/cn bw females, were tested, they were found to have lowered mean k values, the lowest of these being about 0.82. In addition, these recombinants always had increased variances of k, and accordingly were c a l l e d e i t h e r semistable or unstable, depending upon how great the variance of k was. Unstable l i n e s always had lower mean k values than semistable l i n e s . Sandler and Hiraizumi (1960a) also observed that i f the cn-bearing r e -combinants, derived from SD/cn bw females, were made heterozygous with bw-bearing semistable recombinants, then the semistable recombinant became stable. Unstable bw-bearing recombinants could be made more stable i n t h i s manner, but could not be made completely stable. These observations suggested to these authors that a s t a b i l i z e r of jSD was located i n 2R. One rather perplexing observation was that by s e l e c t i n g males with low k values a semistable l i n e could be made unstable and the mean k could be re-duced, but by s e l e c t i n g males with high k values an unstable l i n e could not be made semistable. However, s e l e c t i o n f or only one generation was e f f e c t i v e i n either s l i g h t l y increasing or s l i g h t l y decreasing the mean k values of an un-stable l i n e , depending upon the d i r e c t i o n of s e l e c t i o n . Sandler and Hiraizumi (1960a) reasoned that, since male v a r i a t i o n i n k showed some evidence of h e r i t a b i l i t y for one generation, s e l e c t i o n for high k values should have made an unstable l i n e semistable. Because s e l e c t i o n could not produce t h i s e f f e c t , they suggested that the high k \"states\" of unstable l i n e s have a high mutation rate back to low k \"states\". That i s , although a number of genetic modifiers may combine i n a male to produce a high l e v e l of d i s t o r t i o n i n that male, the modifiers could not maintain t h e i r e f f e c t when selected, because they mutated, at a high' frequency, to an a l l e l i c state that did not contribute to a high l e v e l of d i s t o r t i o n . I assume that t h e i r model implies that s e l e c t i o n was e f f e c t i v e i n the opposite d i r e c t i o n because there were c e r t a i n a l l e l i c states of the modifiers present i n the cn bw and populations that mutated at a very low frequency and did not enhance the degree of d i s t o r t i o n . Sandler and Hiraizumi (1960a) made i t quite clear that they could con-ceive of two possible r o l e s of the s t a b i l i z e r that were consistent with t h e i r data. One r o l e would be simply to increase the degree of d i s t o r t i o n a s sociat-ed with any given SD state. This would cause s t a b i l i z a t i o n because almost a l l states would then have k values of 1.0 and consequently there would be no opportunity for male to male v a r i a t i o n . The other possible r o l e they suggest i s to decrease the mutation rate to the lower k states. This would r e s u l t i n an accumulation of males with high k states'. However, as I stated i n the pre-vious paragraph, I f e e l that i n order to explain the s e l e c t i o n of an unstable l i n e from a semistable l i n e there must e x i s t c e r t a i n a l l e l i c forms of the modifiers that do not mutate and do not enhance k. If t h i s were the case and the s t a b i l i z e r operated by preventing mutation to low k states, then stable l i n e s should have bimodal k d i s t r i b u t i o n s . Since they do not, t h i s l a s t pos-s i b l e r o l e for the s t a b i l i z e r i s not consistent with the data a v a i l a b l e . Miklos and Smith-White (1971) have c r i t i c i z e d the hypothesis of Hiraizumi and Sandler (1960a) on the grounds that i t \"involves the introduction of new concepts; of jSD states, SD-state mutation, and a s t a b i l i z e r which controls t h i s mutation\". As an a l t e r n a t i v e to these new concepts, Miklos and Smith-White (1971) propose a model that predicts increasing variances of k with de-creasing mean values of k, i n the range of k's observed by Sandler and Hiraizumi (1960a). Acceptance of t h i s model also requires that one accept that mean k i s a deceptive measure of the degree of d i s t o r t i o n of a given l i n e . This i s why th i s model i s being examined i n t h i s chapter. Before proceeding to describe the model, I should point out that the model does not i n any way attempt to explain the pertinent observations of Sandler and Hiraizumi (1960a) that led them to hypothesize 'SD state mutations. The concept of SiD state mutations i s the only r e a l l y novel concept that Miklos and Smith-White should have c r i t i c i z e d . The concept of SD states i t s e l f should not be construed as being novel. The term _SD state, as used by Sandler and Hiraizumi (1960a), i s surely meant to imply the p o t e n t i a l d i s t o r t i n g a b i l i t y of a given male, depending upon that males genotype with respect to the modifiers of SD. Also, one cannot c r i t i c i z e Sandler and Hiraizumi's ent i r e model on the basis of the proposed r o l e of the s t a b i l i z e r i n reducing the frequency of mutation to lower k states, since an a l t e r n a t i v e r o l e for the s t a b i l i z e r was c l e a r l y proposed, that i s , increasing the k value of a l l SD states. This leaves only the concept of SD state mutation to be c r i t i c i z e d . SD state mutation was proposed i n order to explain the observation that a semistable l i n e could be made unstable by s e l e c t i o n , but an unstable l i n e could not be made semistable by s e l e c t i o n . Miklos and Smith-White (1971) do not discuss t h i s issue at a l l , instead they concentrate s o l e l y on describing the re l a t i o n s h i p between mean k and the variance of k. Miklos and Smith-White (1971) use the concept of \"make\" i n developing t h e i r model. This concept i s outlined i n Rendel's (1967) analysis of the development of s c u t e l l a r b r i s t l e s . Rendel defined make (m) as the resultant of a l l factors leading to the development of a s c u t e l l a r b r i s t l e . The u t i l i t y of such a concept i s that i t forces one's attention upon the resu l t a n t , and i n so doing does not require unnecessary speculation about the magnitude of var i a b l e s that one cannot measure. Miklos and Smith-White (1971) define make as the \" t o t a l l e v e l of a l l systems leading to the e x t i n c t i o n 'of SD + i n hetero-zygous males\". The proportion of SD + sperm that f a i l to function, c a l l e d e, i s r e l a t e d to k as follows: e = 2 - (^). . They further reason that \"Extinc-t i o n or non-extinction of SD + involves a l t e r n a t i v e responses, and therefore there must be some l e v e l of m which acts as a threshold or switch point. This threshold i s denoted P, and the e x t i n c t i o n c o e f f i c i e n t measures the proportion of gametes i n which m exceeds P\". However, i t would appear that they have not used the term make as Rendel (1967) proposed that i t should be used, because they have resolved two factors contributing to sperm dysfunction, make and threshold of make. They next assume that the make values of the sperm within a given male are normally d i s t r i b u t e d and that the thresholds of make of a l l sperm within a male do not vary. (The l a t t e r assumption I f i n d rather d i f f i c u l t to accept. This point w i l l be returned•'to l a t e r . ) Given these assumptions the e x t i n c t i o n c o e f f i c i e n t i s defined by: If one measures P i n standard deviation units from the mean, then one can deal with a normal d i s t r i b u t i o n with a mean of zero and a standard deviation of one. I.e. \\ -h m^ e S, P-m / yjr- dm . See Figure 3 for a diagram showing the r e l a t i o n s h i p between P-m e and Q, . One can see that i f e varies between males, then the best v a r i a b l e m „ — P-m for measuring t h i s v a r i a t i o n w i l l be i . e . , the number of standard devia-nt — P-m tions of P from the mean of the make d i s t r i b u t i o n . The value -py- can be ^m c a l l e d m', the make l e v e l of an i n d i v i d u a l male. If m' values between males are assumed to have a normal d i s t r i b u t i o n , then one can examine the d i s t r i b u t i o n of k values between males at d i f f e r e n t mean k values, while maintaining the variance of m' constant. I t w i l l be easier, however, to j u s t examine the variance of k. The variance of k and the variance of m' are approximately re l a t e d as follows: Var k = (dk/dm') 2Var m' 1 i 2 where dk/dm' = k 2 e ^ m ^ . The re l a t i o n s h i p between dk/dm' and mean k i s STfr shown i n Figure 4. I t i s apparent that the variance of k w i l l increase as k i s reduced to 0.77, then the variance of k w i l l begin to decrease again. One should note here that these variances do not take into account binomial sampling variance. FIGURE 3 The relationships between the distribution of make values within a male (m), the threshold of dysfunction (P), the proportion of SD+ sperm that dysfunction (e), and the make level of a male (m1). Make (m) is measured in standard deviations from the mean of the distri bution. T2 22 FIGURE 4 The relationship between dk/dm1 and mean k. E 24 Miklos and Smith-White (1971) proposed t h i s model i n order to describe the increased v a r i a b i l i t y of bw-bearing SD recombinants, as compared to unre-combined £>D l i n e s , without the necessity of hypothesizing mutable _SD states. However, Sandler and Hiraizumi (1960a) did not hypothesize mutable &D states i n order to describe the increased v a r i a b i l i t y of bw-bearing SD recombinants, rather they were hypothesized i n order to explain the observation that un-stable l i n e s could not be made semistable by s e l e c t i o n . Sandler and Hiraizumi (1960a) had already proposed a p e r f e c t l y reasonable hypothesis to explain the increased v a r i a b i l i t y of bw-bearing SD recombinants. This hypothesis was that the s t a b i l i z e r of SD operated by simply increasing the k values of a l l SD states. When the s t a b i l i z e r was absent the k values of a l l JSD states would be lowered so that v a r i a b i l i t y between the states could then be observed. It i s c l e a r l y of importance to choose between these two hypotheses, since acceptance of Miklos and Smith-White's hypothesis requires that one should measure k as m'. In order to make a choice one must f i r s t examine the weak points of these hypotheses. In t h i s respect, I f e e l that Miklos and Smith-White's assumption that the thresholds of m within a male are constant, while m may vary normally, i s d e f i n i t e l y suspect. I cannot think of any a_ p r i o r i reason why the thresholds of m should be any less v a r i a b l e than m. Since there i s no way of measuring ei t h e r P or m, at present, i n order to evaluate the generality of the model one should ask: what are the e f f e c t s of P varying within males? I f the variance of P equaled the variance of m, as i n Figure 5, then one can see i n t u i t i v e l y that the variance of k w i l l be smallest at k equal to 0.5, but w i l l be greatest close to a k of 1.0. However, at some point close to 1.0 a ce r t a i n proportion of males w i l l have k equal to exactly 1.0, i . e . a l l m are greater than P. This truncation e f f e c t w i l l begin to re-duce the variance of k again as mean k approaches 1.0. Nevertheless, the way i n which variance of k changes with mean k w i l l be considerably d i f f e r e n t i f FIGURE 5 A model similar to Miklos and Smith-White's, except P is allowed to vary and ^ equals (f^. The units on the abscissa are standard deviations of m from the mean of each distribution. one relaxes the assumption that P i s constant. In addition, i f P v a r i e s , then the r e l a t i o n s h i p between k amd m', as defined by Miklos and Smith-White (1971) , w i l l be al t e r e d . In the remainder of t h i s chapter I w i l l present some r e s u l t s from large scale experiments that demonstrate how the variance of k i s rela t e d to mean k and then I w i l l compare these r e s u l t s with predictions made by the model of Miklos and Smith-White (1971) as we l l as with predictions made by a model I s h a l l develop, based on my i n t e r p r e t a t i o n of Sandler and Hiraizumi's (1960a) discussion of i n s t a b i l i t y . 28 MATERIALS AND METHODS The d i s t r i b u t i o n of segregation r a t i o s of s i x d i f f e r e n t types of SD second chromosomes were determined when the chromosomes were heterozygous with a cn bw chromosome. A l l s i x chromosomes were o r i g i n a l l y maintained as balanced stocks over In(2LR)SMl,Cy, an e f f e c t i v e balancer for chromosome 2 (see Lindsley and G r e l l 1968). In order to approximate isogenic backgrounds between the d i f f e r e n t stocks, a l l of the balanced l i n e s were backcrossed : as males to cn bw/ cn bw females for four generations. Af t e r the fourth backcross generation males that were from two to four days old were c o l l e c t e d and mated to females that were homozygous for the genetic markers, cn bw; K i pP bx sr e s. Since these females were homozygous for K i pP bx sr e s, one could e a s i l y d i s t i n g u i s h a white-eyed SD + f l y , which would be heterozygous for K i p p bx sr e s , from a white-eyed f l y that arose from a non-virgin female. Single \"SD\"/cn bw males were placed with two homozygous cn bw; K i pP bx sr e s v i r g i n females i n s h e l l v i a l s containing standard Drosophila medium. The parents were l e f t i n the v i a l s for four days and then discarded. The pro-geny were scored 12 to 13 days l a t e r . The temperature was maintained at 24+1 C dur ing the experiment. The modified SD_ chromosomes used were derived from SD mapping experiments that w i l l be described i n the next chapter. The following £>D chromosomes were examined: 1. SD-5, an unrecombined SD chromosome. 2. R(SD-5) pk cn, a recombinant of SD-5 with almost a l l of 2R replaced by the r i g h t arm of a b pr I t pk cn chromosome. This recombinant should have l o s t the s t a b i l i z e r of Sandler and Hiraizumi (1960a). 3. Df(2L)(SD-5)-8, a 2L heterochromatic deficiency on SD-5 for Group VIII (It) and the Group VII s i t e including EMS 56-4 (see H i l l i k e r 1976) . RR(SD-5)It, a double recombinant of SD-5 i n which the centromeric heterochromatin of SD-5 was replaced by that of the b pr It pk cn chromosome. R(SD-5) b pr-5, a recombinant derived from SD-5 and b pr It pk cn. RR(SD-5) pr I t , a double recombinant of SD-5 and b pr It pk cn. RESULTS AND DISCUSSION The observed frequency d i s t r i b u t i o n s of k for the s i x d i f f e r e n t \"SD\" chromosomes are shown i n Figure 6. I t i s apparent that as mean k decreases the variance of k increases f o r the four chromosomes with mean k above 0.5 and then the variance of k again decreases for the two low k chromosomes. Sandler and Hiraizumi (1960a) suggested that one way the s t a b i l i z e r might operate i s by simply increasing the k of a l l SD states such that most states would then have k's of 1.0 and thus v a r i a b i l i t y would not be observed. In order to examine t h i s proposal more c a r e f u l l y , I s h a l l develop i t quantita-t i v e l y . One can s t a r t by assuming that a number of modifiers of d i s t o r t i o n , both genetic and environmental, act i n such a way that the p o t e n t i a l k's of a given population of males w i l l be normally d i s t r i b u t e d . Furthermore, assume that t h i s d i s t r i b u t i o n of p o t e n t i a l k's i s unaffected by changes i n mean k. This w i l l undoubtedly not be true as mean k approaches 0.5, because when mean k equals 0.5, SD i s not operating and one would riot expect modifiers of SD to have any e f f e c t upon the p o t e n t i a l segregation r a t i o . However, at high mean k's another factor comes into play, and i t i s t h i s factor which I believe Sandler and Hiraizumi (1960a) alluded to. Since the variance of p o t e n t i a l segregation r a t i o s i s unaffected by mean k, at high mean k's a ce r t a i n per-centage of the males w i l l have p o t e n t i a l segregation r a t i o s greater than or equal to 1.0. Cl e a r l y a l l p o t e n t i a l segregation r a t i o s greater than or equal to 1.0 should be placed together into one group with a p o t e n t i a l segregation r a t i o of 1.0 i . e . no chance of recovering an SD + sperm. This truncates the normal d i s t r i b u t i o n of p o t e n t i a l k values. As the mean of the untruncated d i s t r i b u t i o n increases, a greater percentage of the population f a l l s into the FIGURE 6 The observed k distributions of the 6 \"SD\" chromosomes tested The unit of measurement on the ordinate is the percentage of males having k's within a given interval. The unit of the abscissa is k k interval is 0.03. n = the number of males tested Hn = the harmonic mean number of males = the unweighted mean of k's of all the males &> •= the standard deviation of the k's 32 100-i 90-70-5 0 -3 0 -10-o + S D - 5 n = 1148 H k =.985 H n =61 .3 CTk=.022 0 .1 .3 J 40H 30-1 20-J 10^ R(SD-5)pk cn n =1156 Hn-\"60.4 r .932 (Tk = 056 .9 1j0 yt-rffl 0 .1 T\" 9 1.0 25-1 20-Df(2L)(SD-5)-8 15-10H 5-^ n*1143 JJ k =.879 H n = 6 5 . 2 =.067 0 .1 .3 .7 .9 1.0 25n 20H 15-10J 5 H RR(SD-5)lt 11=1142 Hn=55.8 \" k = .713 V .506 25-2 0 -15-10 -5 -•5 .7 !I i*.o RR(SD-5)pr If n = 1145 H n -55.2 Mk .463 0 .1 tk T 9 ilo truncated portion and t h i s i n turn reduces the variance of the truncated d i s -t r i b u t i o n . Thus, on t h i s model the s t a b i l i z e r i s only a strong p o s i t i v e modifier of that s h i f t s up the d i s t r i b u t i o n of p o t e n t i a l k values such that the majority are truncated at 1.0. This model predicts that p o t e n t i a l segregation r a t i o s w i l l be d i s t r i b u t e d as truncated normal d i s t r i b u t i o n s . However, observed segregation r a t i o s have a binomial sampling component to t h e i r d i s t r i b u t i o n s and t h i s must be taken into account when observing actual d i s t r i b u t i o n s of segregation r a t i o s . In order to check i f the p o t e n t i a l segregation r a t i o s follow truncated normal d i s t r i b u t i o n s one can use a semigraphical method outlined by Sokal and Rohlf (1969). This involves p l o t t i n g the cumulative frequencies of successive i n t e r v a l s of a d i s t r i b u t i o n versus the mean value of the i n t e r v a l . Such a p l o t w i l l be s-shaped i f the d i s t r i b u t i o n i s normal. I f , instead of cumula-t i v e frequencies of successive i n t e r v a l s , one uses the probit transformation of that frequency, then the p l o t w i l l follow a s t r a i g h t l i n e i f the d i s t r i b u -t i o n i s normally d i s t r i b u t e d . For the present purposes, t h i s technique has been modified somewhat. The i n d i v i d u a l k values were ranked from the lowest to the highest, and then the quotient of the rank divided by the t o t a l number of observations was subjected to the inverse normal i n t e g r a l transformation and then these values were i n d i v i d u a l l y plotted versus t h e i r respective k's. The points were then joined by s t r a i g h t l i n e s . A l l of t h i s was performed on a PDP 11 computer. The graphs are shown i n Figure 7. Fig.- 8' shows graghs con-structed i n the same manner, except that k was measured as a r c s i n Vk i n order to attempt to remove the e f f e c t s of binomial sampling. The angular tr a n s f o r -mation i s e f f e c t i v e i n normalizing a binomial d i s t r i b u t i o n . However, these k d i s t r i b u t i o n s are a mixture of a binomial component, and a b i o l o g i c a l compo-nent. I have hypothesized that the b i o l o g i c a l component follows a truncated FIGURE 7 A diagrammatic test for normality of the k distributions of the 6 \"SD_\" chromosomes tested. The unit on the ordinate is standard deviations and the unit on the abscissa is k. 35 + 4 n SD-5 + 4 T RfSD-5)pk cn + 2-- 2 H - 2 H -1 1.0 H 1 1 1 1 .2 4 .6 .8 1.0 + 4 -+ 2-Df(2t)(SD-5)-8 OH 2H - 4 -.2 f 4-1 + 2H -2H -T 1.0 - 4 RRfSD-5) /f .6 10 + 4 - R(SD-5)b pr-5 + 4 T RR(SD-5)pr /f + 2- + 2H O H - 2 H -2H - 4 - T .2 .4 -1 1.0 - 4 -.8 1.0 36 FIGURE 8 A diagrammatic test for normality of the distributions of arc sin of the 6 \"SD\" chromosomes tested. The unit on the ordinate is standard deviations and the unit on the abscissa is arc s inVT. 37 + 4-| +2-oH •2H SD-5 20 40 60 80 + 4 + 2H OH -2H -4-R(SD-5)pk cn 90 20 40 60 80 90 + 4-i + 2H OH -2H Df(2L)(SD-5)-8 •4 + 0 20 40 60 80 90 + 4. + 2H OH -2H -4-RR(SD-5)/f 20 -1 1 1 — 1 40 60 80 90 + 4-t R ( S D - 5 ) b p r - 5 + 4 - , RR(SD-5)pr It + 2H + 2H OH OH 2H -2H - 4 . r 20 4 0 60 80 90 - 4 -20 40 60 \"~I 1 80 90 38 normal d i s t r i b u t i o n . Because the angular transformation i s applied to the hybrid d i s t r i b u t i o n , i t w i l l only be approximately e f f e c t i v e i n removing the e f f e c t s of binomial sampling. The d i s t r i b u t i o n s of a r c s i n >/k shown i n Figure 8 are truncated normal d i s t r i b u t i o n s . These observations support the hypothe-s i s that the b i o l o g i c a l component of v a r i a t i o n follows a truncated normal d i s -t r i b u t i o n , but the support should not be considered to be too strong, because of the previously mentioned reservations concerning the effectiveness of the angular transformation i n t h i s case. From Figure 8 i t i s also apparent that the slopes of the resolved portions of the d i s t r i b u t i o n s of the four \"SD\" stocks with mean k's above 0.5 are approximately equal. Although one might be tempted to suggest that t h i s indicates that the variances of the untruncated p o t e n t i a l k d i s t r i b u t i o n s of the four \"SD\" stocks with k's above 0.5 are equal, I would h e s i t a t e to do so, again because of reservations concerning the e f f i c a c y of the angular transformation i n removing the e f f e c t s of binomial sampling from an observed k d i s t r i b u t i o n . From the d i s t r i b u t i o n s shown i n Figure 8, i t i s not at a l l unreasonable to assume that the p o t e n t i a l segregation r a t i o s approximately follow truncated normal d i s t r i b u t i o n s . If t h i s assumption i s made, then one can determine whether or not truncation alone can account f o r changes i n the observed v a r i -ance of k between the \"SD\" stocks with mean k greater than 0.5. Let j be the po t e n t i a l segregation r a t i o of a male. Binomial sampling from j gives the ob-served segregation r a t i o k. The var i a b l e j has a normal d i s t r i b u t i o n : Since j i s an abstraction, i t can possess what seem to be rather peculiar pro-p e r t i e s . One of these p e c u l i a r properties i s that i t can have values less than 0 or greater than 1, as w e l l as the usual range from 0 to 1. The more j exceeds 1, the \"more impossible\" i t becomes to recover a f l y with the SD + chromosome. P r a c t i c a l l y , however, one cannot d i s t i n g u i s h between d i f f e r e n t degrees of i m p o s s i b i l i t y . Thus, although i n abstract terms j follows a normal d i s t r i b u t i o n , i n p r a c t i c a l terms j follows a truncated normal d i s t r i b u t i o n , with the d i s t r i b u t i o n being truncated at 0 and 1. What one wishes to deter-mine i s how the mean and variance of the truncated d i s t r i b u t i o n vary, when the mean of the untruncated d i s t r i b u t i o n i s altered but the variance of the un-truncated d i s t r i b u t i o n i s maintained at a constant value. In order to s i m p l i f y the mathematics of the problem, I s h a l l transform j i - &i xnto z, where z = — — — . Now the v a r i a b l e z w i l l follow a normal d i s t r i b u -te. 3 z 2 t i o n with a mean of 0 and a standard deviation of 1; -f(z) = e . Instead Jrfr i - Mi of being truncated at 1, t h i s d i s t r i b u t i o n w i l l be truncated at — — — , which w i l l be designated as z'*\". The mean of the truncated z d i s t r i b u t i o n , /J. , can be determined by d i v i d i n g the d i s t r i b u t i o n into two parts, one part le s s than z*- and the other part greater than z f c. The part greater than z1\" w i l l , because of truncation, have a value of z1\". The proportion of the t o t a l d i s t r i b u t i o n contained i n t h i s component i s given by ~ 2 z 2 ^ dz Z2 Thus, t h i s component w i l l contribute z f c e 2 dz towards the mean of the truncated z d i s t r i b u t i o n . The part of the d i s t r i b u t i o n with values of z less z'fc w i l l c o n stitute: 2 -oo z \"2 , = \\ Z dz e dz of the t o t a l d i s t r i b u t i o n . In order to calculate the mean of t h i s part of the d i s t r i b u t i o n one must define t h i s part i n such a way that the area under t h i s part of the d i s t r i b u t i o n i s equal to 1. This can be done by multiplying f ( z ) -1 by / ^ [ a . z d z lzI . The mean of a d i s t r i b u t i o n can be given by ^ (x) • f(x) dx, i f ^ f(x) dx = 1. Thus the mean of the part of the z d i s t r i b u t i o n l e s s than z*\" i s given by: ,t N - l J~ 00 t z Z dz / \\ \\ z • Z dz / . Since t h i s constitutes ^ Z dz of the t o t a l d i s t r i b u t i o n , t h i s part of the d i s t r i --00 t czbution contributes \\ z . Z dz towards the mean of the truncated z U-o° d i s t r i b u t i o n . Thus the mean of the truncated z d i s t r i b u t i o n i s given by: Z dz / + z t 2 2 The s o l u t i o n of the i n t e g r a l on the r i g h t i s -e , therefore YTTF . t 2 oo \\ __z The variance of the truncated z d i s t r i b u t i o n can be determined i n much the same manner, i . e . by d i v i d i n g the d i s t r i b u t i o n into two parts, one part less than z c and the other part greater than zt. In t h i s case one uses the re l a t i o n s h i p O\"^ = C*3 f ( x ) ( x - A * x ) ^ dx, where f(x) dx = 1. Using t h i s procedure: . t2 = t t2 One now knows how to c a l c u l a t e fj( and (J . In terms of j these para-Z 3 meters are JU ^ = Mfc CT. + JU. and = . In order to compare parameters predicted i n terms of j with parameters measured i n terms of k one ' 2 must attempt to remove the e f f e c t s of binomial sampling from yU and £7^ This was done by estimating the variance of a binomial d i s t r i b u t i o n with 6 equal to jU , and the number of progeny per v i a l the same as i n the observed •l 10 aoueueA (1971) model. The agreement between t h i s p r e d i c t i o n and the observed data points i s not much better. In p a r t i c u l a r , the variance of Df(2L)(SD-5)-8/cn bw males i s far below the p r e d i c t i o n of e i t h e r model. .Since the variance and mean of RR(SD-5) l t / c n bw males was used as the reference point through which the curves of both models passed, i t i s j u s t as l i k e l y that, as compared to males with mean k's above the mean k of RR(SD-5) l t / c n bw males, the SD_ system i n RR(SD-5) l t / c n bw males was abnormally s e n s i t i v e to modifiers of SD. 2 Curve a i n Figure 9 was constructed on the assumption that Cf ^ was con-stant for a l l values of / U j . However, as I previously mentioned, when >Uj i s near 0.5 the E!D system i s either not operating, or only operating at a very low l e v e l . Under such circumstances modifiers of segregation d i s t o r t i o n should 2 have no e f f e c t , i . e . U . should be zero. The observations shown i n Figure 9 2 support t h i s contention. Thus i t would appear that CT . must increase as /-(_. _2 increases above 0.5. Curve a i n Figure 9 i s based on the assumption that CT does not change a f t e r i t reaches a value of 0.01 at some point less than jU.. equal to 0.7. Since curve a i n Figure 9 does not f i t the data, the assumption used to construct i t must be i n c o r r e c t . Perhaps the model would be improved 2 i f O . was i n some manner dependent upon jU... The simplest assumption i s that 2 2 i s d i r e c t l y proportional to (M j ~ 0'5), i . e . Cf ^ = C ( / U j - 0.5). Again, the variance and mean of RR(SD-5) l t / c n bw males can be used to estimate c as equal to 0.047. Curve c i n Figure 9 shows the p r e d i c t i o n when t h i s assumption i s applied to the model used to construct curve a. Although t h i s modification renders the model more r e a l i s t i c at low values of jU^> I t does not f i t the observations any better at high values of JJ, ^ . Since none of the predictions shown i n Figure 9 f i t the observations, i t i s l i k e l y that the JSD chromosomes examined i n t h i s study d i f f e r i n t h e i r sen-s i t i v i t i e s to modifiers of segregation d i s t o r t i o n . I f t h i s i s the case, i t must be a consequence of the particular mutations on each of these chromosomes. In this regard, i t is interesting to note that RR(SD-5)lt and Df(2L)(SD-5)-8 differ in that the former chromosome was constructed by recombination, while the latter chromosome bears a deficiency of the heterochromatin close to, and including, the Lt locus. CONCLUSION Although the data obtained i n t h i s study do not enable one to choose between the models I have proposed and that of Miklos and Smith-White (1971), I favour my models because t h e i r common basic assumption i s open to d i r e c t r e f u t a t i o n . This common basic assumption i s that the p o t e n t i a l segregation r a t i o s (j) i n a population of SD males follow a truncated normal d i s t r i b u t i o n . I attempted to test t h i s assumption, as shown i n Figure 8. Figure 8 shows that i n a l l cases arc s i n k i s d i s t r i b u t e d as a truncated normal d i s t r i b u -t i o n , but, as previously indicated, I have reservations concerning the e f f i c a c y of the angular transformation i n t h i s s i t u a t i o n , and, accordingly, I consider Figure 8 to be rather weak evidence i n support of the assumption that j follows a truncated normal d i s t r i b u t i o n . However, the important point i s that, given enough time, i t i s conceivable that one could compute the exact frequency d i s t r i b u t i o n of k expected for a truncated normal d i s t r i b u t i o n of j and a given d i s t r i b u t i o n of numbers of progeny per v i a l . Thus t h i s assumption, although not rigourously tested here, i s at l e a s t p o t e n t i a l l y v e r i f i a b l e . On the other hand, Miklos and Smith-White's (1971) model i s predicated upon three basic assumptions. The f i r s t assumption i s that the values of m within a male follow a normal d i s t r i b u t i o n . The second assumption i s that within a given male a l l values of P are i n v a r i a n t . The t h i r d assumption i s that the values of m' between males of a given population follow a normal d i s t r i b u t i o n . The f i r s t two assumptions cannot be tested d i r e c t l y at present, and l i k e l y they never w i l l be able to be tested d i r e c t l y . Furthermore, one cannot ascertain the d i s t r i b u t i o n of a v a r i a b l e l i k e m' u n t i l one f i r s t affirms the v a l i d i t y of the v a r i a b l e , and I have j u s t pointed out that t h i s cannot be done. The only way to test the v a l i d i t y of t h i s t r i a d of assump-tions i s to test the predictions of the e n t i r e model. In support of t h e i r model Miklos and Smith-White (1971) compared calcu-lated frequency d i s t r i b u t i o n s of k, which they generated using t h e i r model, with observed frequency d i s t r i b u t i o n s of k. The comparison they made was i n -correct, however, because t h e i r predicted d i s t r i b u t i o n s of k did not take into account binomial sampling, while the observed' k d i s t r i b u t i o n s n e c e s s a r i l y had a binomial component. I f one disregards t h i s gross oversight, i t i s apparent from t h e i r published diagrams that the observed and \"predicted\" d i s t r i b u t i o n s agreed reasonably w e l l . As further support for t h e i r model, Miklos and Smith-White (1971) compared the predicted r e l a t i o n s h i p between the standard devia-t i o n of k and the mean of k with the observed r e l a t i o n s h i p . Their p r e d i c t i o n 2 imposed another assumption on t h e i r model, the assumption that Cf , remained m constant f o r a l l values of jU ^ . From the data they presented, the observed r e l a t i o n s h i p agreed reasonably w e l l with the predicted r e l a t i o n s h i p . However, t h e i r model, with a l l four assumptions, did not agree with my observations, as 2 shown iri Figure 9. Thus the assumption that Cf' , remains constant i n f l i e s with s i m i l a r genetic backgrounds i s not generally v a l i d . Proof for the v a l i d i t y of t h e i r model must re s t on predictions made by the o r i g i n a l t r i a d of assumptions. This b o i l s down to comparing observed d i s t r i b u t i o n s of k with predicted d i s t r i b u t i o n s of k and as I have already pointed out, they f a i l e d to do t h i s adequately. Because my model has only one basic assumption, while Miklos and Smith-White's (1971) model has three basic assumptions, my model i s inherently simpler. For t h i s reason, further experiments should concentrate upon attempt-ing to refute my hypothesis, i n preference to Miklos and Smith-White's. I f i t could be adequately v e r i f i e d that j follows a truncated normal d i s t r i b u t i o n , 2 then one could confidently examine changes of Cf .. between d i f f e r e n t SD stocks and t h i s might enable one to use fj. as a measure of d i s t o r t i o n , i n l i e u of k. However, at present, not enough i s known about d i s t r i b u t i o n s of k i n order to justify any measure of distortion other than k. Accordingly, in what follows I will use either k or k as the measure of distortion. CHAPTER II ON THE LOCATION AND PROPERTIES OF SOME OF THE LOCI AFFECTING SEGREGATION DISTORTION 50 INTRODUCTION To introduce the reader to my r e s u l t s on the l o c a t i o n of some of the components of the SD system, I s h a l l f i r s t discuss the r e s u l t s of H a r t l (1974), since h i s experiments have avoided some of the complications that had confound-ed e a r l i e r studies on t h i s topic (Hartl and Hiraizumi 1976). H a r t l (1974) examined the properties of recombinants recovered from R(SD-36)-l/Tft cn females. The T f t cn chromosome was s p e c i f i c a l l y chosen as i t c a r r i e s no suppressors of SD. The k value of R(SD-36)-l/Tft cn males was 0.95. The r e -combinants bearing T f t f e l l into three classes based on t h e i r a b i l i t y to d i s -t o r t cn__bw and t h e i r s e n s i t i v i t y to another J3D chromosome. These three classes were: 1) i n s e n s i t i v e d i s t o r t e r s (unrecombined chromosomes are i n -s e n s i t i v e d i s t o r t e r s ) , 2) i n s e n s i t i v e nondistorters, and 3) s e n s i t i v e non-d i s t o r t e r s . The cn recombinants also f e l l i nto three classes. The f i r s t c lass was composed of i n s e n s i t i v e d i s t o r t e r s . The second and t h i r d classes were s e n s i t i v e nondistorters. These s e n s i t i v e nondistorters could be further subdivided depending upon t h e i r response to the i n s e n s i t i v e nondistorter T f t recombinants. One class of the s e n s i t i v e nondistorter cn recombinants con-tained chromosomes that were themselves d i s t o r t e d by i n s e n s i t i v e nondistorter T f t recombinants. The chromosomes of the other class of s e n s i t i v e nondistorter cn recombinants were not d i s t o r t e d by the i n s e n s i t i v e nondistorter T f t recom-binants . These r e s u l t s suggest that two _SD l o c i are located between T f t and cn. One of these i s c a l l e d Sd_ and i s located to the l e f t of the second locus c a l l -ed Rsp (Responder). A Rsp + a l l e l e responds to a £>d a l l e l e , r e s u l t i n g i n the dysfunction of sperm bearing Rsp +. If a male bears a Rsp + a l l e l e , but does not have a Sd_ a l l e l e , then there i s nothing to cause the Rsp\"*\" a l l e l e to res-pond, and again segregation w i l l appear normal. If segregation i s to appear 51 abnormal, there must be at l e a s t one a l l e l e and one each of Rsp and Rsp +. Thus the recombinants can be c l a s s i f i e d genotypically as follows: T f t insen-s i t i v e d i s t o r t e r s are Sd Rsp, T f t i n s e n s i t i v e nondistorters are Sd + Rsp, T f t s e n s i t i v e nondistorters are Sd + Rsp +, cn i n s e n s i t i v e d i s t o r t e r s are Sd Rsp, and cn s e n s i t i v e nondistorters can be divided into two classes since Sd Rsp + + + + + i s d i s t o r t e d by T f t Sd Rsp while Sd Rsp i s not d i s t o r t e d by T f t Sd Rsp. A 4- + s e n s i t i v e chromosome such\" as cn bw, therefore, i s Sd Rsp . Ganetzky (1977) has extended Hartl's (1974) analysis by using r a d i a t i o n induced deletions rather than recombinants to map the components of SD. Ganetzky (1977) i r r a d i a t e d SD-5 males with 5000 rads from a cobalt 60 source and then tested the treated chromosomes for t h e i r a b i l i t y to d i s t o r t cn bw. From 4,000 tested chromosomes, eight were recovered that evidently f a i l e d to d i s t o r t cn bw. However, when tests for d i s t o r t i o n of cn bw were repeated using female controls (as I have discussed i n Chapter I) i t was found that three of these had corrected k values of approximately 0.7, while the other f i v e had corrected k's near 0.5. Those SJJ semi-revertants that had Kc's (corrected k values) near 0.7 were also a l l mutant for I t . Furthermore, when they were complemented with a ser i e s of proximal deletions i n 2L ( H i l l i k e r and Holm 1975), these SD chromo-somes behaved as i f they a l l had deletions extending from the p o s i t i o n of Group IX l e t h a l s to Group VII l e t h a l s ( H i l l i k e r 1976). These r e s u l t s suggest-ed that an enhancer of SD i s located i n the centromeric heterochromatin of 2L. Ganetzky (1977) designated t h i s locus as E(SD). The f i v e complete SD revertants were tested for t h e i r a b i l i t y to induce s e l f d i s t o r t i o n of a Sd Rsp + chromosome. It was found that four of them did cause Sd Rsp to dysfunction, but one did not. Supposedly the one that did not induce s e l f d i s t o r t i o n of Sd Rsp + was a powerful suppressor of SD. How-ever, the other four were candidates for mutations at the SD_ locus. These four chromosomes were complemented with a ser i e s of deletions covering the base of 2L from polytene chromosome band 36F to 40A. They were also examined c y t o l o g i c a l l y . Three of the four chromosomes were f u l l y v i a b l e with the dele-tions and f a i l e d to show any new c y t o l o g i c a l abberations. The fourth chromo-some, however, was l e t h a l with some of the deletions and i n addition a de l e t i o n was observed on t h i s chromosome that extended from 38A6-B2 on the r i g h t to 37D2-7 on the l e f t . This d e l e t i o n i s located between the s i t e s of T f t and pr and i t suggests that the JJd. s i t e of H a r t l (1974) i s located there also. Ganetzky (1977) also tested the s e n s i t i v i t y , i n males, of 5,160 cn bw chromosomes that were i r r a d i a t e d with 5000 rads of gamma rays. Five chromo-somes were recovered that were i n s e n s i t i v e to d i s t o r t i o n by SD-72. Three of these chromosomes had acquired a new recessive l e t h a l and complementation tests revealed the three l e t h a l s to be a l l e l i c . By recombination the newly induced i n s e n s i t i v i t y s i t e was placed between pr and cn. This i s approximately the region where H a r t l (1974) placed Rsp. If the newly induced i n s e n s i t i v i t y a l l e l e s are newly induced Rsp a l l e l e s , then these i n s e n s i t i v e chromosomes should cause d i s t o r t i o n of Sd Rsp +. Ganetzky found t h i s indeed to be the case. Since three of the i n s e n s i t i v e cn bw chromosomes had the same recessive l e t h a l i n common, i t was decided that mapping t h i s l e t h a l should provide i n -formation on the l o c a t i o n of Rsp. Accordingly, these three chromosomes were complemented with a ser i e s of deletions and point mutations of the centromeric heterochromatin of 2R ( H i l l i k e r and Holm 1975; H i l l i k e r 1976). I t was found that the common recessive l e t h a l was a mutation at the r l locus. One of these chromosomes was mutant only at _ r l , while the other two appeared to be dele-tion s , one extending into Group I l e t h a l s and the other extending both into Group I and into Group I I I l e t h a l s ( H i l l i k e r 1976). These data alone suggest that Rsp i s located near i r i . However, H i l l i k e r (personal communication) has tested h i s proximal 2L and 2R d e f i c i e n c i e s and has found that at l e a s t one de f i c i e n c y from each Group i s s e n s i t i v e to d i s t o r -t i o n . Furthermore, Df(2R)M-S2^, which appears to be both c y t o l o g i c a l l y and g e n e t i c a l l y d e f i c i e n t f o r a l l of the 2R centromeric heterochromatin ( H i l l i k e r and Holm 1975), i s s e n s i t i v e to d i s t o r t i o n (Ganetzky 1977). In view of these r e s u l t s , i t i s most unreasonable to maintain that Rsp i s located near r _ l . In order to explain h i s r e s u l t s i n view of H i l l i k e r ' s f i n d i n g s , Ganetzky (1977) proposes that the Rsp s i t e may be located proximal to Group I. I f t h i s were the case, then the i n s e n s i t i v e cn bw chromosome that was l e t h a l only for Group II would have to be the r e s u l t of a double \" h i t \" of some type, one h i t causing l e t h a l i t y at Group II and the other h i t mutating the Rsp s i t e which i s p r o x i -mal to Group I. Ganetzky also points out that, conceivably, Rsp might occupy d i f f e r e n t s i t e s on d i f f e r e n t chromosomes. In the remainder of t h i s chapter I s h a l l present the r e s u l t s of my f i n d -ings on the l o c a t i o n and properties of j3d, E(SD) , and Rsp. MATERIALS AND METHODS 54 The SD chromosomes used i n t h i s study were SD-5 and SD-72. These chromo-somes were obtained by David Holm from the Pasadena stock centre about ten years ago. SD-5 has been maintained as a balanced stock over In(2LR)SMI, Cy and SD-72 over In(2LR)SM5, Cy since that time. SD-5 has two inversions, In(2R)45C-F; 49A and In(2R)NS, while SD-72 has In(2R)NS and a p e r i c e n t r i c i n -version, In(2LR)39-40; 42A (Lewis 1962). A l l experimental crosses were performed i n s h e l l v i a l s containing standard Drosophila medium. Segregation r a t i o s from males were always determined from the mean of a number of i n d i v i d u a l males each mated to two females. Segrega-t i o n r a t i o s from females were always determined from the mean of a number of i n d i v i d u a l females each mated to two males. Usually parental f l i e s were allow-ed to mate and lay eggs for three to four days and then they were discarded. A cobalt-60 source was used for r a d i a t i o n treatment. Mature males were treated with 2000 rads and then mated with females immediately, i n order that mature, i r r a d i a t e d sperm were sampled. A l l k values are calculated as the proportion of the t o t a l progeny bear-ing the chromosome that i s written f i r s t or on top. For example, for the a b cross — mated to — (a/b mated to b/b) k refe r s to the number of progeny bear-b b ing a divided by the t o t a l number of progeny. For a d e s c r i p t i o n of any v i s i b l e genetic markers used i n t h i s study, see Lindsley and G r e l l (1968). 55 RESULTS AND DISCUSSION In order to test i f there were any components of SD i n the centromeric heterochromatin of 2L, I wanted to induce a number of deletions i n that region of a SD chromosome. Since It ( l i g h t eyes) has an e a s i l y detectable phenotype and i s located i n the centromeric heterochromatin of 2L ( H i l l i k e r and Holm 1975) I decided to use ^ - r a y s to induce mutations of JLt on SD-5. This was done by i r r a d i a t i n g SD-5/SMI males, mating them to homozygous b pr l t pk cn females and then scoring the non-curly winged progeny f or It eyes (the SMI chromosome c a r r i e s the dominant gene Cy_, curly wings). The SD-5 chromosomes with newly induced lt ^ mutations were then balanced over SMI. Immediately a f t e r being balanced the \"It\" SD-5 chromosomes were checked for l e t h a l i t y with Df(2L)PR31, a def i c i e n c y of proximal 2L that includes the locus for lit . Since i t i s known that d e f i c i e n c i e s of rt are recessive l e t h a l s ( H i l l i k e r and Holm 1975), those \" l t \" SD-5 chromosomes that were l e t h a l with Df(2L)PR31 were r e -tained for further examination, since i t was l i k e l y that some of them would have heterochromatic d e f i c i e n c i e s . Nine l t SD-5 chromosomes were recovered that were l e t h a l with Df(2L)PR31. A l l nine chromosomes were tested f o r t h e i r a b i l i t y to d i s t o r t cn bw. In addition, they were complemented with a series of deletions of the heterochro-matin of 2L ( H i l l i k e r and Holm 1975) as well as with some recessive l e t h a l point mutations that comprise Group VII ( H i l l i k e r 1976). The r e s u l t s of the complementation and d i s t o r t i o n tests are shown i n Figure 10. Four of the chromosomes have reduced k values, while the remaining f i v e have k values that are unaffected by the mutations around lt^. From the complementation patterns i t i s apparent that a component of SD i s located between EMS 56-4 and the Group VI l e t h a l s of H i l l i k e r (1976). The loss of t h i s component r e s u l t s i n reducing k from close to 1.0 to approximately 0.6. L i k e l y the reason that FIGURE 10 The complementation relationships between the putative SD-5 deficiencies and Hill iker's (1976) groups of lethal mutations in the proximal heterochromatin of 2L. The centromere is to the right of Group VI, as drawn in the diagram. Df(2L)(SD-5)-27 was unfortunately lost before its ability to survive with EMS 56-24 was confirmed and a female.control to determine k^ was made. The only other chromosomes for which female controls were made were Df(2L)(SD-5)-2 and DF(2L)(SD-5)-61. The mean k shown for these is E(kc) and the error term is the 95 per cent confidence limits of E(k ) . 57 Gp IX GpVIII GPVII GpVI EMS 56-4 EMS 56-24 SD-5 del. no. mean k ^ 2 .595 ^ 0 2 7 1 _| 61 -607 i ° 2 6 8 - -\\ 2 7 .53 8 .883 45 .997 47 .995 44 .988 1 0 .984 40 L O Df(2L)(SD-5)-8 has an intermediate k of 0.88 i s that the a b i l i t y of t h i s SD component to function has only been s l i g h t l y affected by the proximity of the del e t i o n to i t , but the d e l e t i o n has not removed the SD s i t e . However, i t i s also possible that more than one jSD_ s i t e resides between Group VIII and Group VI and that Df(2L).(SD-5)-8 has deleted fewer of these than the other three SD-5 d e f i c i e n c i e s with reduced k's. I t i s also i n t e r e s t i n g to note that Df(2L)(SD-5)-8 has permitted me to order the two Group VII l o c i of H i l l i k e r (1976). It i s apparent that the locus s p e c i f i e d by EMS 56-4 must be d i s t a l to the locus s p e c i f i e d by EMS 56-24. Ganetzky (1977) also reported d i s c l o s i n g an important component of SID located i n proximal 2L. He did not, however, prove that the component, which he designated as E(SD), was located i n heterochromatin. The data presented here confirm the existence of E(SD) and also shows that i t i s located i n the proximal heterochromatin of 2L. As I did not know of Ganetzky's (1977) work when I had uncovered the component of jSD near rt, I wanted to confirm my r e s u l t s i n some other manner. Accordingly, I undertook a recombinational analysis of SD-5. The marker chromosome used was b pr I t pk cn. This chromosome was chosen both because of the s u i t a b l e l o c a t i o n of i t s markers and also because i t i s very s e n s i t i v e to SD-5 (from a cross of SD-5/b pr I t pk cn males mated to homozygous b pr I t pk cn females, out of 16,008 t o t a l progeny, only s i x were b pr I t pk cn). From 156 single SD-5/b pr It pk cn females each mated with two homozygous b ;>pr I t pk cn males the following progeny were recovered: 7,058 wild type, 6,711 b pr I t pk cn, 554 b_, 467 It pk cn or pr It pk cn (these could not be d i s t i n -guished without further t e s t s ) , 4 pk cn, 46 b pr, 2 cn, 2 b pr I t pk, 5 b pr I t , 18 p_r, 2 b pk cn, 1 pr I t . It i s i n t e r e s t i n g to note the high frequency of apparent jvr double crossovers. These w i l l be discussed l a t e r . F i f t e e n of the bj-bearing recombinants were tested f o r t h e i r a b i l i t y to d i s t o r t both cn bw and b pr It pk cn. The r e s u l t s are shown i n Table 1. Eleven of the 15 had high k's over both cn bw and b pr It pk cn. However, four had reduced k's. When heterozygous with cn bw these four had k's of about 0.55 and when heterozygous with b pr I t pk cn they had k's of about 0.8. This suggests that some component of SD i s located between b_ and pr. Table II gives the r e s u l t s of tes t s to determine the a b i l i t i e s of the b pr-bearing recombinants to d i s t o r t cn bw and b pr It pk cn. Fourteen chromo-somes were tested and 13 had k's of about 0.55 over cn bw and 0.58 - 0.84 over b pr It pk cn. One chromosome had a high k over both cn bw and b pr It pk cn. For reasons to be presented l a t e r , i t i s assumed that t h i s b pr chromosome was not the r e s u l t of a single exchange between p_r and rt. I f one disregards t h i s anomaly, then a l l of the b pr recombinants had reduced k's. This r e s u l t i s i n agreement with the proposition that there i s a component of SD located between b_ and _pjr. Before examining more c a r e f u l l y the properties of t h i s component, I s h a l l present the d i s t o r t i n g a b i l i t i e s of some of the other recombinants. Table I I I shows the a b i l i t i e s of some of these recombinants to d i s t o r t cn bw and b pr I t pk cn. The k values of R(SD-5) b pr I t - 1, R(SD-5) b pr It pk - 1, and R(SD-5) b pr It pk - 2 are quite v a r i a b l e but they tend to be near 0.5 over both cn bw and b pr I t pk cn. These chromosomes would be expected to have k values near 0.5, since they should not have e i t h e r the S I D component between b_ and £r_ or the component near I t . The pr It double recombinant also has a value near 0.5. The two pr SD-5 chromosomes that were tested both had high k values. I t was rather p e c u l i a r that these pr SD chromosomes should have high k's. I t was also rather p e c u l i a r that such a high frequency of _p_r \"double cross-overs\" were obtained fromSD-5/b pr It pk cn females. Because I could not TABLE I The d i s t o r t i n g a b i l i t i e s of the b SD-5 recombinants. 60 Number of Total 95% Recombinant Cross* males tested progeny Mean k confidence l i m i t s of k b-20 1 5 348 .99 2 12 729 .98 b-8 1 4 344 .96 _ 2 11 664 1.00 b-14 1 5 494 .99 _ 2 10 426 1.00 b-7 1 5 350 1.00 2 12 765 1.00 b-13 1 5 481 1.00 _ 2 12 1077 1.00 b-5 1 4 325 , .97 _ 2 12 835 1.00 b-10 1 5 424 .98 — 2 12 778 .98 b-2 1 5 443 1.00 — 2 11 890 1.00 b-1 1 5 484 1.00 2 12 1123 1.00 b-3 1 5 487 1.00 _ 2 11 889 1.00 b-6 1 5 546 1.00 _ 2 12 1018 1.00 b-12 1 11 1153 .56 .53 - .59 2 21 1553 .79 ** .72 - .85 ** b-4 1 11 1119 .54 .50 - .58 2 22 1490 .82 ** .75 - .88 ** b-11 1 10 973 .54 .50 - .58 2 20 1618 .80 ** .74 - .86 ** b-9 1 6 538 .55 .50 - .60 2 19 1288 .79 .** .72 - .85 ** TABLE I (Continued) * 1- R(SD-5)b-x male mated to cn bw females, cn bw cn bw 2. R(SD-5)b-x male mated to b pr It pk cn females, b pr It pk cn b pr 1t pk cn ** Calculated from arc sinv/k\". Note: None of these k values have been corrected for relative viability of SD and SD_+ progeny. TABLE II 62 The d i s to r t ing a b i l i t i e s of the b pr SD-5 recombinants. Number of Total 95% Recombinant Cross* males tested progeny Mean k** confidence l i m i t s of k' b pr -8 1 5 330 1.00 -2 10 412 .97 -b pr-7 1 11 866 .51 .47 - .55 2 24 2136 .70 .66 - .74 b pr -1 1 11 958 .54 .50 - .58 2 16 1185 .66 .59 - .73 b pr -9 1 10 687 .55 .48 - .62 2 14 1147 .70 .64 - .76 b pr -4 1 11 1003 .59 .55 - .63 2 19 1833 .81 .75 - .86 b pr-12 1 10 971 .57 .53 - .61 2 23 2299 .65 .61 - .67 b pr-10 1 10 739 .54 .49 - .59 • 2 21 1701 . .78 .71 - .84 b pr-15 1 11 956 .57 .53 - .61 2 23 2184 .58 .53 - .63 b pr -3 1 10 856 .54 .50 - .58 2 19 1256 .79 .70 - .87 b pr-14 1 11 1215 .55 .52 - .58 2 22 2221 .71 .65 - .76 b pr -5 1 9 836 , .60 .53 - .67 2 18 1479 .75 .70 - .80 b pr -2 1 11 1062 .54 .51 - .57 2 23 2047 .71 .64 - .78 b pr-16 1 11 1023 .58 .55 - .61 2 24 2321 .71 .68 - .74 b pr -6 1 10 742 .56 .50 - .62 2 16 1371 .84 .76 - .90 63 TABLE II (Continued) * 1. R(SD-5)b pr-x male mated to cn bw females, cn bw cn bw 2. R(SD-5)b pr-x males mated to b pr l t pk cn females, b pr It pk cn b pr It pk cn ** Calculated from arc sin 7k . Note: None of these k values have been corrected for relative viability of SD and SD_+ progeny. TABLE III The d i s to r t ing a b i l i t i e s of miscellaneous SD-5 recombinants. 64 Number of Total Unweighted 95% Recombinant Cross* males tested progeny Mean k confidence l i m i t s of F b pr l t - 1 1 11 1046 .48 .46 - .51 2 17 1162 .38 .33 - .42 b pr I t pk-1 1 11 1121 .56 .52 - .60 2 11 941 .48 .44 - .52 b pr I t pk-2 1 11 1107 .50 .47 - .53 2 13 1011 .39 .31 - .48 pr l t - 1 1 11 995 .49 .45 - .52 2 18 1127 .55 .52 - .57 pr -3 1 5 445 .99 _ 2 10 578 1.00 -pr-6 1 4 369 1.00 _ 2 11 583 1.00 -* 1. R(SD-5)-x male mated to cn bw females. cn bw cn bw 2. R(SD-5)-x male mated to b pr I t pk cn females. b pr I t pk cn b pr I t pk cn Note: None of these k values have been corrected for r e l a t i v e v i a b i l i t y of SD and SD+ progeny. phenotypically d i s t i n g u i s h pr l t pk cn and l t pk cn progeny, from the above cross, I cannot c a l c u l a t e exactly what the observed frequency of recombination was between b_ and £r and p_r' and lt^. However, since there were 554 b_'s, 46 b pr' s, 467 pr I t pk cn' s or l t pk cn's, and 17 p_r' s, I can approximate these frequencies of exchange as: i, 554 + 18 + L 1554 + 46 .. n , b - pr = x 100 = 6.750% 14,870 and 14,870 i - 46 + 18 + 1 \\ 554 + 46^ 1 A _ _ pr - l t = x 100 = 0.67% The number of observed double exchanges between b_ and j>£ and j>r and _lt should have been approximately equal to the product of 0.0675, 0.0067, and 14,870, which i s 6.74. Thus I should have recovered about four £r's, when i n fact 18 were recovered. Moreover, although the numbers were small, four females pro-duced one 2J£_ each, one female produced two '_p_r' s, and four females produced three pr' s each, and these _p_r' s appeared to be recovered as c l u s t e r s . Because of the high frequency of observed n_r's and the i n d i c a t i o n of c l u s t e r i n g , i t i s quite possible that these p_r' s were not the r e s u l t of a double exchange, i n -stead they might have been the r e s u l t of mutation. This would also explain why the two p_r' s tested f or t h e i r a b i l i t y to d i s t o r t cn bw had high k's. This hypothesis also provides a reasonable explanation for the observation that R(SD-5) b pr -8/cn bw males had high k values. Perhaps R(SD-5) b pr - 8 was the r e s u l t of a single exchange between b_ and _p_r, along with a mutation at pr. If t h i s were the case, then i t would be quite l i k e l y that R(SD)-5 b pr - 8 would s t i l l be capable of strongly d i s t o r t i n g cn bw. In order to further characterize the component of SD between b_ and £r_, I examined the segregational properties of the pr l t pk cn, I t pk cn, and pk cn recombinants when heterozygous with SMI. H a r t l (1975) had shown that In(2L + 2R)Cy ca r r i e d Rsp, and since SMI was derived from In(2L + 2R)Cy (Lindsley and G r e l l 1968), I f e l t that i t was reasonable to assume that SMI might also have Rsp. The r e s u l t s of the experiment are shown i n Table IV. A l l of the pr I t pk cn recombinants had k values near 0.5. However, both the It pk cn recombinants had k's of about 0.3. This suggests that the sperm bearing R(SD-5) I t pk cn were dysfunctioning. I t i s most reasonable to assume that t h i s i s because the R(SD-5) I t pk cn chromosomes have the component of SD between b_ and j>r, but they also must have Rsp\"1\", i . e . they have the genotype Sd Rsp + and SMI i s Sd + Rsp, which together i s a \"suicide combination\" (Hartl 1974). This indicates that the component between b_ and pr can operate i n c i s and trans to cause d i s t o r t i o n , providing one chromosome has Rsp and the other has Rsp +. I have avoided c a l l i n g the component between b'and pr, Sd, and the com-ponent near It E(SD) because the l o c i I have examined do not appear to behave i n exactly the same manner as the l o c i described by Ganetzky (1977). The differen c e resides i n the properties of E(SD). When Ganetzky (1977) deletes Sd, which i s located between b_ and p r , the deleted SD chromosome i s no longer capable of d i s t o r t i n g cn bw. However, when I replaced the Sd s i t e with a wild type region from the b pr - I t pk cn chromosome, the r e s u l t i n g R(SD-5) b pr chromosomes s t i l l appeared to have a s l i g h t capacity for d i s t o r t i n g cn bw (see Table I I ) . In order to tes t more c a r e f u l l y for any r e s i d u a l d i s t o r t i o n i n the b pr recombinants, I performed the crosses shown i n Table V. Since i t would have required too much work to test a l l of the recombinants, I selected only one recombinant, R(SD-5) b pr-5, for extensive examination. When k i s close to 0.5, i t i s important to correct for r e l a t i v e v i a b i l i t y of the two progeny classes, i f one wishes to demonstrate a s i g n i f i c a n t amount of d i s t o r t i o n i n TABLE IV The d i s t o r t i n g a b i l i t i e s of R(SD-5)-x/SMl males mated to homozygous cn bw females. 67 95% confidence n f E ( k c ^ l i m i t s of k c pr l t pk cn-14 .50 12 .50 12 .50 .46 - .55 pr I t pk cn-15 .50 12 .56 12 .43 .38 - .49 pr l t pk cn-16 .50 12 .54 12 .46 .40 - .52 pr 1t pk cn-17 .48 12 .54 12 .44 .41 - .47 pr l t pk cn-18 .51 12 .55 12 .47 .41 - .53 pr l t pk cn-19 .47 11 .50 12 .47 .42 - .53 pr l t pk cn-20 .53 12 .54 10 .49 .43 - .55 pr l t pk cn-22 .48 11 .55 10 .48 .43 - .52 pr l t pk cn-23 .49 12 .57 10 .42 .36 - .49 pr l t pk cn-24 .51 11 .52 12 .49 .43 - .55 1t pk cn-1 .29 12 .53 11 .26 .18 - .35 I t pk cn-2 .26 12 .48 10 .28 .19 - .37 pk cn-1 .46 12 .49 12 .47 .42 - .52 TABLE V 68 The distorting abilities of R(SD-5) b pr-5 under various circumstances, 95% confidence Experiment km nm n^ ) limits of E(kc) b pr-5 cn bw .560 49 .508 48 .551 .518 - .584 cn bw cn bw b pr-5 b pr It pk cn .530 49 .473 48 .555 .526- .584 cn bw b pr It pk cn b pr-5 „ cn bw .637 48 .506 49 .630 .586 - .674 b pr It pk cn cn bw b pr-5 b pr It pk cn x .794 45 .608 45 .715 .666 - .764 b pr It pk cn b pr It pk cn the segregation r a t i o . As I described at the beginning of Chapter I, I f e e l that the best way of doing t h i s i s by executing a r e c i p r o c a l cross. However, I noted there that one drawback to t h i s technique i s that conceivably there could be v i a b i l i t y maternal e f f e c t s associated with one of the progeny classes. In order to check i f such a problem might be occurring, R(SD-5) b pr-5 was made heterozygous both with cn bw and- with b pr I t pk cn and then both classes of heterozygous males and females were mated with homozygous cn bw and with homozygous b pr I t pk cn males or females, as shown i n Table V. When R(SD-5) b pr-5 was heterozygous with cn bw, ECk^) was e s s e n t i a l l y the same, with both cn bw and b pr It pk cn homozygous parents (0.551 and 0.555 r e s p e c t i v e l y ) . However, when R(SD-5) b pr-5 was heterozygous with b pr It pk cn, ^(k^) was 0.630 with cn bw as the homozygous parent and 0.715 with b pr It pk cn as the homozygous parent. This large difference i n E(k c) must be due eit h e r to a v i a b i l i t y maternal e f f e c t , or to a female genotype e f f e c t on the a b i l i t y of d i f f e r e n t sperm classes to function during f e r t i l i z a t i o n . Without further experiments i t i s impossible to say what causes t h i s d i f f e r e n c e . This r e s u l t demonstrates that one must not forget the drawbacks of t h i s type of a v i a b i l -i t y c o n t r o l . In Chapter I, I described the r e s u l t s of a large scale experiment with R(SD-5) b pr-5/cn bw males mated to homozygous cn bw; K i pP bx sr e s females. The mean k value of the experiment was 0.506, from 825 males tested. The standard deviation of k was 0.064. In order to check for any v i a b i l i t y pro-blems the r e c i p r o c a l cross of R(SD-5) b pr-5/cn bw females mated to homozygous cn bw; K i pP bx sr e s males was performed. In t h i s experiment, from 31 females tested, k^ was 0.478 with a standard deviation of 0.047. I t would appear that one reason why k was below 0.55 was because of reduced v i a b i l i t y m of progeny with both the R(SD-5) b pr-5 chromosome and the K i pP bx sr e s chromosome. Aft e r correcting for v i a b i l i t y , E(k ) equaled 0.529 and the 95 per cent confidence i n t e r v a l of the mean was 0.509 to 0.546. Even a f t e r cor-rections f or v i a b i l i t y E(k^) was s l i g h t l y less than 0.55, although i t was s i g n i f i c a n t l y greater than 0.5. Perhaps ECk^) was s l i g h t l y less than 0.55 because of a maternal v i a b i l i t y e f f e c t . Although there does appear to be a problem with ei t h e r maternal e f f e c t s or female genotype e f f e c t s , the fa c t that a l l four experiments shown i n Table V have E(k^) greater than 0.5 should c e r t a i n l y be taken as reasonable evidence that segregation d i s t o r t i o n i s s t i l l occurring with R(SD-5) b pr-5. This chromosome has replaced Ganetzky's (1977) Sd_ s i t e , located between b_ and pr, with the wild type s i t e from b pr I t pk cn. However, when Ganetzky (1977) deleted t h i s s i t e on h i s SD-5 chromosome, there was no evidence for r e s i d u a l d i s t o r t i o n . The difference i s that I used recombination, while Ganetzky (1977) used deletions. One p o s s i b i l i t y which might account for t h i s d i f f e r e n c e i s that the Sd + s i t e on b' pr l t pk cn could be semiactive. In order to check i f t h i s was the case, I examined a wild-type recombin-ant recovered from a cross of R(SD-5') b pr-5/cn bw females mated to b pr l t pk cn males. This recombinant I w i l l c a l l R(SD-5) b + pr +-5. It w i l l l i k e l y be genotypically the same as R(SD-5) b pr-5, except i t w i l l have most of euchro-matic 2L from b pr l t pk cn replaced with euchromatic 2L from cn bw. As shown i n Table VI, R(SD-5) b + pr +-5 appears to d i s t o r t cn bw, giving E(k^) equal to 0.543. Since t h i s recombinant can also d i s t o r t cn bw, i t must be either because S d + on both cn bw and b pr It pk cn are semi-active, or because the S>d_ s i t e between b_ and p_r i s not the only Sd_ s i t e on the p a r t i c u l a r SD-5 chromo-some used i n t h i s study. If the Sd_ s i t e between b_ and _p_r i s not the only S<1 s i t e , then I f e e l that the component near It i s a l i k e l y candidate for the other Sd_ s i t e . Henceforth I s h a l l r e f e r to the s i t e between b_ and p_r_ as Sd^ and the s i t e near l t as Sd 0. Each of these s i t e s alone provides a c e r t a i n TABLE VI Tests to determine i f Sd + , on b pr It pk cn is semi-active. 71 95% confidence Experiment km nm k f n f E(kc) limits of E(kc) RR(SD-5)b+ pr + -5 cn bw x cn bw R(SD-5) b pr It b pr It pk cn x b pr It pk cn b pr It pk cn SMI x b pr It pk cn ,515 50 .472 40 .543 .520 - .567 ,337 43 .370 43 .464 .407 - .522 .424 45 .355 49 .572 .547 - .597 72 capacity to d i s t o r t , while both together enable maximal d i s t o r t i o n . I t i s now necessary to decide whether or not Sd*, on b pr It pk cn i s semi-active (Sd^ s). A good test would be to observe the segregation r a t i o of Sd^ S Rsp/Sd^ s Rsp + . This should have a k greater than 0.5. The chromosome R(SD-5) b pr l t - 1 (see Table III) i s Sd-^ Rsp . That i t c a r r i e s Rsp w i l l be discussed l a t e r . The chromosome b pr I t pk cn i s Sd^ s Rsp* . Accordingly, I determined E(k c) for R(SD-5) b pr l t - l / b pr I t pk cn. It was 0.464 and the 95 per cent confidence l i m i t s of t h i s mean were 0.407 - 0.522,(see Table VI). The test did not show any s i g n i f i c a n t d i s t o r t i o n and accordingly suggests that Sd*^ , on b pr I t pk cn i s not semi-active (Sd^ s). Furthermore, i t i s very u n l i k e l y that the crossover which produced R(SD-5) b pr l t - 1 would have occur-red between jLt and Sd^. I f t h i s were the case, then the genotype of R(SD-5) b pr l t - 1 could have been SD-^+ 5d^~ Rsp. Since i t did not show d i s -t o r t i o n when heterozygous with b pr I t pk cn, i t i s reasonable to a t t r i b u t e the r e s i d u a l d i s t o r t i o n of R(SD-5) b pr-5 to Sd^ and not to some other s i t e i n 2R of SD-5. As another approach to determine i f Sd*, on the b pr I t pk cn chromosome, is. semi-active, I examined the segregation r a t i o of b pr I t pk cn/SMl when mated to b pr I t pk cn homozygotes. The heterozygote could be symbolized as s + + + + s Sd^ Sd 2 Rsp /Sd^ Sd^ Rsp . I f Sd^ i s t r u l y semi-active, then one would expect that E ( k c ) would be le s s than 0.5. As shown i n Table VI, E ( k c ) was 0.572 and the 95 per cent confidence l i m i t s of the mean did not overlap 0.5. Since was not le s s than 0.5 t h i s evidence also suggests that Sd*, on b pr I t pk cn i s not active. It i s p e c u l i a r , however, that E ( k c ) was greater than 0.5. I f e e l that i t i s quite reasonable to a t t r i b u t e t h i s to a maternal v i a -b i l i t y e f f e c t . I t might be that progeny which are homozygous b pr It pk cn are less v i a b l e i f the female parent i s homozygous for b pr I t pk cn rather than i f the female parent i s heterozygous for b pr l t pk cn. This hypothesis i s also i n agreement with the observation that E(k ) for R(SD-5) b pr l t - 1 / c * b pr l t pk cn mated to b pr I t pk cn/b pr l t pk cn was s l i g h t l y less than 0.5 and also with the observation that Mk^) for R(SD-5) b pr-5/b pr l t pk cn mated to b pr l t pk cn/b pr l t pk cn was greater than E(k^) for R(SD-5) b pr-5/ b pr l t pk cn mated to cn bw/cn bw. The evidence presented here suggests that both Sd^ and Sd^ , are capable of inducing a c e r t a i n amount of d i s t o r t i o n , providing one chromosome has Rsp and the other has Rsp*. On the other hand, Ganetzky (1977) has presented evidence suggesting that only Sd^ can operate on i t s own to induce d i s t o r t i o n . His r e s u l t s lead him to c a l l Sd.^ a n enhancer of SD, E(SD). There are two obvious p o s s i b i l i t i e s that could explain t h i s discrepancy. F i r s t l y , i t i s possible that the SD-5 chromosome used by Ganetzky (1977) had a Sd^ , a l l e l e that could not d i s t o r t alone. Secondly, perhaps an SD chromosome that has had Sd^ deleted behaves d i f f e r e n t l y than an SD chromosome that has had Sd^ replaced with Sd* by recombination. This l a s t p o s s i b i l i t y would appear to be the case with respect to Sd^,. This was demonstrated by replacing Sd^ with Sd* from b pr l t pk cn. Females of the genotype R(SD-5) b pr lt/R(SD-5) pk cn were mated to b pr l t pk cn males and recombinants that were rt and pr It were retained. These were c a l l -ed RR(SD-5) l t and RR(SD-5) pr l t . Since R(SD-5) b pr I t i s Sd* Sd* Rsp and R(SD-5) pk cn i s Sd^ Sd^ Rsp, the J^t recombinant would be Sd^ Sd^ + Rsp and the pr l t recombinant would be Sd* Sd* Rsp. As shown i n Chapter I the d i s t r i b u t i o n of 1,145 RR(SD-5) pr l t / c n bw males mated to cn bw; K i pP bx sr e s females gave a mean k of 0.463 and a standard deviation of .076. A r e c i p r o c a l cross inv o l v i n g 33 females gave a mean k of 0.467 and a standard deviation of 0.063. These r e s u l t s give an E ( k c ) of .504 with 95 per cent confidence l i m i t s of .481 to .528. This chromosome does not d i s t o r t , as one would expect. RR(SD-5) I t / cn bw males, on the other hand, had a mean k of 0.713 and a standard deviation of 0.118 (the sample s i z e was equal to 1,142). A r e c i p r o c a l cross had a k^ of 0.47. Since only 17 females were examined, I w i l l not calculate k . The mean c k of 0.713 was far greater than one would expect from the r e s u l t s with dele-m tions of Sd„, which had k 's of about 0.6. Since the dele t i o n of Sd 0 had a — 2 c — 2 d i f f e r e n t k than the recombinant that replaced Sd_^ with Sd*, i t i s apparent that, with respect to SD, a de l e t i o n does not necessa r i l y behave the same as an a l l e l i c s u b s t i t u t i o n through recombination. The reason that i t i s of some importance to know whether or not Sd^ i s capable of operating without Sd^ i s that Holm (personal communication) has observed that compound 2R chromosomes constructed from SD-72/cn bw are capable of d i s t o r t i n g compound 2L chromosomes from'the It stw s t r a i n . SD-72 has a p e r i c e n t r i c inversion with break points at 39-40 and 42A (Lewis 1962) and be-cause of t h i s , a compound 2R chromosome constructed from SD-72 has a f a i r l y good chance of having Sd^,, but i t could not have Sd^, since Sd^ i s d i s t a l to the l e f t break point of the inversion. Since an SD-72 compound 2R can d i s t o r t a compound 2L, i t would be most reasonable to assume that i t i s Sd^ , that i s operating i n the compound 2R, e s p e c i a l l y i n view of the evidence that has been presented here. Some SD-72 compound 2R's can cause almost t o t a l e limination of c e r t a i n compound 2L's (Holm, personal communication). This observation demonstrates that some l o c i i n 2R or proximal 2L of SD-72 are capable of inducing a high l e v e l of d i s t o r t i o n under c e r t a i n conditions. This suggests that i t should be possible to exchange most of 2L on SD-72 without appreciably a f f e c t i n g the k of the recombinant. In order to test t h i s I obtained b pr recombinants from SD-72/b pr It pk cn females. Five of these recombinants were retained and tested f o r t h e i r a b i l i t y to d i s t o r t ' cn bw. The r e s u l t s are shown i n Table VII TABLE VII The distorting abilities of R(SD-72) b pr-x/cn bw males mated to cn bw/cn bw females. 75 95% confidence Recombinant km nm k f n f E(kc) limits of kc b pr-1 .493 48 .507 43 .489 .462 - .512 b pr-2 .513 49 .499 47 .514 .490 - .538 b pr-3 .503 49 .505 43 .498 .477 - .518 b pr-4 .497 50 .499 49 .498 .476 - .520 b pr-5 .504 50 .500 45 .504 .478 - .530 In a l l f i v e cases E(k c) was close to 0.5 and the 95 per cent confidence i n t e r -v a l of the mean included 0.5. There was absolutely no evidence for d i s t o r t i o n . This r e s u l t c l e a r l y presents a paradox. I f a compound 2R constructed from SD-72 can d i s t o r t a compound 2L, then why can't R(SD-72) b pr d i s t o r t cn bw? My r e s u l t s have shown that under c e r t a i n circumstances Sd^ , alone can cause a r e s i d u a l amount of d i s t o r t i o n i n the SD-5 chromosome studied here. In order to resolve the previously mentioned paradox one must determine what the circum-stances are that allow segregation d i s t o r t i o n to operate i n SD-72 compound 2R's. The r e s u l t s presented here suggest that Sd^ , i s a l i k e l y s i t e on which to i n i t i a t e t h i s study. Another s i t e of i n t e r e s t i s Rsp. As mentioned i n the introduction to t h i s chapter, Ganetzky (1977) has presented evidence which suggests that Rsp i s located near what H i l l i k e r and Holm (1975) have c a l l e d Group I I , i n the proximal heterochromatin of 2R. However, Ganetzky (1977) r e a l i z e s that such a placement i s inconsistent with the s e n s i t i v i t y of a large heterochromatic dele t i o n such as Df(2R) M-S2~*\"^ . Ganetzky (1977) concludes by i n f e r r i n g that i t i s most l i k e l y that Rsp i s located proximal to Group I and that h i s d e f i -ciency mapping of the s i t e was somewhat misleading because of possible multiple h i t events. I had planned to determine the s e n s i t i v i t i e s of the SD-5 recombinants that I recovered, but t h i s proved d i f f i c u l t to do because of a pe c u l i a r type of s e m i - s t e r i l i t y i n males of the genotype SD-5/SD-72. Many of the SD-5 r e -combinants also showed t h i s s e m i - s t e r i l i t y i n combination with SD-72 and con-sequently i n s u f f i c i e n t progeny were recovered to warrant reporting the r e s u l t s . Although I s h a l l not report the s e n s i t i v i t i e s of a l l of the SD-5 recombinants, I s h a l l discuss two p a r t i c u l a r recombinants. R(SD-5) pr lt-l/SD-72 males mated to b pr l t pk cn females gave a mean k of 1.0. Ten males were tested and 680 progeny recovered. On the other hand R(SD-5) b pr lt-l/SD-72 males mated to b pr l t pk cn females gave a mean k of 0.593 with a 95 per cent con-fidence i n t e r v a l from 0.529 to 0.657. Eleven males were tested and 842 pro-geny were recovered. Although, f or some reason, the l a t t e r k was s i g n i f i c a n t -l y greater than 0.5, i t i s apparent that R(SD-5) b pr l t - 1 was e s s e n t i a l l y i n s e n s i t i v e to SD-72, while R(SD-5) pr l t - 1 was very s e n s i t i v e to SD-72. Since Ganetzky (1977) has shown by recombination that Rsp l i e s between _p_r and cn, i t i s u n l i k e l y that the differ e n c e i n the s e n s i t i v i t i e s of these two r e -combinants could be a r e s u l t of an exchange d i s t a l to pr. I t i s most reason-able to a t t r i b u t e the s e n s i t i v i t y difference to differences i n the s i t e of exchange between lt_ and pk, i . e . t h i s data suggests that Rsp l i e s between l t and pk. Furthermore, since the frequency of exchange i n the centromeric heterochromatin of 2R i s very low ( H i l l i k e r 1975), the r e s u l t suggests that Rsp l i e s i n the euchromatin of 2R proximal to pk. I found t h i s r e s u l t extremely perplexing i n view of the observation that a compound 2R constructed from SD-72 can d i s t o r t c e r t a i n compound 2L's (Holm, personal communication). If £>D compound autosomes behave the same as standard SD autosomes, then C(2R)SD should have Rsp and C(2L) should have Rsp +. Since compound autosomes r a r e l y have duplications or d e f i c i e n c i e s of proximal euchromatic regions ( H i l l i k e r and Holm 1975) , i f Rsp and Rsp + are located i n the proximal euchromatin of 2R, i t i s very u n l i k e l y that a compound 2L could have Rsp +. On the other hand, i f Rsp + were located i n the centromeric hetero-chromatin of 2R, i t would be quite l i k e l y that a compound 2L could have Rsp +. In order to more c a r e f u l l y examine the l o c a t i o n of Rsp, I sought a chromosome that was free of inversions, but c a r r i e d Rsp. The chromosome b pr r l cn appeared to be i n s e n s i t i v e to SD-72, since Sd-72/b pr r l cn males mated to cn bw females gave a mean k of 0.5 with the 95 per cent confidence i n t e r v a l 78 of the mean being 0.46-0.54. Eleven males were tested and 719 progeny were recovered. If b pr r l cn was i n s e n s i t i v e because i t c a r r i e d Rsp, then i t should be able to induce s e l f - d i s t o r t i o n of R(SD-5) I t pk cn, which i s Sd^ Sd^ Rsp . In the cross R(SD-5) It pk cn-l/b pr r l cn mated to b pr r l cn, E(k^) was 0.266 and the 95 per cent confidence i n t e r v a l of t h i s mean extended from 0.222 to 0.309. The mean k of the male cross was 0.250 (25 males tested) and the mean k of the female cross was 0.480 (25 females tested). C l e a r l y b pr r l cn can induce s e l f - d i s t o r t i o n of R(SD-5) I t pk cn-1. Because of t h i s , i t i s quite l i k e l y that b pr r l cn c a r r i e d Rsp. I f b pr r l cn c a r r i e d Rsp, then i t was a su i t a b l e chromosome to use to map Rsp, since I have observed that b pr r l cn does not have any inversions. Accordingly, b pr r l cn was recombined with a s e n s i t i v e second chromosome from the wild-type Oregon-R stock. Six separate recombinants of every geno-type r e s u l t i n g from a si n g l e exchange were retained and tested for s e n s i t i v i t y to an SD chromosome. A l l recombinants that did not have cn were tested f or s e n s i t i v i t y to R(SD-5) pk cn. The r e s u l t s are shown i n Table VIII. A l l of the b_ and b pr recombinants were more or le s s s e n s i t i v e , since they a l l had mean k's s i g n i f i c a n t l y greater than 0.5. On the other hand, a l l of the b pr r l recombinants had mean k's whose 95 per cent confidence i n t e r v a l s included 0.5. The r e c i p r o c a l recombinants of those shown i n Table VIII were tested for t h e i r s e n s i t i v i t y to R(SD-5) b-8. The r e s u l t s are shown i n Table IX. The cn recombinants had high mean k's of about 0.97. A l l but one of the r l cn and pr r l cn recombinants had mean k's whose 95 per cent confidence i n t e r v a l s included 0.5. The one exception was r l cn-2 and i t had a mean k of 0.43. Although I do not know why i t had a mean k s i g n i f i c a n t l y less than 0.5, i t c e r t a i n l y did not show any evidence of s e n s i t i v i t y . These r e s u l t s suggest that Rsp i s t i g h t l y linked to r l on the b pr r l cn TABLE VIII The sensitivities of b-bearing recombinants, from Or - R/b pr rl cn females, when heterozygous with R(SD-5) pk cn. 79 Number of Total 95% confidence Recombinant males tested progeny Mean k limits of mean k b-1 10 1222 .71 .65 - .77 b-2 10 1272 .70 .64 - .76 b-3 10 1190 .76 .67 - .85 b-4 10 1127 .66 .56 - .76 b-5 10 1191 .71 .67 - .75 b-6 9 1072 .75 .66 - .84 b pr-1 10 848 .80 .72 _ .88 b pr-2 10 920 .61 .54 - .68 b pr-3 10 1081 .78 .69 - .87 b pr-4 10 1467 .59 .56 - .62 b pr-5 10 1416 .75 .68 - .82 b pr-6 9 1005 .71 .62 - .80 b pr r l -1 10 1026 .50 .47 _ .53 b pr r l -2 9 947 .54 .48 - .60 b pr r l -3 10 1347 .50 .45 - .55 b pr r l -4 10 1053 .52 .50 - .54 b pr r l -5 10 699 .47 .42 - .52 b pr r l -6 9 581 .51 .44 - .58 Note: None of these k values have been corrected for relative viability of SD and SD+ progeny. TABLE IX £ The sensitivities of on-bearing recombinants, from Or-R/b pr rl cn, when heterozygous with R(SD-5)b-8. Number of Total 95% confidence males tested progeny Mean k limits of mean k cn-1 cn-2 cn-3 cn-4 cn-5 cn-6 10 10 10 10 10 9 1426 1387 1313 1348 1404 1372 .95 .97 .95 .97 .97 .99 rl cn-1 rl cn-2 rl cn-3 rl cn-4 cn-5 cn-6 rl rl 9 9 10 10 10 10 1134 1215 1612 1464 1625 1625 .48 .43 .49 .47 .50 .47 .44 .39 .47 .44 .48 .44 .52 .47 .51 .50 .52 .50 pr rl cn-1 pr rl cn-2 pr rl cn-3 pr rl cn-4 pr rl cn-5 pr rl cn-6 10 10 10 10 10 10 1297 1763 1593 1600 1639 1334 .51 .49 .48 .50 .49 .49 .47 .45 .46 .46 .46 .45 .55 .53 .50 .54 .52 .54 Note: None of these k values have been corrected for relative viability of SD and SD* progeny. 81 chromosome. This i n turn implies that Rsp could be located i n heterochromatin, since exchange between heterochromatic l o c i i s very rare ( H i l l i k e r 1975). However, t h i s proposal contradicts the observation that R(SD-5) pr l t - 1 was s e n s i t i v e , while R(SD-5) b pr l t - 1 was i n s e n s i t i v e . In order to avoid the contradiction one must eit h e r assume that Rsp i s i n fact located i n euchroma-t i n and i t was only chance that prevented me from separating Rsp from r l i n any of the 12 recombinants between r l and cn, or one must assume that R(SD-5) b pr l t - 1 was the r e s u l t of a rare heterochromatic exchange. In view of Holm's (personal communication) observation that a compound 2R constructed from SD-72 can d i s t o r t a compound 2L, I f e e l that i t would be best to assume that R(SD—5) b pr l t - 1 was the r e s u l t of a rare heterochromatic exchange. In an attempt to obtain firmer evidence on the l o c a t i o n of Rsp, I r e -covered four cf-ray induced l e t h a l mutations of Group II on cn bw (these mutations were induced i n sperm). I f any one of these Group II l e t h a l s was a + + d e l e t i o n that also deleted Rsp , and i f Rsp was located proximal to Group I, then Group I would ne c e s s a r i l y also be deleted. A d e l e t i o n of Rsp + on cn bw should cause cn bw to be i n s e n s i t i v e (Ganetzky 1977). The s e n s i t i v i t y of three of the four Group II l e t h a l s to SD-5 i s shown i n Table X. It i s appar-ent that the three chromosomes retained t h e i r s e n s i t i v i t y to SD-5. The fourth Group II l e t h a l was s t e r i l e with SD-5 i n both males and females. This problem was avoided by t e s t i n g i t s s e n s i t i v i t y to SD-72. Ten SD-72/(rl)-l-cn bw - 4 males were mated to cn bw females and 1,139 progeny were recovered. The mean k was 1.0. Thus a l l four l e t h a l s of Group II were s e n s i t i v e and they could not be used to provide any p o s i t i v e information on the l o c a t i o n of Rsp. Nevertheless, complementing these l e t h a l s with H i l l i k e r ' s (1976) EMS induced l e t h a l s i n the heterochromatin of 2R could provide information on where Rsp i s not located. The r e s u l t i n g complementation patterns are shown i n Figure 11. The chromosome ( r l ) - l - c n bw-1 appears to be a deletion extending from Group II TABLE X 82 The sensitivities of cn bw chromosomes with c^-ray induced lethals on them. A. SD-5 (rl) - l -cn bw-x x cn bw ¥ * Lethal Number Number of males tested Total progeny Mean k 1 2 3 44 47 42 2746 2934 2773 .993 .980 .982 B ' SD-5 (r l ) - l -cn bw-x * x cnbw ^ cn bw Lethal Number Number of females tested Total progeny Mean k 95% confidence 1imits of mean k 10 10 10 1006 1044 951 .52 .52 .49 .50 - .54 .49 - .55 .46 - .52 FIGURE 11 The complementation relationships between the cn bw chromosomes bearing putative deficiencies and Hill iker's (1976) groups of lethal mutations in the proximal heterochromatin of 2R. This is to the left of Group I. G p l Gpll Gp III GpIV GpV lethal t *\" V n a EMS 45-10 EMS 31 \\ » j j — — 1 : ! : 1 I I 1 — r — i I i I 2 j j , j r j j j 3 j i i — A 1 i i — I i I 4 to the Group I s i t e s p e c i f i e d by EMS 31. I t complements with a l l four of the l e t h a l s (EMS 45-10, EMS 45-84, EMS 45-91, and EMS 45-87) i n the other s i t e of Group I. Thus Rsp would not appear to be located between Group II and the s i t e s p e c i f i e d by EMS 31. This r e s u l t also demonstrates that the EMS 31 s i t e i s d i s t a l to the other Group I s i t e s p e c i f i e d by EMS 45-10, EMS 45-84, EMS 45-91, and EMS 45-87. The other % -ray induced l e t h a l s on cn bw provide l i t t l e information of any sort. The chromosome ( r l ) - l - c n bw-4 was peculiar i n that i t was the r e s u l t of a multiple h i t event, even though only 2000 rads were used to generate these l e t h a l s . CONCLUSION This £!D mapping study was p a r a l l e l to Ganetzky's (1977) work i n many ways. Both studies have demonstrated the existence of an important component on SD-5 that i s located j u s t d i s t a l to _p_r. They also have demonstrated the e x i s -tence of another component, on SD-5, that i s located near lt^. Both studies have contributed a meagre amount of evidence i n support of the contention that Rsp i s located i n the proximal heterochromatin of 2R. This agreement serves to increase the c r e d i b i l i t y of the points agreed upon. The only major d i f f e r -ence between our r e s u l t s i s that my findings suggest that the component of SD_ located near Lt i s capable of inducing d i s t o r t i o n i n the absence of the SD component located j u s t d i s t a l to _p_r, providing one chromosome has Rsp and the other chromosome has Rsp +. Accordingly, the s i t e that i s located j u s t d i s t a l to pr_, I have c a l l e d Sd.. and the s i t e that i s located near Lt, I have c a l l e d 87 LITERATURE CITED Crow, J .F . , C. Thomas, and L. Sandler, 1962 Evidence that the segregation-distortion phenomenon in Drosophila involves chromosome breakage. Proc. Nat. Acad. Sci. U.S.A. 48: 1307 - 1314, Das, C C , B.P. Kaufmann, and H. Gay, 1964 Histone protein transition in Drosophila malanogaster. I. Changes during spermiogenesis. Exp. Cell Res. 35: 507 - 514. Ganetzky, B., 1977 On the components of segregation distortion in Droso- phila melanogaster. Genetics 86_: 321 - 355. Gould-Somero, M. and L. Holland, 1974 The timing of RNA synthesis for spermiogenesis in organ cultures of Drosophila melanogaster testes. Wilhelm Roux1 Archiv. 174: 133 - 148. Hartl, D.L., 1974 Genetic dissection of segregation distortion. I. Suicide combinations of SD genes. Genetics 76: 477 - 486. Hartl, D.L., 1975 Genetic dissection of segregation distortion. II. Mechan-ism of suppression of distortion by certain inversions. Genetics 8TJ: 539 - 547. Hartl, D.L. and Y. Hiraizumi, 1976 Segregation distortion. In: The gene- tics and biology of Drosophila. Edited by M. Ashburner and E. Novitski, Academic Press, Londbrr: Hartl, D.L., Y. Hiraizumi, and J.F. Crow, 1967 Evidence for sperm dysfunction as the mechanism of segregation distortion in Drosophila melanogaster. Proc. Nat. Acad. Sci. U.S.A. 58: 2240 - 2245. Hill iker, A . J . , 1976 Genetic analysis of the centromeric heterochromatin of chromosome 2^ of Drosophila melanogaster: Deficiency mapping of EMS-induced lethal complementation groups. Genetics 83_: 765 - 782. Hill iker, A.J. 1975 Genetic analysis of the proximal heterochromatin of chromosome 2^ of Drosophila melanogaster. Ph.D. Thesis, University of British Columbia. Hill iker, A.J. and D.G. Holm, 1975 Genetic analysis of the proximal region of chromosome 2 of Drosophila melanogaster. I. Detachment products of compound autosomes. Genetics 81: 705 - 721. Hiraizumi, Y., 1961 Lethality and low viability induced by the segregation distorter locus (symbol SDj in Drosophila melanogaster. Ann. Rep. Nat. Inst. Genet. Jap. 12: 1-2. Hiraizumi, Y. and K. Nakazima, 1967 Deviant sex ratio associated with segregation distortion in Drosophila melanogaster. Genetics 55: 681-697. Johnson, N.I. and S. Kotz, 1969 Discrete distributions. Houghton Mifflin Co., Boston. 88 Kettanah, N.P. and D.L. Hartl, 1976 Histone transition during spermiogenesis is absent in segregation distorter males of Drosophila melanogaster. Science 193: 1020 - 1021. Lewis, E.B., 1962 Salivary gland chromosome analysis of segregation distorter lines. Dros. Inf. Ser. 36_: 87. Lindsley, D.L. and E.H. Grell, 1968 Genetic variations of Drosophila melano- gaster. Carnegie Institute of Washington Publ. No. 627^ Mange, E.J., 1968 Temperature sensitivity of segregation-distortion in Drosophila melanogaster. Genetics 58: 399 - 413. Miklos, G.L.G. and S. Smith-White, 1971 An analysis of the instability of segregation-distorter in Drosophila melanogaster. Genetics 67_: 305-317. Mood, A.M., F.A. Graybill, and D.C. Boes, 1974 Introduction to the theory of statistics. McGraw-Hill, New York. Novitski, E. and I. Sandler, 1957 Are all products of spermatogenesis regularily functional? Proc. Nat. Acad. Sci. U.S.A. 43: 318 - 324. Peacock, W.J. and J. Erickson, 1965 Segregation-distortion and regularily nonfunctional products of spermatogenesis in Drosophila melanogaster. Genetics 51: 313 - 328. Peacock, W.J., K. Tokuyasu, and R.W. Hardy, 1972 Spermiogenesis and meiotic drive in Drosophila. In: Edinburgh symposium on the genetics of the spermatozoon. Edited by R.A. Beatty and S. Gluecksohn-Waelsch, Bogtryk-keriet Forum, Copenhagen. Rendel, J.M., 1967 Canalisation and gene control. Logos Press, London. Sandler, L. and Y. Hiraizumi, 1960a Meiotic drive in natural populations of Drosophila melanogaster. IV. Instability at the segregation-distorter locus. Genetics 45: 1269 - 1287. Sandler, L. and Y. Hiraizumi, 1960b Meiotic drive in natural populations of Drosophila melanogaster. V. On the nature of the SD region. Genetics 45: 1671 - 1689. Sandler, L., Y. Hiraizumi, and I. Sandler, 1959 Meiotic drive in natural populations of Drosophila melanogaster. I. The cytogenetic basis of segregation-distortion. Genetics 44_: 233 - 250. Sokal, R.R. and F.J. Rohlf, 1969 Biometry. W.H. Freeman and Co., San Francisco. Tokuyasu, K.T., 1974 Dynamics of spermiogenesis in Drosophila melanogaster. IV. Nuclear transformation. J. Ultrastruct. Res. 48: 284 - 303. 89 Tokuyasu, K.T., W.J. Peacock, and R.W. Hardy, 1977 Dynamics of spermio-genesis in Drosophila melanogaster. VII. Effects of segregation distorter (SD) chro~mosome. J. Ultrastruct. Res. 58: 96 - 107. Zimmering, S. and G.L. Fowler, 1968 Progeny: sperm ratios and nonfunctional sperm in Drosophila melanogaster. Genet. Res. (Camb.) 12: 359 - 363. "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0094064"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Zoology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Location and properties of some of the major loci affecting the segregation distortion phenomenon in Drosophila Melanogaster"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/20482"@en .