mmm'rm. SURVWAL mom; spam mmm‘ms FRQM HETEROZYGOUS MALE ' mommm mammomsm $919331: PM file Mme- Qf Pb. E33. :‘A!€H§GAE€ STATE umfimw Mécifiwfi E. MysmwsE-{i E965 THE“ LIBRARY . Micki ,, .11 State .3 University (.2 This is to certify that the thesis entitled DIFFERENTIAL SURVIVAL AMONG SPERM POPULATIONS FROM HETEROZYGOUS MALE Drosophila me lanogaster . presented by Michael E . Myszewski has been accepted towards fulfillment of the requirements for Ph.D. degree in ZOOLOGY A. F. Yandew or professor Date July 21, 1965 0-169 ABSTRACT DIFFERENTIAL SURVIVAL AMONG SPERM POPULATIONS FROM HETEROZYGOUS MALE DROSOPHILA MELANOGASTER by Michael E. Myszewski Aberrant progeny ratios may arise from errors occurring at meiosis, selection during the gametic stages, or non— random survival after fertilization. While all sperm have until recently been considered to be equally capable of achieving fertilization, increasing evidence indicates that this capability does not hold for all cases which have been studied. The present experiments were performed to deter- mine if gametic selection could be detected as a differen- tial recovery of progeny produced by male Drosophila heterozygous for second chromosome markers. Heterozygous males carrying combinations of the marker chromosomes, Curly, Plum, and lethal-9, were mated to OR (wild-type) or §§§E.females. Progeny produced from a twelve-day laying period were scored with regard to their sex and the autosomal markers which they carried. Similar matings were performed to provide a measure of the egg hatch and larval mortality so as to rule out postfertili- zation phenomena. Michael E. Myszewski Significant deviations were found in the relative re- covery of the autosomes under consideration. Some matings gave evidence of differential gametic viability although zygotic mortality could not be ruled out as a cause in other crosses. No particular homologue was found to be favored throughout all of the matings. The sperm stored in the female reproductive tract seemed to be acted upon in a differential manner dependent on their chromosomal makeup and on the genotype of the female. It is concluded that the female reproductive tract can cause some sperm, less capable of maintaining themselves in the storage organs, to be expelled. In a related series of experiments, Cygle—9 males were irradiated and crossed to §§§g_females to study the rela- tive association of the marker chromosomes with newly induced sex-linked recessive lethals. A tendency was noted for newly induced sex-linked lethals to be asso- ciated with the lg:2_chromosome rather than the Curly chromosome. This tendency appears to operate independ- ently of the sex ratio. Whatever mechanism produces this effect acts on a single homologue (in this case Curly) rather than a combination of chromosomes. DIFFERENTIAL SURVIVAL AMONG SPERM POPULATIONS FROM HETEROZYGOUS MALE DROSOPHILA MELANOGASTER Bytr — \ Michael E. Myszewski A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Zoology 1965 ACKNOWLEDGEMENTS I would like to express my sincere thanks to Dr. Armon F. Yanders, both for the privilege of working with him and for the counsel, interest and encouragement he has freely given to me both in and out of the laboratory during the course of my graduate work. I would like to acknowledge and thank the members of my guidance committee: Drs. J. R. Shaver, G. B. Wilson, P. J. Clark and J. V. Higgins. Their assistance to me in the classroom and in conversation has proven to be the better part of my graduate education. The sad loss of Dr. Clark from this committee was sorely felt. To Mrs. B. R. Henderson go my express thanks for her warm interest in my work and family; for her generous direction and advice that is not to be found in catalogs or memos from the graduate office. This study was conducted during the tenure of fellow— ships awarded by the National Science Foundation and sup- ported in part by a grant to Dr. A. F. Yanders from the Atomic Energy Commission (Contract AT (ll-l) — 1033). ii To my wife, H. Barbara TABLE OF CONTENTS Acknowledgements . . . . . . . . . . List of Tables . . . . . . . . . . . . SECTION INTRODUCTION . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . METHODS AND MATERIALS . . . . . . Stocks . . . . . . . . . . . . Food and Temperature . . . . . Ratio Experiments . . . . . . Sex-linked Lethal Experiments Tests for Mortality . . . . . RESULTS . . . . . . . . . . . . . Ratio Experiments . . . . . . Tests for Mortality . . . . . Sex-linked Lethal Experiments DISCUSSION . . . . . . . . . . . Ratio Experiments . . . . . . Sex-linked Lethal Experiments SUMMARY . . . . . . . . . . . . . BIBLIOGRAPHY . . . . . . . . . . iv Page ii 26 26 27 28 32 33 36 36 47 53 6O 6O 64 66 68 LIST OF TABLES Page Summed data from initial ratio experiments .... 37 Results of later experiments testing segre— gation of autosomal markers including progeny totals of all crosses, breakdown of data with regards to sex and autosomal markers, le-9 ratio, sex ratio and X2 values for deviation from 1:1 ratio .. ........ ................ ...... 39 X2 values for 2 X 2 contingency chi-square tests comparing all the crosses listed in Table 2 for intercross variation. X2 values in boxes 1 to 36 represent sex ratios; X2 values in boxes 37 to 72 represent 1e-9 ratios. Values listed in column one ending in (b) represent calculations based only on numbers of females ......................... 43 Matings of later crosses grouped with regards to male homologues (Section I) and maternal genotype (Section II) ..... ....... 45 Progeny counts from temperature mortality experiments ........... ..... ................... 49 Results of egg and larval mortality experiments OCOO...OOOOOOOOOOOOOOOOOOOO00...... 49 "Correction" of two progeny ratios through the addition of "presumptive progeny" to the smallest phenotypic class ................. 52 Comparison of the experimental groups of the sex-linked lethal study (Cy/le-9dd X M-S 99) 00.00.00.00000.00000000000000000ooooooo 54 Table 10. Page X2 tests and values for comparisons made within and between F1 and F2 generations of the sex-linked lethal study .... ............ 55 Progeny counts for the F1 and F2 genera— tions of the sex-linked lethal experiments .... 56 vi INT RODUCTI ON Deviations from expected progeny ratios may develop from aberrations arising during the meiotic divisions, at the gamete stage or after fertilization. A study of differen- tial selection in any one of these categories must account for possible variation from the other areas. Although some investigators have suggested that the genes will not function in sperm and that all sperm will have ‘an equal probability for fertilizing an egg (Muller and Settles, 1927), there is increasing evidence from a number of studies that this position is no longer tenable. It has been postulated that some sperm may be regularly non- functional as a consequence of meiotic segregation (Novitski and Sandler, 1957); the non-functional proportion may regu- larly include one-half of all the sperm produced in Drgy sophila melanogaster (Peacock and Erickson, 1965). Depending on their genotype, Drosophila sperm will exhibit a differential response to irradiation (Lindsley t al,, 1963), temperature shock (Frydenberg and Sick, 1960), and retention in the female's storage organs after double matings (Lefevre and Jonsson, 1962b). Aging will also produce the same response (DeVries, 1962, 1964). The experiments described here were performed to deter— mine if selection could be detected at the gamete level within a sperm population heterozygous for second chromo— some markers. Preferential association of non-homologues could be studied, since any cross would produce at least four classes of progeny (each sex would exhibit the two autosomal markers). For each cross, a study of post fertilization mortality was made to eliminate this influ- ence from the progeny ratios. LITERATURE REVIEW The subject of sperm survival and competition may well be prefaced by a brief description of the major events which make up the life of a sperm. The following description of spermatogenesis in Drosophila is taken largely from Cooper (1950). Spermatogenesis consists of the various divisions and transformations which primordial male germ cells undergo to become spermatozoa. The testis of a newly emerged larva of a male Drosophila melanoqaster will contain only spermato- gonia, and until shortly before pupation only spermatogonia and primary spermatocytes will be found. After the onset of pupation, the testis will contain all stages in sperma- togenesis, including sperm. The testis of the imago principally contains spermatids and maturing sperm but other stages of spermatogenesis are also present in male Drosophila. Spermatogenesis will apparently continue even after males have become infertile with age, due to failures in the sperm transfer mechanism. Two types of spermatogonia occur in Drosophila: "pri- mary" spermatogonia, which do not occur in groups and which do not undergo synchronous divisions, and "secondary" spermatogonia. The original secondary spermatogonia will undergo four synchronous divisions; all of these division products will be contained in a well defined cyst. Enter- ing meiotic prophase, each cyst will contain sixteen primary spermatocytes. The two subsequent meiotic divisions will result in a bundle of sixty—four spermatids which differ- entiate into spermatozoa. The maturing sperm pass from the testis into the seminal vesicles. Here they are described as either being still grouped in bundles (Miller, 1950), or as no longer being in bundles, with the sperm heads dispersed throughout the vesicle (Lefevre and Jonsson, 1962b). Spermatozoa leave the vesicles by means of ejaculatory ducts; accessory glands empty their products into the ejaculatory duct at its proximal end. A sperm pump is located near the distal end of the ejaculatory duct. The mature sperm are transferred to the genital chamber of the female and migrate to the sperm-storage organs (in Q, melanogaster two spermathecae and a seminal receptacle), where they are retained within these storage organs until they are utilized in fertilization. Utilization of sperm appears to be non-random and any progeny sample of suffi- cient size should reveal an equal recovery of homologues. The various events of spermatogenesis, and the sub- sequent transfer, storage, and utilization of the sperm, may be considered as periods when errors may occur which can alter normal reproduction. Malfunctions may act in a non-specific way, to simply reduce the number of poten- tial progeny. In other instances, the aberration may act selectively to produce a progeny ratio which will deviate from the expected frequencies by a reduction or elimination of a given category of progeny. The degree to which these errors are operative will be reflected as major or minor variations in the resultant progeny ratio. While an off- ratio is generally considered to result from some abnor- mality, there is conjecture that, for some cases anyway, off-ratios may be considered "normal" (Hanks, in press). The deviations which are generally the easiest to study are those involving the sex chromosomes. The X, Y sex determination mechanism of Drosophila provides an expectation that both sexes be recovered in equal numbers. One of the earliest accounts of abnormal sex ratio was reported by Morgan (1911). Female Drosophila carrying the mutant for rudimentary wings are almost completely sterile when mated to mutant males. The same females when crossed to nonémutant males produce daughters but rarely sons. Lynch (1920) observed similar results for rudimentary as well as other sex—linked mutants. The embryology of the rudimentary mutant was studied (Counce, 1956c); it was found that development was normal up to primary organogenesis when there was a complete cessation of differentiation between the thirteenth and sixteenth hours of develOpment. At times certain genes may act to alter sex ratios. Bell (1954) reported a mutant (daughterless, ga) which produces only male progeny. The mutant transformer (tra) will result in transforming females into males (Sturtevant, 1954). The factor known as "sex ratio" (SR) has been found to be transmitted maternally and acts by disturbing the development of male zygotes to a degree that almost uni- sexual progenies are recovered. The agent causing this disturbance, which had been described as a plasma gene, cytoplasmic particle or virus, was demonstrated to be a small spirochete (Poulson and Sakaguchi, 1961). Another phenomenon, also referred to as "sex ratio," was first described by Gershenson (1928) in a natural population of Drosophila obscuga, Sturtevant and Dob- zhansky (1936) found a similar condition in Drosophila pseudoobscura. In both cases, females were recovered in far greater number than males. Gershenson attributed the differential recovery to a sex-linked mutant whose action was analogous to a gametic lethal. The greatest proportion of those sperm which should act to determine males do not. participate in fertilization. Sturtevant and Dobzhansky indicated that their cytological studies showed an equa- tional division of the X chromosome at each meiotic division accompanied by a degeneration of the Y chromosome. The meiotic products would consist wholly of sperm all carry- ing an X-chromosome. Re-examination of this problem (Novit- ski, Peacock, and Engel, 1965) has demonstrated that although the Y chromosomes do degenerate, there is no extra replica- tion of the X chromosome. The sperm population would, in the latter study, consist of functional X-bearing sperm and non-functional sperm containing a Y chromatin mass. A male sex ratio (MSR) condition in Drosophila affinis was described by Novitski (1947). This condition acts to produce greater numbers of males than females. Although the particular mode of action has not been examined cy- tologically, it may be similar to the female sex ratio condition described above for Drosophila pseudoobscura. While deviations from expected values are most obvious when they involve the sex chromosomes, it may also be said that autosomal factors will act in similar ways to produce aberrant ratios. Since the present research involves the segregation and subsequent action of autosomal elements, it may be of interest to examine the more obvious stages in the life of a sperm in considering the more general question of whether all sperm are capable of fertilizing an ovum. There now exists a number of studies which would indicate that in special cases at least, not all sperm will function to fertilize an egg. Examples of differential ability to effect fertilization may be traced to various factors act- ing during spermatogenesis, sperm transfer, storage and utilization. Examination of the relative recovery frequencies of gametes from males carrying the translocation T l, 4 BS [Translocation (l, 4) Bar of Stone] led Novitski and Sandler (1957) to postulate that not all meiotic products are functional. Complementary classes were not recovered in equal numbers among the progeny of the translocation males. To determine the possible cause for the inequality, zygote mortality was tested by egg count and ruled out as a possibility. Gamete lethality was considered but pro- geny classes bearing the genetically normal meiotic products (normal X and normal Y) were also recovered at a differential rate relative to one another. As an explanation, it was postulated that not all products of spermatogenesis are functional. This would not be ordinary gamete lethality because the geometry of the meiotic divisions would deter- mine which gametes would function. The term meiotic drive has been suggested (Sandler and Novitski, 1957) to describe a force potentially capable of altering gene frequencies which functions as a consequence of the mechanics of the meiotic divisions. The Bar of Stone translocation described above would answer this description as would the various "sex ratio" anomalies (Gershenson, 1928; Sturtevant and Dobzhansky, 1936; Novitski, 1947). The case involving the translocation may be referred to as chromosomal meiotic drive whereas qenic meiotic drive describes the "sex ratio" conditions. Another example of genic meiotic drive is to be found in the segregation-distorter (S2) mutant in Drosophila melano- - gaster (L. Sandler, Y. Hiraizumi, and I. Sandler, 1959). Located in the right arm of chromosome 2, the §Q_locus is recovered in greater frequency than its allele among the 10 progeny of heterozygous males. This phenomenon does not occur in females and apparently requires synapsis. Sand— ler §£_§1, (1959) postulated that SQ acts in some way to prevent the sperm bearing the normal allele from forming, or alternatively, from functioning normally once it had been formed. These authors presented a cytogenetic model based on preliminary evidence in which the §Q_locus would cause a break in its homologue while synapsed. A sister strand reunion would result, causing the formation of a dicentric bridge at anaphase II which would prove to be lethal to the affected cells. An extensive cytogenetic study of the SQ locus was undertaken by Peacock and Erick- son (1965) to describe the action of this mutant. Cytologi- cal examination of meiotic stages not only failed to con- t gl, (1959) but firm the preliminary results of Sandler indicated that the meiotic process was quite normal. The postmeiotic cells were traced through maturation and no difference was found in the number of sperm contained in sperm bundles of the §Q_and control groups. Dissections indicated both §Q_and §Qf sperm were motile as they entered the seminal vesicle. Because no basis for §Q_function had been found in the male, the study was extended to sperm counts in the female to ascertain if some form of selection 11 would occur there. Young males transferring relatively low numbers of sperm were utilized. Comparisons were made between the number of sperm deposited in the vagina and the number which were stored. These counts clearly indicated that both §Q_and ng sperm were stored so any selection would have to act between the time of sperm storage and fertilization. Males were mated to two females, one of which was dissected for counts of stored sperm, the other . kept for progeny counts. Counts for both Oregon-R (wild- type) and §Q_males demonstrated that the sperm count was on the order of twice the progeny count. Peacock and Erick- son interpreted these rather unexpected results on the basis of Novitski and Sandler's (1957) proposal that some portion of the sperm regularly produced by a Drosophila melanoqaster male may be non-functional. The fact that the Oregon - R males exhibit the same differential between sperm counts and progeny counts as do the S2.males suggests that one-half of all Drosophila sperm may be regularly non- functional. While the basis for the aberrant ratio is directed back to the first meiotic division when §Q_separ- ates from its §Qf counterpart, the actual cause of the non- functioning is unknown. Orientation on the metaphase I plate must be such that the SDebearing chromosomes are directed to the "functional" pole. 12 The segregation of homologues at anaphase I to func- tional and non-functional poles has also been suggested to explain some sex ratio phenomena. Yanders (1965) has found a paternal age-related sex-ratio shift in which a greater number of female progeny are recovered as the paternal male ages. This shift has been attributed to a greater tendency for the Y chromosome to segregate to the non—functional pole as the age of the paternal male increases. The significant differences in the mean sex ratios of Oregon - R and Canton - S strains of Drosophila melanogaster have been explained by Hanks (1964) on the basis that the X chromosome will preferentially segregate to the functional pole and the Y chromosomes to the non- functional pole. The size of homologues may also influence their re- covery (Novitski, 1951). When homologues are of unequal size, the smaller of the homologues is recovered with about twice the frequency of the larger homologue. This non-random disjunction was further studied by Novitski and Sandler (1956) using a tandem metacentric and tandem acrocentric compound X chromosome with or without a homologue. When the homologue was absent, the number of newly generated single chromosomes formed by crossing over was reduced due 13 to some portion of these single chromosomes acting as lethals. Preferential recovery of the smaller homologue was also noted in the BE_translocation studied by Novitski and Sandler (1957). The authors felt there was only a superficial resemblance between this phenomenon and the non—random disjunction observed earlier, since crossing over was a necessary requirement for non-random disjunc- tion - a feature not found in the Drosophila male. Although it is usual for non-homologous chromosomes to assort themselves independently at meiosis, exceptions have been found (Grell, 1959). A non-random assortment was found between the Y chromosome, the free fourth chromosome and a 3-4 translocation carried by a Droso- phila female. The Y and free fourth chromosomes would segregate to opposite poles with a frequency as high as 92 percent. The non-random assortment was interpreted as the result of an association between the Y and the fourth with a subsequent disjunction. A sperm which has managed to avoid any complications during the process of spermatogenesis and maturation still is faced with any number of pitfalls before fulfilling its raison d'etre. The mature sperm must be transferred 14 to a female, be retained within one of the sperm-storage organs and maintain itself until participating in fertilization. As many as 3,000 to 4,000 sperm may be transferred to a female during a single mating (Kaufman and Demerec, 1942). A high proportion of wastage is experienced at this stage since the storage organs will only accommodate 500 to 700 sperm (Kaplan, §E_§1,, 1962; Lefevre and Jonsson, 1962b). Because storage space is limited, only a fraction of the total sperm population will be retained. Any sperm left in the genital chamber when the first egg is laid after insemination are voided with the egg. A potential for selection will exist at this level for some sperm may be better able to gain immediate entry into the female's storage organs. Once the sperm are stored, their utilization is quite efficient. Early accounts of polyspermy (Sonnenblick, 1950) indicated that five to eight sperm are typically found within the Drosophila egg although the number may go as high as thirty. More recently, Hildreth and Lucchesi (1963) found one sperm present per fertilized egg in most instances. Some dispermy was found but only in a few cases; no polyspermy was found. 15 Additional evidence for monospermy comes indirectly from the sperm-progeny count study of Lefevre and Jonsson (1962b). This study found individual females producing in excess of 500 progeny from single matings indicating a high degree of efficiency in sperm utilization after storage. An early description (Guyenot, 1913) of sperm resorption stored in the ventral receptacle has not been collaborated in a more recent study (DeVries, 1962, 1964). Sperm may be lost from the storage organs by means of cold temperature shocks (Muller, 1944; Novitski and Rush, 1948; Scossiroli, 1954; Frydenberg and Sick, 1960; Lefevre and Jonsson, 1962a). Frydenberg and Sick report a dif- ferential degree of resistance on the part of the sperm to cold shocking the inseminated female. Some strains are very sensitive, producing only one or two fertilized eggs after the treatment; other strains will consistently deposit twenty or more fertilized eggs. Their preliminary results indicate that the response of the sperm is somehow influenced by the chromosomes which are carried, although the nature of this influence was not discussed. Sperm are not passive cells which lie inert within the female prior to fertilization. The female reproduc- tive tract, while adapted to storing sperm, still presents 16 an environment which is different from that which the sperm experienced while yet in the male. These differences appear as various interactions which have been observed to occur between sperm and the female reproductive tract. Smith (1956) found that Drosophila subobscura females rarely mated a second time. An adequately inseminated female will remain unreceptive to further mating for the remainder of her life even though her sperm supply has been exhausted. Smith explained this reluctance to mate as the consequence of a factor which would switch off the receptivity mechanism. The absence of sperm would not turn it back on again. Drosophila melanogaster females are more receptive than are 2, subobscura to mating a second time. Manning (1962) postulated two ways whereby insemination turns off the receptivity mechanism and sperm exhaustion switches it on again. Sperm may signal their presence by a synchronized lashing movement within the storage organs or some chemical may be produced by the sperm themselves or by the interaction of the sperm with the female tissues. Manning felt that the evidence favored the second alternative. Chemical interactions between gametes have been described for mollusces, annelids, echinoderms and chordates (C.B. Metz, 1955). 17 To postulate the presence of some similar chemical inter- action in Drosophila may not be unwarranted. Patterson and Stone (1952) describe an "insemination reaction" which occurs soon after coitus in some species of Drosophila. The reaction consists of a rapid secretion of fluid into the vaginal cavity causing the vagina to swell to three or four times its normal size. In homo- gametic (intraspecific) matings the contents of the en- larged vagina remain soft and are soon expelled allowing the vagina to return to normal size within a few hours. The intensity of the reaction varies with the particular species, ranging from no apparent reaction, to a slight reaction, to a strong reaction. Drosophila melanggaster was classified as a species which showed no apparent reac- tion. While there is no apparent reaction, there is evi- dence that some sort of interaction between sperm and the female genital tract does occur in Drosophila melanogaster (Yanders, 1963). The quantity of sperm found in the seminal receptacle was the measure of insemination success and was scored by means of the Insemination Index (I. I.). (Yanders, 1963, 1964; DeVries, 1962, 1964). I. I. scores were de- termined for sixteen male-female combinations of four geo- graphical strains. The I. I. values indicate that in certain 18 combinations of strains, the female tract has an inhibi- tory effect on the rate of sperm migration to the seminal receptacle resulting in a type of insemination reaction differing from that described by Patterson and Stone (1952). These studies of insemination success for inter- and intra—strain matings were continued by DeVries (1962, 1964). Females from the various matings were dissected after inter— vals of twenty-four hours, eight days, fifteen days or twenty-nine days. The I.I. values indicated the quantity of sperm stored was different for the various crosses but seemed to be consistent for a particular male-female com- bination. Sperm from different strains of males were lost at a differential rate over a period of time when mated to the same female strain. The variation in the initial quan- tity of stored sperm and the different rates of sperm loss over a period of time suggested that the sperm would ex- perience a differential viability within the female tract. Hildreth and Carson (1957) considered the possibility of inseminated females exerting a differential mutagenic effect on the X chromosomes of irradiated sperm at some time between insemination and fertilization. The §y_ ("2_§£Ei,§_l2&2_z_!E/z_g_£,XE?) females produced sex- linked lethals at a rate 7.8 times that recovered from S Basc (sc 1 §_InS w3_sc8) females. 19 Sperm from wild-type males stored in the ventral re- ceptacles of glossy (an allele of lozenge - symbol 1&3) females become inviable or non-motile within forty-eight hours of insemination (Oliver and Anderson, 1942). These females were still fertile after being mated a second time, the progeny produced, being exclusively from fertili- zation by the sperm of the second mating. The uni-sexual progenies found in Sciara (C. W. Metz, 1929) were attributed to an elimination or inactivation of those sperm which would normally determine the sex of the deficient class. There was no direct evidence to show if the elimination or inactivation was caused by the egg or by something operating in the storage organs or ducts of the female before the sperm reach the egg. In addition to an interaction with the female repro- ductive tract, sperm have been found to compete amongst themselves for space within the storage organs. This competition is best demonstrated when females are mated to two successive males, each of which carries genetic markers allowing the relative success of each insemination to be detected in the progeny. Evidence for sperm competition was presented by Loba- shov (1939) based on double matings of females with mutant 20 (w_and s y_§_car) and wild-type males. The wild—type sperm were more successful in displacing the mutant sperm used for the initial insemination and also in maintaining themselves in the storage organs when mutant sperm were introduced from the second mating. The wild-type sperm were credited as being more viable than the mutant sperm and therefore better able to function in fertilization. Lefevre and Jonsson (1962b) also found wild-type sperm more effective in displacing mutant (y_§3:_ga£) sperm than vice versa. Although these authors did not feel that the advantage of the wild-type sperm was due to a difference in viability, it was thought that the wild—type sperm were more efficient in storing themselves regardless of what the mechanism might be. Sperm competition of a different sort is also indicated. Recall that Lefevre and Jonsson (1962b) found a high effi- ciency in the utilization of stored sperm providing in- direct evidence against regular polyspermy in Drosophila. Over 500 sperm, close to the upper limit of capacity for the storage organs, would produce progeny. This is in apparent contradiction to the study of Peacock and Erick- son (1965) which indicated that one-half of Drosophila 21 sperm would not function in fertilization but could be stored along with the functional sperm. The young males used by Peacock and Erickson did not transfer a sufficient number of sperm to fill the storage organs to capacity. In this case all transferred sperm could be stored and the two classes of sperm distinguished. In contrast Lefevre and Jonsson used older males which deposited larger numbers of sperm. It was proposed that in the heavier inseminations, the functional sperm would gain access to the storage organs while the non-functional sperm would be competitively excluded. t al, (1965) Rather than using genetic markers, Gugler studied sperm displacement in twice-mated females by labeling the sperm of the second male with tritiated de— oxycytidine. Within twenty minutes of the second mating more than half of the sperm from the first mating was replaced by labeled sperm. Gametic selection has been studied by other methods in both Drosophila and mice. Irradiation of males prior to insemination indicates that some aspect of sperm behavior has been affected (Yanders, 1959, 1964) since irradiated sperm are less likely to be retained over periods of storage than are non-irradiated sperm. Physiological damage resulting 22 in loss of motility could account for the observed differences. Studies by Lindsley _§_§1, (1963) indicated that the gametic genotype will influence the survival of a sperm following irradiation where survival is measured by the ability of a particular sperm to produce an adult fly. X-bearing sperm and Y—bearing sperm show the same sensi- tivity to radiation. Increasing sperm sensitivity may be found in sperm containing zero, one or two sex chromosomes. Heterochromatin, in its normal position proximal to the centromere of the X—chromosome, causes no effect in the sensitivity of the sperm carrying it. When this proximal heterochromatin is shifted to the distal end of the chromo- some the sensitivity of X-bearing sperm is increased. In the mouse, Bateman (1960b) has found a T_a11e1e (353 Which will influence the sperm carrying it to enter an egg of a favored genotype. When a choice of eggs is presented by heterozygous, (T/§_and T/EE) females, the t3_- bearing sperm unite more frequently with the normal (+) than the brachyury (T), and more frequently with brachyury than 3: eggs. 23 Six other T_alle1es were studied by Braden and Weiler (1964) who found heterogeneity of progeny ratios existing not only between males carrying the same tfallele but also between litters from the same male. In order to detect these differences the authors utilized a mean square con- tingency test--a test useful where cell frequencies are too low to permit use of a chi-square test. The authors con- cluded that tfalleles have an effect on the function of the sperm in which they reside. The type or magnitude of the effect can be influenced by relatively small changes in the physiological state of the female. There is evidence that deletions in the dilute-short ear region of chromosome 2 in the mouse may affect sperm function (Russell and Russell, 1960). Sperm carrying the deletions are transmitted at less than normal frequency to the progeny due to a semi-lethal effect on the gametes. Should a given sperm be used to fertilize an egg, several factors should still be considered to avoid confusing their effect on the zygote with a pre-fertilization phenomenon. Many lethals will act for a variety of reasons to arrest development and cause the death of the zygote (Hadorn, 1961). Discussion here will be limited to a few examples of factors 24 Which have an influence on fertilization or which will act soon after fertilization to affect progeny number. In Drosophila the gene deep orange will act even before fusion of the gamete nuclei (Counce, 1956a). The gene suppressor of erupt will produce a gene product or sub— strate which is present in full amount within eight minutes after fertilization (Glass and Plaine, 1950). Maternal or cytoplasmic effects have been observed to influence egg hatch or zygote development. Redfield (1957) noted a higher incidence of unhatched eggs, larval and pupal deaths when the mothers carry Qg£1y_and gayng_inver- sions. These same effects were considerably lower when neither parent carried the inversions or when the inver- sions were carried by the male. Bateman (1962) found that in two inbred lines (B and O) of Drosophila melan0gaster an incompatability developed between eggs with O cytoplasm and sperm of type B or B0 (the F1 hybrid with B as the female parent). The incompatability observed took the form of unhatched eggs. Meyer (1954) reported a high rate of embryonic deaths in Drosophila due to abnormal or in- adequate material supplied by the female parent. Females homozygous for the markers bw_§p; ;i_§_(chromosomes 2 and 3) 25 were mated to al_b_gg_§p_ma1es. The reciprocal cross-was also made. Fewer eggs produced larvae in the first cross than in the second. All eggs were known to be fertilized, the proportion of flies developing from emerged larvae was very similar for both crosses, and the possibility of a sex- linked embryonic lethal was eliminated. Aside from lethal and sub—lethal mutants, different strains of Drosophila will tend to develop patterns of egg hatch and egg laying which are unique to the particular strain. Johansen (1963) found hatchability to vary from one inbred wild stock to another. Hybrids between the inbred strains gave a heterotic effect whereby the egg hatch of the hybrid was higher than either parent. Dif- ferent strains have also been found to lay eggs at a dif- ferent rate (Gowen §E_§1,, 1946). If this is the case over a specified period of time used for egg-laying in an experi- ment, one might expect to obtain different quantities of eggs laid. This would also reflect a differential sampling of sperm assuming a high correlation between eggs laid and eggs fertilized. METHODS AND MATERIALS Stocks One wild type (Oregon-R) and eight mutant stocks of Drosophila melanoqaster provided the flies used in these experiments. Of the mutant stocks, two are the common laboratory stocks, Muller-5 (Basc) and Curly/Plum (Cy/Pm). The Muller-5 (M-5) stock is routinely used in the detection of sex linked lethals and carries the markers B wi_sc8 (Bar, white-apricot, inversion scute-8) on the X-chromosome. The mutants Cy and Pm_are dominant chromosome II markers, each of which is associated with extensive inversions in both arms. Either marker is usually lethal when homozygous and when both are present in the same fly, one on each homolOgue, they constitute a balanced lethal system. Five second chromosome recessive lethals were balanced with the Qy_- bearing homologue and used for early experi- ments (Curly/lethal). They were designated as le-2 (lethal - g), 1e-3, le-4, le:§, and 1e-22. Lethal - 9 was used exten- sively in later experiments. These recessive lethals were induced by exposure to approximately 2,000 r units of X-rays applied at about 26 27 136r/minute. The source of the X-rays was a General Elec— tric Maximar - 250-111 operated at 250 kv, 15 ma, with a .50mm copper filter. Second chromosome lethals were de- tected by standard tests and maintained as balanced lethal stocks. One of these lethals, lethal - 9 was balanced with the Plum chromosome (Pm/lethal - 9). Food and Temperature All stocks were maintained by mass transfers and kept on modified Carpenter's nutrient medium (Carpenter, 1950). Flies selected for experimental matings were placed on a sugar-agar medium (modified after Offerman, 1936) to pre- vent the freshly inseminated females from laying eggs. The flies were kept in a constant temperature room at 230C. except when brought out to room temperature for mat- ings, counting of progeny, and other routine handling. Some of the early experiments involved storage at 10°C. as dis- cussed below. High relative humidity and elevated room temperature during July and August of 1964 may have contributed to the high mortality rates noted in the sex-linked lethal studies. These same environmental factors may well have been evident at other times but were not critical. The other experiments 28 only involved enumeration of the F1 generation rather than remating of the F1 females as is the case for the sex-linked lethal test. Ratio Experiments Initial Experiments Initial experiments are described extensively elsewhere (Myszewski, 1962) and the present discussion will be limited to a brief resumé of these experiments. §y1§g_males and males from each of the Cy/lethal stocks were mated to Oregon-R females. The OR females were collected as virgins and stored for seven days at 10°C. After this storage period, the females were mass-mated overnight to males of the various genotypes. After the mating period, the males were discarded and the females divided into two groups of equal size. One group was placed individually in shell vials on nutrient medium and allowed to lay eggs immediately. The second group was collectively placed in a bottle containing Offer— man's medium and returned to cold storage at 100C. for 15 days. At the end of this time they were moved from cold storage, and individually placed in shell vials containing nutrient medium. Both groups were allowed to lay eggs for six days before being transferred to a fresh vial for another 29 six day period. At the end of the twelve day laying period the females were discarded. Each of these crosses was done twice. Later Experiments After the initial experiments, more extensive ratio testing was accomplished by extending the number and vari- ety of stocks used for the matings. The following crosses were made: Cy/lethal - 9 dd x M—5 99 Cy/lethal - 9 66 X OR 99 Pm/lethal 9 66 x OR 99 Pm/lethal - 9 55 X M-5 9? The males from the above crosses were routinely col- lected from stock bottles prior to mating. These stock bottles had been used two or three days earlier to collect virgin females from the first emerging flies from the cul- ture. In this way, while the age of the males varied, it would generally range from a minimum of two days to a maxi- mum of seven days. If it was necessary to store the males for a few days (while the females were aging) they were kept on nutrient medium. For one series of matings CyZlethal - 9 55 X M-5 9?), the males were irradiated 30 prior to mating. These experiments will be described in the section on sex-linked lethals below. Female flies were collected as virgins and stored on nutrient medium for at least 3 but not more than 5 days. For these experiments there was no cold storage prior to mating - the females were kept at 230C. Matings were accomplished by placing etherized flies of both sexes in bottles containing sugar-agar medium. They were under observation until the majority of flies had recovered at which time they were returned to the constant temperature room. Most matings were made up in the afternoon and allowed to continue until the following morning. In no experiment was the time interval for the matings less than 16 hours. Matings were made using a minimum ration of 3 65 to 2 99 ... although where enough male flies were available 3 66‘:1 99 were used. By using these mating procedures, it was hoped to achieve maximum insemination of the female. After mating, males were separated from the females and discarded. The females were placed in individual shell vials containing nutrient medium and allowed to lay eggs for twelve days; six days in each of two vials. The females were then discarded. Each of 31 the Cy/lethal - 9 dd X M-5 9? crosses was done once. All of the other later ratio experiments were repeated twice; the Cy/lethal - 9 55 X OR 99 was repeated five times. Progeny began to emerge from the culture vials 11 days after commencement of egg laying and all of the progeny were counted on alternate days. The counting period covered 10 days starting from the appearance of the first emerging flies. The progeny were counted and the data tabu- lated in such a way as to obtain the maximum amount of information. Both the sex and phenotype of each fly were recorded. In the Pmllethal - 9 55 X M-5 9? cross the action of the wi_locus prevented expression of the Pm_marker in the hemizygous males. This interaction made it impossible to separate Pp_bearing males from those having the lethal - 9 homologue. Females carrying the M;§_chromosome heterozygously could be classified with regards to the autosomal markers. The counting procedures prevented many flies from be- coming stuck in the food. Those which were stuck were usually able to be identified. Those which were stuck and unable to be identified were extremely few and did not in- fluence the total results. 32 Sex-Linked Lethal Experiments Cy/lethal - 9 and OR males were exposed to 0, 2,500, or 5,000r units of X-rays. Radiation was administered by a General Electric Maximar - 250-111 operated at 250 kv, 15 ma with a .50mm copper filter. The dose rate was cal- culated to be 152r/minute. Both the gy/lethal - 9 and OR males were mated to M;§ females. The OR X Ml; cross con- stituted the controls for the other mating. The female progeny of these crosses were mated to M;§_males and the F2 generation was scored for sex-linked recessive lethals. Some procedures for the experiments were modified for the Cy/lethal - 9 55 X M—5 9? mating from the usual sex-linked lethal test. The primary modification was the collection of the F1 females as virgins. Because the association of newly induced sex-linked recessive lethals with autosomal markers was being tested, the females had to be mated with M;§_males rather than their male sibs who also were carry— ing autosomal markers. In addition to scoring all of the F1 females for reces- sive sex-linked lethals, all the F1 progeny were scored for sex and genotype. A number of the F1 females (3-10%) from each Pl female were selected at random at the time they 33 were being placed in their individual shell vials. All of the progeny from these sampled F1 females were counted, recording both the sex and phenotype in a manner identical to the ratio experiments. The induced sex-linked recessive lethals were recorded for association with a Curly or lethal - 9 chromosome in addition to their frequency. Tests for Mortality Cy/lethal - 9 males were crossed to OR females for the first of two experiments designed to detect and measure mortality during developmental stages. The collecting and mating procedures described previously were followed. The females were placed in individual shell vials on nu- trient food and allowed to lay eggs for a similar period. Transfers were continued for 12 days for all females. On alternate days the vials were placed either in an 180C. or a 23°C. constant temperature room. Six vials of progeny from a given female would then develop at 180C., six vials of progeny would develop at 230C. To compensate for any differences from day to day, half of the vials from all the females were placed at 18°C. the other half at 23°C. as indicated in the following table: 34 Day 91 92 93 94 ..... . 9N 1 23° 18° 23° 18° 2 18° 23° 18° 23° 3 230 180 23° 18° 4 18 23° 18° 23° 0 O O O 12 18 23 18 23 Development at the normal rate proceeds at a tempera- ture of 23°C., but is retarded at the lowered temperature (18°C.). Since less viable individuals would be more likely to be eliminated from the marginal situation than the normal ones, compiling the data in this way makes it possible to detect differences within the progeny of a given female as well as to sum the data for all the females in order to test for temperature dependent variations. This experiment could only test for a differential via- bility in emerging progeny and it was necessary to evaluate the total potential for mortality in the developing individ- ual. To this end, development was followed by counting the total eggs laid, the number of eggs hatched and the number of adults which developed from the hatched eggs. This 35 procedure was utilized for the following crosses: M-5 99 X M-5 99 Pmlle—9 96 x OR 99 OR 96 X OR 99 Cy/Pm 66 X OR 99 Cy/Pm 96 X Cy/Pm 99 Cygle-g 69 X OR 99 Cygie-9 99 x Cygle-9 99 pmg1e-9 99 x M—5 99 Pm/le-9 66 x Pm/le-9 99 Cy/le-9 99 x M-S 99 The collecting and mating procedures described previ— ously were followed. Ten females were started for each mating. The females were transferred on 12 successive days to fresh shell vials containing nutrient medium. After transferring the female, the eggs which she had laid that day were counted. Egg hatch was determined twenty-eight to thirty-six hours after the female had been removed. The progeny from each vial was counted; sex and non-egg hatch. . E a 't autosomal markers were recorded gg mort 11 y ( total number) and larval mortality (total egg hatch - number emerging adults) total egg hatch were computed for each cross. RESULTS Ratio Experiments Initial Experiments The tabulated data for the initial ratio experiments involving the second chromosome recessive lethals are shown in Table 1. These data represent the summed results of both experimental runs. Each separate run as well as the summed data was analyzed by 2 X 2 and 2 X 4 contingency chi-square tests for the six stocks. Three stocks, lethal - 2, lethal - 4, and lethal - 22 showed no heterogeneity for any of these tests. The differences found in lethal - 3 were attributed to a greater recovery of lethal-bearing females than lethal—bearing males in the stored group. While this difference was significant,there was almost an equal recov- ery of §p£1y_and lethal - 3 chromosomes for both the stored and non-stored groups. The differential recovery of the sex chromosomes may have been a chance fluctuation, in that the total Curly :lethal - 3 ratio for the stored group did not significantly differ from a 1:1 recovery. The Cngm data also showed significant differences for portions of the data. Within the total number of Plum pro- geny, both the stored and non—stored groups produced more 36 37 TABLE 1 Summed data from initial ratio experiments Curly Lethal Total Lethal males females males females non-stored 2954 2992 2941 3047 11864 le-2 stored 3460 3512 3424 3579 13975 Total 6414 6434 6365 6626 25839 non-stored 2437 2541 2420 2387 9785 1e‘3 stored 3075 3074 2918 3187 12254 Total 5512 5615 5338 5574 22039 non-stored 3859 4140 3869 3910 15778 1e-4 stored 2335 2437 2326 2403 9501 Total 6194 6577 6195 6313 25279 non—stored 2316 2506 1980 2156 8958 le-9 stored 2215 2285 2121 2280 8901 Total 4531 4791 4101 4436 17859 non-stored 3561 3707 3409 3522 14199 1e-22 stored 3437 3565 3268 3556 13826 Total 6998 7272 6677 7078 28025 Curly Plum Total males females males females non-stored 10851 11789 10886 11748 45274 Cy/Pm stored 11341 12339 11222 12663 47565 Total 22192 24128 22108 24411 92839 38 females than males. This difference was seen again when only the stored group was analyzed. Within the stored group, both Curly and Plpm_categories contained more fe- males than males. The Curly and glpm_chromosomes were recovered in equal numbers. The experiment with lethal - 9 showed a difference in the recovery of §y_y§_lethal - 9 chromosomes. Significant at the 1% level, this difference was most evident in the non-stored group. Although still favoring the §y_chromosome, the difference was reduced upon storage to a point where it was no longer significant. When the cross(Cy/1ethal - 9 55 X OR 99) was repeated the same effect was noticed. The results of the initial Cy/lethal - 9 56 X OR 99 crosses are added to the data from these repeated experiments and are found listed on line 2 of Table 2 as Cygle-9 X OR NS. To summarize these initial experiments it may be noted that of the six mutant stocks, lethal - 9 alone produced what appeared to be a differential recovery of the two second chromosomes. This effect was more evident in the non-stored group than the stored group. The two homologues Curly and 2123 were recovered with a frequency not signifi- cantly different from one another for both the stored and non-stored group. 39 TABLE Results of later experiments testing segregation le-9 ratio, totals of all crosses, breakdown of data with sex ratio and X2 values Treat- Total Total Total Total 99 x 99 ment 96 99 le-9 Plum Pm/le-9 x M-5 N87 3085 3689 1753 19361 Cy/le-9 x OR NS 13967 15251 14237 ----- Cy/1e-9 x OR 87 10661 11814 11218 _____ Cy/Pm x OR NS 21737 23537 ----- 22634 Cy/Pm x OR 8 22563 25002 ----- 23885 Pm/le-9 x OR NS 4794 4819 4903 4710 Cy/le-9 x M-5 2500r 4302 5151 4938 ----- Cy/le-9 x M-5 5000r 362 480 441 ----- Cy/le-9 X M-5 Or 1181 1369 1336 ----- a l. 1e-9 due to w 2. 3. 4. 5. 6. Sex Ratio 7. NS Nonstored; S 8. Females only--ma1es could not be distinguished as gm_or locus on chromosome one. Total progeny = 95 + 99, not le-9 + gm, Computed as le-9/Tota1 for females only. 1e—9 Ratio = le-9/Total Progeny. Computed as the Pm/Total Progeny. Total 99/Tota1 Progeny, 55 + 99. Stored. N.S. = Not significant. 40 2 of autosomal markers including progeny regards to sex and autosomal markers, for deviation from 1:1 ratio X2 Value for Deviation Level 4 6 from 1:1 of Total Grand le-9 Sex Ratio Signif- Curly Total Ratio Ratio for le-9 icance ----- 6774 .47513 .5445 9.07 1% 14981 29218 .4872 .5219 18.94 1% 11257 22475 .4991 .5256 0.06 N.S.8 22640 45274 .49995 .5198 0.0+ N.S. 23680 47565 .50245 .5256 0.88 N.S. ----- 9613 .5100 .5013 3.87 5% 4515 9453 .5223 .5449 18.92 1% 401 842 .5237 .5700 1.90 N.S. 1214 2550 .5239 .5368 5.83 2.5% 41 Later Experiments The later experiments consisted of the crosses listed in Table 2 with the exception of Cngm X OR (both S and NS). The Cngm X OR (both S and NS) data are included here for the sake of comparison and are the same as listed in Table l. The Cy/le—9 X M;§ matings represent the F1 generation which was counted for the sex linked recessive lethal test. The data for the various crosses are ranked with regards to the relative recovery of the lethal - 9 chromosome. This is referred to in Table 2 as the l§;§_ratio and represents the number of 1g;2_bearing progeny divided by the total number of progeny. The ratio for the gyggm_x OR crosses is computed as Plpg_progeny over total progeny. An equal recovery of lg32.and gy chromosomes would be represented by an l§:2_ratio of .5000. Deviation from this value may possibly be taken as a measure of differential survival of one of the homologues. The totals listed for the szle-9 <59 x 114;; 99 in the "total lgzgj" "total gm? columns represent only the totals for the female prOgeny. The wi_locus on the female X chromo- some, transmitted to all of the males, prevents any expres- sion of the glpm_or wild-type eye color. All of the males from this cross look alike and autosomal markers cannot be scored. All of the statistical tests which were used to 42 assess differences between the crosses were done in two ways when the ngle-9 95 X Ml; 99 cross was involved. When the relative recovery of autosomes was being compared,the tests were run using (1.) only females and (2.) both males and females. There was only one instance where significance was shown by the total data and not by the females alone (Table 3, Box 58a, 58b). Because all the tests but one could be reproduced using only female totals,it was felt that the inclusion of the Pm/le-9 X M:§_data calculated in this man- ner was justified. It may be noted in passing that the one test which failed to show agreement was the comparison with the Cygle—9 X M-5 (5000r) cross which represented the smallest sample (Table 2). Table 4 presents the data from Table 2 grouped with regards to the homologues carried by the male parent (Section I) and with regards to the maternal genotype (Section II) rather than the 1§:2_ratio. Examination of this table re— veals that no one homologue is favored or selected against with any degree of regularity between crosses. For example, the lethal — 9 chromosome is recovered among the progeny with a frequency of .4751 when balanced with Elpm_and mated to a M25 female (Table 4, Section I, Pm/le—9 X M55). When mated to the same type of female (M-S) but balanced with Curly, 43 TABLE 2 . . . X values for 2 x 2 contingency chi-square tests comparing X2 values in boxes 1 to 36 represent sex ratios; X2 values column one ending in (b) represent calculations d5 Pm/le-9 Cy/le—9 Cy/le-9 96 x 99 99 M-S (NS) OR (NS) OR (S) - (1) ,,$2) Pm/le-9 x M—5 11.22 .7.55_ (37) (9) Cy/le-9 x OR 1.95 (a) 0.69 (NS) 0.98 (b) <38) <39) Cy/le-9 x OR ”7:29 (a) ”6.98 (s) 6.57 (b) * _(40) (41) (42) Cy/Pm x OR 78.47 (a) 11.40 0.10 (NS) 7.24;(b). ‘ (43) (44) (45) Cy/Pm x OR (37.10 (a) 15.99 0.75 (S) 12.69 (b) a “147) (48) (49) Pm/le-9 x OR .12297 (a) 15.05 3.16 6.50'(b) - _(52> . (53) .(54) Cy/le-9 x M-5 T23.56”(a) 35.28 14.44 (2500r) ‘25.47 (b) "‘ (58) ‘ (59) (60) Cy/le-9 x M-5 6.35 (a) 4.40 1.97 (5000r) 2.12 (b) .*(65) _, _(66) (67) Cy/le-9 X M-5 14.13 (a) 12.61 _7:02 ,(Or) 4.91 (b)‘ 1 ' NS = Nonstored sperm; S = Stored sperm. 44 3 all the crosses listed in Table 2 for intercross variation. in boxes 37 to 72 represent le-9 ratios. Values listed in based only on numbers of females Cy/Pm Cy/Pm Pm/le-9 Cy/le-9 Cy/le—9 Cy/le-9 OR (NS) OR (S) OR (NS) M5 2500r M5.5000r M5 Or (3) (4) (5) (6) (7) (8) 14.50 8.48 29.82 0.0+ 1.94 0.42 (10) (11) (12) (13) (14) (15) 0.40 0.96 12.45 15.10 7.44 2.09 (16) (17) '(18) (19) (20) (21) 2.01 0.0+ 15.99 (9.88 7.58 1.18 (22) (23) (24) (25) (26) 3.10 10.91 19.69 8.14 2.78 (46) ,127) (28) (29) (30) 0.46 19.05 11.75 6.63 1.20 (50) (51) (31) _ (32) ((33) 3.14 4.80, 36.38 14.50 10.28 (55) (56) (50) (34) (35) 15.66 12.82 2.92 3.27 0.51 (61) (62) (63) (64) (36) 2.25 1.56 0.62 0.0+ 2.80 (68) (69) (70) (71) (72) ,5.57 '4.64 1.55 . 0.0+ 0.0+ I O 2 I O O I Boxes containing X values show1ng Significance are indicated as follows: green = 1%; blue = 2.5%; red = 5%. 45 TABLE 4 Matings of later crosses grouped with regards to male homologues (Section I) and maternal genotype (Section II) Section I Matings grouped with regards to male homologues dd 99 Treatment le-9 ratio Pm/le-9 X M-S NS .4751 Pm/le—9 X OR NS .5100 Cy/le-9 X OR NS .4872 Cy/le-9 X OR 8 .4991 Cy/le-9 X M-5 Or .5239 Cy/le-9 X M-S 2500r .5223 Cy/le-9 X M-S 5000r .5237 Section II Matings grouped with regards to maternal genotype dd 99 Treatment le—9 ratio Pm/le-9 X M-S NS .4751 Cy/le-9 x M-5 Or .5239 Cy/le-9 X M-5 2500r .5223 Cy/le-9 X M-5 5000r .5237 Cy/le-9 x OR NS .4872 Cy/le-9 X OR 8 .4991 Cy/Pm X OR NS .4999 Cy/Pm x QR S .5021 Pm/le-9 X OR NS .5100 46 the l§;2_ratio of the progeny is .5239 (Table 4, Section I, Cygle-9 X M;§J OR). In a similar way, the l§:g_ratio re- flects a differential recovery for the lethal — 9 chromosome when the genotype of the mother is changed; .5100 and .4751 for OR and M:§_females respectively when crossed to Pm/le-9 males (Table 4, Section I). The irradiation of the Cygle-9 males did not seem to influence the qualitative recovery of the autosomes as indi- cated by the 13:9 ratio. Closer analysis of the data seems to contradict this initial impression when the 5000r group is compared to the 2500r and Or groups for intercross vari- ation. For four comparisons (Table 3, Boxes 58b, 60, 61, 62) significance is not achieved as measured by 2 X 2 contingency X2 tests. Both of the other groups of the irradiation cross show significance for the same comparisons (Table 3, Boxes 52b, 54, 55, 56 and 65b, 67, 68, 69). An explanation for this is suggested by the observation that within the 5000r group the recoveries of the two categories, gy_and lg:2_are not significantly different from one another (X2 = 1.900). Similar tests for deviation from 1:1 recovery show the X values for the Or and 2500r groups to be significant at the 1% and 2.5% levels respectively. The small sample size of the 5000r group appears to be the reason for the lack of 47 significance since the le;9_ratio is the same as recorded for the other two groups. These later experiments indicate that the differential recovery of the various homologues depends on the genotype of the mother and the combination of homologues that the parental males carry. The differences appear within a given cross as differential recovery of the homologues and between crosses as differential recovery of alternative homologues when different females or stored sperm are used. While these experiments may serve to demonstrate a dif- ferential recovery among classes of progeny it must be realized that they do not in themselves constitute a critical analysis of the problem. The enumeration and subsequent categorization of progeny is possible only after a number of developmental stages. In order to attribute the observed results to differential sperm mortality it is first neces- sary to consider other causes of aberrant progeny ratios and effectively eliminate them as an influence on the observed results. Tests for Mortalipy While meiotic events may be considered normal without need for experimentation (see Discussion), it was necessary 48 to test for postfertilization phenomena. Such phenomena, which could contribute to aberrant ratios, would include egg hatch and larval viability. Temperature Experiment The first experiment to test for developmental mortality (Table 5) did not indicate a reduction of viability between the progeny reared at 180C. and those reared at 230C. A re- duction of progeny might occur at the lower temperature since development will be slower and any intrinsic weaknesses at critical stages in the developing larvae would be pro- longed. There was no significant difference between the categories of progeny within each of the temperature groups. While no reduction of viability after slower development was demonstrated, this experiment could not provide the infor- mation needed to properly evaluate developmental mortality. The experimental design provided no way of knowing how many individuals might have developed in each group. No enumer- ation had been made of the eggs which could have potentially developed. Egg and Larval Mortality Experiments The second mortality experiment was designed to correct the faults of the temperature experiment for mortality. 49 TABLE 5 Progeny counts from temperature mortality experiments Curly Lethal—9 66 dd 99 Total* 18°C. 398 476 384 537 1795 23°C. 354 467 368 549 1738 752 943 752 1086 3533 *Based on progeny from 23 females. TABLE 6 Results of egg and larval mortality experiments mating cgggt Hgizh £3613: figgort::::Z1 M-5 66 x M—5 99 831 811 737 2.5 9.2 08 66 x OR 99 1243 1222 1158 1.7 5.3 Cy/Pm 96 x Cy/Pm 99 718 682 309 5.0 54.6 Cy/le-9 69 x Cy/le—9 99 699 678 363 3.0 46.4 Pm/le-9 96 x Pm/le—9 99 998 939 326 5.9 65.2 Pm/le—9 96 x OR 99 884 837 777 5.4 7.2 Cy/le-9 96 x OR 99 993 974 935 2.0 4.1 Cy/Pm 69 x OR 99 682 561 504 17.8 10.2 Pm/le-9 96 x M-5 99 558V 442 345 20.8 22.0 Cy/le-9 96 x M-5 99 936 901 836 3.8 7.3 50 From the tabulated data (Table 6) egg and larval mortalities were determined. The mortality figures for the OR X OR and M;§_X M:§_crosses were used to represent the "background" genetic mortality for the OR and M;§_females of the other crosses. When the background mortality is subtracted from the particular mortality figures for a given cross the re- mainder is referred to as the "residual" mortality--peculiar to the given cross. The residual mortality is used to bias the data by assuming all of this mortality has been con- tributed by the smallest observed category of progeny. By multiplying the percentage of residual mortality times the total progeny of a cross, the percentage may be transformed to presumptive progeny which are assumed (for purposes of intentional bias) to have belonged to the smallest progeny class. For example, the Cyzle-9 95 X M:§_99 cross showed an egg mortality of 3.8% and larval mortality of 7.3% (Table 6). Subtracting the background egg and larval mortality figures of the M:§_X M:§_cross (2.5% and 9.2% respectively) gives a residual mortality of 1.3%. If the value for the residual mortality is negative, it is not included in the final value (3.8% - 2.5% = 1.3% and 7.3% - 9.2% = -l.9%, but the residual mortality is computed as 1.3%). The 1.3% residual mortality is multiplied by the total progeny recovered from the cross 51 (Table 7). The additional progeny (123) are added to the smallest progeny class (Curly) and to the total progeny value. In this way the original deviation would be corrected to some value closer to the 1:1 ratio which would be expected from genetic.theory. The residual mortality percentages for most of the crosses, when corrected for background mortality, are too high to effectively rule out post fertilization mortality as the basis for the observed autosomal deviations. It should be realized that the sample size used to determine egg and larval mortality were generally very small in com— parison to the total number of progeny that had been counted in the ratio experiments. It is also unlikely that all of the residual mortality would fall into the smallest progeny category. In spite of these restrictions two categories were able to accept the bias of residual mortality and still remain significant (Table 7). The crosses involved consist of the same male but different females. Both crosses had significant deviations from the expected 1:1 ratio and re- tained this deviation after the bias had been introduced. The difference between the two crosses is also extreme as indicated by the 2 X 2 contingency chi-square test (X2 = 20.84). 52 m®.va mmmv. QOMmN Hmmqa mmmfia em.o Hem.mHv Hmeme.v HmHmmmc HHmmeHv HemmeHi mo x man\>o mm.m emHm. meme mmee mmme em.H Ama.mHi Hmmmm.v Hmmemv HmHmeV Hemmev mu: x mumH\so muflamuuoz ms m> oflumm m o a H: o no HmseHmmm x H ex aumH H u e H o m H H H mmmHU oflmhuosmflm ummHHmEm ecu Ou excomoum 0>HumEsmmHQ: mo cofluflppm may nmsounu moflumu hammoum o3u mo :cofluomHHOO= h mqmdfi 53 These differences appear to justify the conclusions drawn from the ratio experiments. A given chromosome may possess some advantage over its homologue as measured by progeny counts. This advantage would be dependent upon the combination of homologues carried by the parental male and would be affected by the maternal environment. Sex-Linked Lethal Experiments The results of these experiments may be divided into the association of newly induced sex-linked lethals with auto- somal markers (Table 7) and into comparisons between the F1 and F2 progeny (Table 9). The rate of spontaneous mutants for both the Cygle-9 and OR males was low but not too low for Drosophila as a whole (Demerec, 1937) or for the particular stock (Yanders and Seaton, 1962). The induced mutation rate is lower than might be expected based on an expectation of 2.89% mutants per 1,000r (Lea, 1962). With this mutation rate, 5000r and 2500r should have produced 14.45% and 7.225% sex-linked recessive lethals respectively. The lowered mutation rate observed in the experiments may be a reflection of the num- ber of sterile and non—test Fl females (Table 8). This category was especially high for both the Cygle-9 and OR 54 mm.MH we.mH mm.HH em.e ma.o MH.e «mm. .......... mHmnumH 9 He em «a mmm oeH mHH m m H mHmHumH emerHuxmm omm NON mmH whom msmH womH semH .......... Hmehoz mm 6 mm OHNH OHA com mm uuuuuuuuuu ummuco: 6cm mHHumum ooem. mmmm. 66mm. meem. emmm. 96mm. memm. emmm. emmm. oHumu xmm mm.mH mm.m em.e ee.me 9H.mm em.om sm.mmH om.mm em.me a\scmmohe mmmumem New Hee Hee mmem mmme mHme ommm emmH vaH Hmuoe owe mew mmm HmHm mmem mmvm meMH was Hem ea mom 66H 66H Nome memm meow HmHH mmo mem 66 "Sawmoum Hm Hmuoe mlwq >HHSU HmDOB mlwq >HHDU Hmuoe mlwq hausu Hooom Hoomm HO Add ml: 66 mImH\>OV Spawn Hmnuma onsflalxmm man no mmsoum amusoefluwmxm on» NO somflHmmEou m mqm6H x6.m 0:» um unmonHamHm44 .H0>0H uHm>mH *6 one um 6660H6H66H64 Hm.o me . He.ms 66 u 66 66666 .66 mo 60 summons .Hm H6.o me . Hm.ms 66 u 66 Hoomm .66 mo mo summons .om 444mm.- we I Hm.ms we I 66 Ho .66 mo 60 summons .mH 4466.6 mm . Hs.ms as u 66 66666 .66 mumH\so mo summons .mH 44466.66 me . Hm.MI we I 66 “comm .66 mu6H\so mo summons .sH 444Hs.m me u Hm.ms 66 n 66 Ho .66 mu0H\so mo scomoum .6H mm 6cm Hm smoBDon mCOmflnmmEou 66.~ mesons 6666.6 .uoom.m .no.ms as u 66 66 mn6H\so no He .mH +6.6 666096 66666 .HOOmm .Ho.ms muwH u so 66 mu0H\so 60 He .6H mm.o HmnumH u Hmeuoz.ms Houucoo u HmucoeHuomxm uooom .mH 4mH.6 HanumH . Hmeuoz.ms Houucoo . HmucmsHumexm Hoomm .NH 66.H HanumH . Hmeuoz.ms mumH . so 66666 .66 6-6H\so .HH 66.H HanuoH u Hmeuoz.ms mumH u so Hoomm .66 mu6H\so .oH mm.~ Houucoo ( HmucmsHumexm.ms 66 u 66 66666 .mm .m 6m.H Houucoo u HmucmeHuomxm.ms 66 n 66 66666 .H6 .6 44466.AH Houucoo . HmucmeHumexm.ms 66 u 66 “comm .me .s HH.o Houucoo . HmucmeHumexm.ms as I 66 Hoomm .HH .6 HH.G Houusou I amusweflummxm.mw do 1 66 HO .mm .m m6.~ Homecoo . HmucmeHumexm.ms 66 u 66 60 .H6 .6 66.m mumH I so.ms ea . 66 uoomm .66 6:6H\so no He .m 66.6 mumH . so.ms 66 I 66 66666 .66 mumeso 60 H6 .m 6m.m meH . so.ms 66 u 66 no .66 mumH\so 60 He .H msam> mx pwumwu msflmn msoum sflnufl3 mumuomumnu pmumwu mason msouw mm 6:6 as casuw3 mcomflumasou spsum Hmnuwa pwxcHHIXmm may no mcoaumumsmm mm can an cmo3uon 626 cfinufl3 0665 msomwnmmeoo How mosam> 6cm mums» X m magma. N Progeny counts for the F1 and F2 generations of the sex-linked lethal experiments 56 TABLE 10 Cy/le-9 99 x M-5 99 OR 96 x M-5 99 OR 2500r 5000r Or 2500r 5000r Number of P1 99 16 149 53 15 24 19 F1 Progeny: 99 1181 4302 362 996 942 154 99 1369 5151 480 1268 1148 175 Total 2550 9453 842 2264 2090 329 F1 Sex ratio .5368 .5449 .5770 .5600 .5492 .5319 Ave. progeny/ 9 159.3 63.4 15.8 150.9 87.0 17.3 Number of F1 99 sampled 160 157 38 137 35 16 Corrected sample* 160 149 34 137 33 15 F2 Progeny: 99 9464 9314 1889 8884 1895 965 99 9686 9693 2226 9158 2283 1045 Total 19150 19007 4115 18042 4178 2010 Corrected F2 Progeny*: 99 9464 9143 1800 8884 1846 924 99 9686 9311 1990 9158 2174 937 Total 19150 18454 3790 18042 4020 1861 Ave. progeny/ 9 119.6 121.0 108.2 131.6 119.3 125.6 Corrected Ave. progeny/ 9* 119.6 123.8 111.4 131.6 121.8 124.0 F2 Sex ratio .5057 .5099 .5409 .5075 .5464 .5199 Corrected F2 Sex ratio* .5057 .5045 .5250 .5075 .5407 .5034 * F2 totals excluding progeny of sex-linked lethal bearing Fl females. 57 groups at 2500r. As mentioned previously, high humidity and temperature in all probability contributed to producing a high rate of mortality when these females were etherized for counting. In all the irradiated classes it may be observed that the lower the mortality rate of the F1 99, the closer the induced mutation rate approached the expected value. The association of sex-linked lethals with autosomal markers had no meaning for the Or group since only a few lethals were recovered. Neither the 2500r nor 5000r group exhibited heterogeneity when tested by 2 X 2 contingency chi—square tests (Table 10, tests 16,17). These tests com- pared the recovery of normal and sex-linked 1ethal-bearing chromosomes in relation to their association with the Curly_ and lethal - 9 autosomes. When comparing progeny of the Cygle-9 males to those of the OR control males, the 2500r group indicated a difference significant at the 5% level (Table 10, test 18). The major contribution to the X2 value comes from the differential between the observed number of lethals (40) and the expected number (53). For the control group, as noted above, the number of F1 females unable to be tested was quite high for this test, possibly causing the deviation. 58 Although chi-square tests could not demonstrate a greater association of sex—linked lethals with either of the autosomes, both the 2500r and 5000r groups showed higher percentages of sex-linked lethals associated with the lethal - 9 autosome. This similarity, while not signifi- cant in itself, may suggest a trend towards a more definite association. Unfortunately for the present data, the 5000r group constitutes too small a sample from which to draw con— clusions and the 2500r group is undoubtedly prejudiced by F1 mortality. Comparisons were made within as well as between the F1 and F2 progeny by means of 2 X 2 and 2 X 3 contingency chi— square tests (Table 10) for all three tests, for both the control and experimental groups. Five of these comparisons were significant. Three tests (Table 10, tests 4, 5, 6) involved differences in the sex ratio between the F1 and F2 generations for the Cygle-9 (experimental) crosses. ’The non-irradiated group of the OR (control) cross showed a similar difference (Table 10, test 7). The major difference between the two genera- tions is a drop in the number of females recovered among the F2 progeny (Table 9). The relatively high sex ratios among the 2500r and 5000r groups of the F1 are somewhat unexpected 59 in that any dominant lethals induced in the irradiated males would promote a reduction rather than an increase in the number of F1 females. A sex ratio difference was also detected in the F2 genera— tion when the control and experimental crosses were compared for the 2500r group (Table 10, test 13). The difference here may be attributed to a higher frequency of female prog- eny in the control group. The results obtained for the association of sex—linked lethals with autosomal lethals suggests a possible relation- ship that cannot be critically tested with the available data. The suggestion is that a lethal - 9 chromosome is more likely to be associated with a newly induced sex—linked lethal than is a Curly chromosome. Comparisons between F1 and F2 generations reveal differences in the sex ratios. The cause of the difference stems from a higher recovery of females in the F1 generation. This higher recovery is unusual in that it is contrary to the expected genetic results for dominant lethals induced on the X-chromosomes. DISCUSSION Ratio Experiments Among the problems of studying sperm viability or lethal- ity is that the phenomenon cannot be observed directly. Other possible factors, operating at meiosis or after fertilization, have to be considered and satisfactorily eliminated in order to invoke, by elimination of other alternatives, selection against particular types of sperm. In the present study, meiotic anomalies may be disregarded as possible causes of the off-ratios which were observed, for off-ratios occurring during the meiotic divisions have been found only when there are homologues of unequal sizes (Novit- ski, 1951; Novitski and Sandler, 1957), or when a specific gene is involved (Gershenson, 1928; Sturtevant and Dobzhan- sky, 1936; Novitski, 1947; L. Sandler, Y. Hiraizumi and I. Sandler, 1959). The homologues in the present study are of equal size and there is not a mutant involved which would pro- duce abnormalities similar to those observed in the other studies cited. Peacock and Erickson (1965) have demonstrated that even sperm from wild-type males may be non—functional, indicating that the chromosomal or genic factors do not neces- sarily have to be present. The mating procedures used in the present experiments assured that only "functional" sperm would 60 61 be stored. The stocks which were used for these experiments were maintained in this laboratory for several years; during this time they were utilized in other experiments, none of which indicated any meiotic selection. Zygotic selection could be studied directly by egg count - progeny count experiments. Without trying to ascertain spe— cific causes for non-hatch or larval deaths, it was still possible to get an estimate of the mortality which did occur after fertilization. The mortality in most instances was too great to rule out zygotic events as a cause of the off- ratios in all but two crosses; but these crosses which still remained significantly different from one another (Table 7) give support to the case for differential gamete viability. Postulating that the sperm will be acted upon in a dif- ferential manner while stored in the female poses the ques— tion of how this selection might act. DeVries (1962, 1964) demonstrated that the quantity of sperm stored by a female varied with the male-female combination used. The rate of loss during storage of the sperm showed differences among males of different strains mated to females of the same strain. The differences in quantity of sperm stored and the rate of loss of the stored sperm were suggested to result from a 62 differential response by the sperm to the female reproduc- tive tract. This response would result in some sperm types being lost as a consequence of having a lower viability than the favored sperm. Moreover, Lefevre and Jonsson (1962b) have suggested wild-type sperm are better able to maintain them- selves in the storage organs than are mutant sperm. The mechanism of the loss that they suggest is related to a postu- lated migration of sperm in and out of the storage organs and from one storage organ to another. The sperm possessing the advantage would be better able to find its way back to the storage organ. It is hard to imagine how this mechanism would operate in an actively laying female, since some pro- portion of sperm would be lost during migration by being pushed out of the genital chamber along with the egg, and the efficiency of sperm utilization under these conditions should not be the nearly perfect value reported (Lefevre and Jonsson, 1962b). Regardless of the mechanism, these investi- gators demonstrated a greater recovery of wild-type than mutant sperm after double matings. In the present study, the variable recovery is found within the sperm from one male rather than between sperm from different strains or from sperm produced by two genotypically 63 different males. It is postulated that an influence is exerted by the female reproductive tract on the two sperm types carry— ing the two different autosomes. One of these sperm types, in a particular female, may be selected against. The same sperm type in another female of a different genotype may be favored. The direction in which the selection will go is probably determined by the female involved. A similar situa- tion has been found to prevail with several Tfalleles in the house mouse (Braden and Weiler, 1964). The magnitude of the effect which the Tfallele has on the sperm can be influenced by relatively small changes in the physiological state of the female tract. It is not meant to be implied that a similar gene is functioning in the present work to produce the same effect found with the Tfalleles in the mouse, but rather that the female reproductive tract may serve to initiate a selec- tive process. Sperm inactivation following exposure to X rays has been suggested by some investigators (Yanders, 1959, 1964; Linds- t al,, 1963). It seems unlikely that differential — ley, sperm inactivation due to irradiation occurred in the present experiments since the le-9 ratios for both the irradiated and control groups were virtually identical. While the sex ratio 64 varied between the control and experimental groups with each level of irradiation used, the direction taken by the sex ratio for each group was different. It would seem from this that no common element would be acting to effect the change. Sex—Linked Lethal Experiments The data from the sex-linked lethal experiments indi- cated a tendency for the newly induced recessive lethal mutants to be associated with the 1§:2_chromosome. This could not be a result of an effect of X rays on the meiotic divi- sions since the sperm utilized for the experiments were mature at the time of treatment and the males were mated shortly after irradiation. This tendency may be superficial in that a greater number of 1§;2_chromosomes were recovered than Curly ones for this experiment. The sex ratio and 1§;2_ ratio appear to be independent of one another, suggesting that whatever selective force is acting would work against sperm bearing a single factor (for this cross the Curly chromosome) rather than a particular combination of chromo- somes. The sex ratio differences between the F1 and F2 are quite uniform for all experiments. All of the sex ratios show a 65 greater recovery of females than males; the F1 exhibits this to a greater extent than does the F2. This is not entirely unexpected as £5 melanoqaster customarily produce more females than males (Hanks, 1964). What is somewhat unexpected is that the F1 should demonstrate such a high sex ratio -- dominant lethals induced in the X chromosome of irradiated sperm will be transmitted to the F1 females and shauld act to reduce the number of females. It should be noted though, that sex ratio shifts following irradiation are not always observed (Lindsley §§_§1,, 1963). The increase in the F1 sex ratio may be a heterotic effect similar to that noted for egg hatchability when two inbred strains are crossbred (Johansen, 1963). This effect would be lost by the F2 generation as the sex ratio would return to a lower value. The data obtained from these experiments support the hy- pothesis that a form of selection operated within the female reproductive tract in such a way as to lessen the number of Sperm effecting fertilization within some sperm classes. The class of sperm selected against depends on the combination of homologues carried by the male parent as well as the genotype of the female parent. A tendency for newly induced sex-linked lethals to be associated with a particular homologue was also noted. SUMMARY The experiments were designed to test for selection among sperm from male Drosophila heterozygous for second chromosome markers. Males bearing various combinations of second chromo- some mutants were mated to OR (wild-type) or §a§g_females; progeny from these crosses were scored with reference to their sex and autosomal markers. Irradiation studies were also carried out to test for a non-random association of newly induced sex-linked recessive lethals with the mutant second chromosomes. Deviations from 1:1 ratios were found among the re- covered progeny. Some of these deviations remained signifi- cant after zygotic mortality had been accounted for. The cause of the deviations is postulated to be an influence on the part of the female acting upon the stored sperm to ex- pell less viable sperm from the storage organs. The type and magnitude of the influence was dependent on the combina— tion of homologues carried by the parental male and by the genotype of the female in which the sperm were stored. Results from the irradiation experiments indicated a greater tendency for the sex-linked lethals to be associated 66 67 with a particular chromosome (1e—9). This tendency is postulated to result from selection against the single homologue, Curly, rather than for a combination of non- homologous chromosomes. BIBLIOGRAPHY Bateman, A. J., 1962. The genetics of egg hatching in Drosophila. Heredity 17:107-113. Bateman, Nigel, 1960a. High frequency of a lethal gene (t3) in a laboratory stock of mice. Genetical Research 1 (2):214-225. , 1960b. Selective fertilization at the T locus of the mouse. Genetical Research 1 (2):226-238. Bell, A. E., 1954. A gene in Drosophila melanoqaster that produces all male progeny. Genetics 39:958-959. Braden, A. W. H. and H. Weiler, 1964. Transmission ratios at the T-locus in the mouse: Inter- and intra-male heterogeneity. Aust. g;_Biol. Sci. 17:921-934. Carpenter, J. M., 1950. A new semi-synthetic food medium for Drosophila. Drosophila Information Service 24:96. Cooper, K., 1950. Normal spermatogenesis in Drosophila. In, Biology 9; Drosophila. M. Demerec (ed.). John Wiley and Sons, New York. pp. 1—62. Counce, S., 1956a. Studies on female sterility genes in Drosophila melanoqaster. I. Effects of deep orange on embryonic development. g;_indukt. Abstamm- EL_Verer- bunqslehre 87:443-461. , 1956b. Studies on female sterility genes in Drosophila melanoqaster. 11. Effects of fused on em— bryonic development. g;_indukt. Abstamm- u;_Vererbungs- 1ehre 87:462-481. , 1956c. Studies on female sterility genes in grosophila melanoqaster. III. Effects of rudimentary on on embryonic development. g;_indukt. Abstamm- u;_Ver- ggbungslehre 87:482-492. Demerec, M., 1937. Frequency of spontaneous mutations in cer— tain stocks of Drosophila melanoanter. Genetics 22:469-478. 68 69 DeVries, J. K., 1962. Studies on insemination and sperm storage in intra— and inter-strain crosses of Drosophila melanogaster. Master's thesis, Michigan State University. , 1964. Insemination and sperm storage in Drosophila melanoqaster. Evolution 18(2):271-282. Frydenberg, Ove and K. Sick, 1960. Desemination by cold shocks. Drosophila Information Service 34:79. Gershenson, S., 1928. A new sex ratio abnormality in Drosophila obscura. Genetics 13:488—507. Glass, H. B. and H. L. Plaine, 1950. The immediate dependence of the action of a specific gene in Drosophila melanogaster upon fertilization. Proc. Nat'l Acad. Sci. (Wash.) 36:627— 634. Gowen, J. W. and L. E. Johnson, 1946. On the mechanism of heterosis. I. Metabolic capacity of different races of Drosophila melanoqaster for egg production. Amer. Nat. 80:149-179. Grell, R. F., 1959. Non-random assortment of non-homologous chromosomes in Drosophila melanogaster. Genetics 44:421- 436. Gugler, H. D., W. D. Kaplan and K. Kidd, 1965. The displace- ment of first mating by second mating sperm in the storage organs of the female. Drosophila Information Service 40:65. Guyenot, E., 1913. Etudes biologiques sur une mouche, Dro- sophila ampelophila Low. VI. Resorption des spermatozoides et avortement des oeufs. Q; R; Soc. Biol. (Paris) 74:389- 391. Hadorn, E., 1961. Developmental_Genetics and Lethal Factors. John Wiley and Sons. New York. Hanks, G. D., 1964. Are off-ratios generally due to abnormal segregation? Genetics 50:253. , in press. Are normal off sex ratios often caused by aberrant segregation? Genetics. 7O Hildreth, P. E. and G. L. Carson, 1957. Influence of the type of inseminated female on the lethal frequency in the X chromosome from the male. Proc. Nat'l Acad. Sci. (Wash.) 43:175-183. ' and J. C. Lucchesi, 1963. Fertilization in Dro— sophila. I. Evidence for the regular occurrence of monospermy. Developmental Biol. 6:262-278. Johansen, I., 1963. A wildtype cross with a high yield of hatching. Drosophila Information Service 38:54. Kaplan, W. D., V. E. Tinderholt and D. H. Gugler, 1962. The number of sperm present in the reproductive tracts of Drosophila melanoqaster females. Drosophila Information Service 36:82. Kaufman, W. D. and M. Demerec, 1942. Utilization of sperm by the female Drosophila melanoqaster. Amer. Nat. 76:445—469. Lea, D. E., 1962. Actions 9f_Radiations Qg_Living Cells. Second Edition. Cambridge University Press, Cambridge. 416 pages. Lefevre, G., Jr., and U. B. Jonsson, 1962a. The effect of cold shock on Drosophila melanoqaster. Drosophila Infor- mation Service 36:86. , 1962b. Sperm transfer, storage, displacement and utilization in Drosophila melanoqaste§,_ Genetics 47:1719- 1736. Lindsley, D. L., C. W. Edington and E. S. Von Halle, 1963. The effect of gametic genotype on the radiation sensitivity of Drosophila sperm. In: Repair from Genetic Radiation Damage and Differential Radiosensitivity in Germ Cells. F. H. Sobels (ed.) New York. Pergamon Press. pp. 63-76. Lobashov, M. E., 1939. Mixture of sperm in case of polandry in Drosophila melanogaster. .E;.B; (Dokl.) g§_l'Acad. Sci. URSS 23:827-830. Lynch, C. J., 1919. An analysis of certain cases of intra- specific sterility. Genetics 4:501—533. 71 Manning, A., 1962. A sperm factor affecting the receptivity of Drosophila melanoqaster females. Nature, London 194: 252—253. Metz, C. 8., 1955. Specific egg and sperm substances and ac— tivation of the egg. In: The Beginnings of Embryonic De- velopment. A. Tyler, R. C. von Borstel and C. B. Metz (ed.). The AAAS Publication no. 48, Wash., D.C. Metz, C. W., 1929. Evidence that "unisexual" progenies in Sciara are due to selective elimination of gametes (sperms). Amer. Nat. 63:214-228. Meyer, H. U., 1954. Influence of maternal genotype on the hatchability of fertilized eggs in Drosophila melanoqaster. Drosophila InformatiogService 28:134-135. Miller, A., 1950. The Internal Anatomy and Histology of the Imago of Drosophila melanogaster. In, Biology 9§_Drosophila, M. Demerec (ed.) John Wiley and Sons, New York. pp. 420-534. Morgan, T. H., 1911. An alteration of the sex ratio induced by hybridization. Proc. Soc. Exp. Biol. 8:83-93. Muller, H. J., 1944. Failure of desemination by nitrogen. Drosophila Information Service. 18:56. , and F. Settles, 1927. The non-functioning of the genes in spermatozoa. gy_indukt. Abstamm- Ey_Vererbungs- 1ehre 43:285-312. Myszewski, M. E., 1962. The Effects of Storage Upon the Differ— ential Survival Among Spermatozoa of Drosophila melanoqaster. Master's Thesis, Michigan State University. Novitski, E., 1947. Genetic analysis of an anomalous sex ratio condition in Drosophila affinis. Genetics 32:526-534. , 1951. Non-random disjunction in Drosophila. Genetics 36:267-280. , 1954. The compound X chromosome in Drosophila. Genetics 39:127-140. 72 .1 W. J. Peacock, and J. Engel, 1965. Cytological basis of "sex ratio" in Qy_pseudoobscura. Science 148: 516-517. , and G. Rush, 1948. Desemination by low-tempera- ture shocks. Drosophila Information Service 22:75. , and I. Sandler, 1957. Are all products of sperma- togenesis regularly functional? Proc. Nat'l Acad. Sci. (Wash.) 43:318-324. , and L. Sandler, 1956. Further notes on the nature of non-random disjunction in Drosophila melanoqaster. Genet- ics 41:194-206. Offerman, C. A. and I. K. Schmidt, 1936. Culture media for Drosophila. ,Drosophila Information Service 6:64. Oliver, C. P. and R. C. Anderson, 1942. Infertility due to sperm non-motility in a female of Drosophila melanoqaster. Genetics 28:85 (Abstr.) Patterson, J. T. and W. S. Stone, 1952. Evolution in the Genus Drosophila. The Macmillan Company. New York. pp. 359-383. Peacock, W. J. and J. Erickson, 1965. Segregation-distortion and regularly non-functional products of spermatogenesis in Drosophila melanogaster. Genetics 51:313-328. Poulson, D. F. and B. Sakaguchi, 1961. Nature of "sex ratio" agent in Drosophila. Science 133:1489-1490. Redfield, Helen, 1957. Egg mortality and interchromosomal effects on recombination. Genetics 42:712-728. Russell, L. B. and W. L. Russell, 1960. Genetic analysis of induced deletions and of spontaneous non-disjunction in- volving chromosome 2 of the mouse. J;_Cell. Comp. Physiol. 56 (supp1.):l69-188. Salisbury, G. W. and J. R. Lodge, 1962. Metabolism of sperma— tozoa. Adv. Enzym. 24:35-104. 73 Sandler, L., Y. Hiraizumi and I. Sandler, 1959. Meiotic drive in natural populations of Drosophila melanoqaster. I. The cytogenetical basis of segregation-distortion. Genetics 44:233-250. , and E. Novitski, 1956. Evidence for genetic homology between chromosomes I and IV in Drosophila melanogaster with a proposed explanation for the crowding effect in triploids. Genetics 41:189-193. , 1957. Meiotic drive as an evolutionary force. Amer. Nat. 91:105-110. Scossiroli, R. E., 1954. Desemination of fertilized Drosophila melanoqaster females. Physiol. 2001. 27:157. Smith, J. M., 1956. Fertility, mating behavior and sexual selection in Drosophila subobscura. gg_Genetics 54:261- 279. Sonnenblick, B. P., 1950. The Early Embryology of Drosophila melanogaster. In, Biology g§_Drosophi1a. M. Demerec, (ed.). John Wiley and Sons, New York. pp. 62-167. Sturtevant, A. H., 1945. A gene in Drosophila melanogaster that transforms females into males. Genetics 30:297-299. , and Th. Dobzhansky, 1936. Geographical distribu— tion and cytology of "sex ratio" in Drosophila pseudo- obscura and related species. Genetics 21:473-490. Yanders, A. F., 1959. The effect of X-rays on sperm activity in Drosophila. Genetics 44:545-546. , 1963. The rate of sperm migration in inter- and intra-strain matings. Drosophila Information Service , 1964. The effects of X rays on sperm activity in Drosophila, Genetics 49:309-317. , 1965. A relationship between sex ratio and paternal age in Drosophila, Genetics 51:481-486. 74 Yanders, A. F., and R. Kirk Seaton, 1962. The lack of mutagenicity of caffeine in Drosophila. Amer. Nat. 96:277-280. RIES “‘11‘111111111711111111111]’