Genetic Studies in Drosophila Pseudoobscura Citation Mampell, Klaus (1943) Genetic Studies in Drosophila Pseudoobscura. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/EQ1V-FY29. https://resolver.caltech.edu/CaltechTHESIS:12142016-110541139 Abstract No abstract. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Biology) Degree Grantor: California Institute of Technology Division: Biology Major Option: Biology Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: Unknown, Unknown Defense Date: 1 January 1943 Record Number: CaltechTHESIS:12142016-110541139 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:12142016-110541139 DOI: 10.7907/EQ1V-FY29 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 9994 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 14 Dec 2016 23:29 Last Modified: 10 Nov 2023 17:38 Thesis Files Preview PDF - Final Version See Usage Policy. 16MB Repository Staff Only: item control page

GENETIC STUDIES IN DROSOPHILA PSEUDOOBSCURA Thesis by Klaus Mampell In partial fulfillment of the requirements for the degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1943 FOREWORD Drosophila is one of the preferred organisms for geneticists. D· melanogaster is the most thorough l y investigated species of the genus. Research with this species for over thirty years has resulted in the establishment o·f a remarkable variety of stocks; this greatly .facilitates the work o.f any geneticist who now chooses this species for his researches. exist for most species of Drosophila. No such stocks It is, of course, more difficult to work with a species where all necessary stocks have to be built up and maintained by the investi= gator himself· This involves time, labor and luck, and even so, the stocks remain incomplete at best. Most of the D· pseudoobscura stocks used in this work were made up in the course of the research. The choice of the species was accidental. The laboratory where the author .first began Drosophila re= search had only D· pseudoobscura, race B, available• A chronological account of the work will show how each part of the research described in this thesis presented itself accidentally and unexpectedly. Thus we get a chain of problems whose only connection is the appearance of one i problem in the course of solving another. It is enter= taining to begin this chronological account with the day when the author began Drosophila research. a mutation was found: short. On that day shortest, an extreme allele of In the course of mapping this gene new mutant genes were found, and mapping those, others were found, etc. One of these new mutant genes was radius rudiment - ary, a sex~linked recessive showing in females only. This seemed analogous to the behavior of bobbed in D· melanogaster where a wild type allele in the Y-ehromosome prevents the appearance of the character in the males. To test for a like explanation in this ease, XO males had to be obtained; there, the Y=chromosome with its dominant allele would be removed, and the recessive character could express itself· It is known that D· pseudoobscura interracial hybrid females produce an increased proportion of XO males. Mating a radius rudimentary male to a hybrid female homozygous for a sex-linked recessive like vermilion should have resulted in vermilion sons and an occasional male wild type for eye color which would then have resulted from the union of an X-bearing sperm with a no-X-egg; this would have been the desired individual. n. pseudoobscura hybrid females ordinarily are fertile, but these vermilion hybrid females proved to be ii sterile• This unexpected phenomenon led to the study described in chapter I· While investigating this prob= lem, one of the race B cultures which was used for mark= ing the race B X-chromosome showed four new sex=linked recessives. II· This initiated the study described in chapter One of the wild type stocks to which the flies were outcrossed was the Seattle strain. Besides the more usual point mutations many mutations suggesting deficiencies began to appear. After a period of confusion about the nature of this phenomenon, it was finally recognized as being due to an independent cause, a gene that had been introduced from the ,' Seattle straine is described in chapter III· The study of this case Already a new phenomenon has appeared in the study of the last problem. It will furnish another chapter in the future. Besides the problems mentioned in the foregoing account, literally hundreds of others suggested themselves in the course of the different researches. Of those, many seemed interesting and profitable to be investigated. Obviously, one investigator can only do a limited amount of work, and it is up to his judgment to select out of the multiplicity of problems those which seem to him t he most interesting and profitable• Thus, the majority of problems iii has to be discarded in favor of a very few. This is regrettable and sometimes painful, but cannot be helped. The 11 Female sterility in interracial hybrids of D· pseudoobscura" was selected, for here we have unusually favorable material for the study of speciation. Race A and race B are still closely enough related to show no significant morphological differences, but the hybrid males are completely sterile. As is described in chapter I, the hybridization of some strains or the two races produces sterile females as well· This is the final step in the divergence or the species. No exchange of genes is possible any longer and the evolution of the two will progress independently. The sterility of the females is due to a beautifully complicated mechanism the analysis of which is interesting regardless of the evolutionary implications& The two other problems deal with mutation. The author's main interest lies in the concepts having to do with the nature of the gene. This is indeed a large field, but any problem dealing with the causes of gene mutation is apt to throw some light on the nature of the gene. The "high mutation frequency in D· pseudoobscura, race B" in chapter I I shows how one gene can control the mutation iv rate of all other genes. This is probably the mechan- ism controlling mutation under natural conditions, for irradiation of all kinds, temperature, etc. have been ruled out as significant natural factors. Therefore, the problem is interesting both for questions concerned with evolution and those concerned with the mechanism of gene mutat.ion. In chapter III "a deficiency inducer in D· pseudoobscura, race B" is described. Here mutations of a different nature are observed. All arguments in favor of studying the former problem can be applied to this one. In addition, an analysis and comparison of the two kinds of mutation, namelya point mutation versus defi- cieny, may show to what extent chemical or physical changes can account for either. The author is indebted to those members of the Biology Department who have contrib~ted suggestions and ... criticism concerned with various phases of these studies. v CHAPTER I Reprinted from the Proceedings of the NATIONAL ACADEMY OF SCIENCES, Vol. 27, No. 7, pp. 337-341. July, 1941 FEMALE STERILITY IN INTERRACIAL HYBRIDS OF DROSOPHILA PSEUDOOBSCURA Bv ' KLAUS MAMPELL WM. G. KERCKHOFF LABORATORIES, CALIFORNIA INSTITUTE OF TECHNOLOGY Communicated May 27, 1941 Male sterility is a rngular occurrence in the hybrids of race A and race B of D. pseudoobscura. 1 A case of female sterility in the same hybrids has been described recently. 2 The genetic analysis of this sterility will here be extended, and new data concerning the map of the X-chromosome will be given. The salivary chromosomes of the A B hybrid female show two inversions in the X-chromosome, one in the right, one in the left arm. The XR inversion has now been established genetically. In race A the genetical map shows the genes: compressed, sepia, ascute, javelin (slender) in the order given. 3 The X-chromosome of race B, based on the map already known 4 and on additional data, shows the order: compressed, ascute, sepia, javelin (table 1). A comparison of the maps shows that the breakpoints of the inversion lie to the right of compressed and to the left of javelin, respectively. Included in the map are some genes which seem to have no counterpart in either race A or D. melanogaster. Filiform is a good dominant viable in male and homozygous female; the macrochaetae are long and slender with tapering tips. The lethal plexus kills the males in the late pupal stage; when occasional males emerge their wings show a plexus-like effect; they are very weak and die soon after emergence. The mutant radius rudimentary is expressed in the female only. The suppression of the effect in the male is not due to the influence of the Y-chromosome (tested in XO males). The expression in the female is rather variable and is influenced by environmental conditions; in flies ·raised at a temperature of 15°C. very 338 GENETICS: K. MAMPELL PROC. N. A. S. marked plexus effects are added to the rudimentary condition of the radius; at 25°C. the character usually overlaps wild type. In race A, Gottschewski 2 has described two genes, lm 1 and lm2, which have a certain lethal action in an environment of hybrid cytoplasm. In race B there are two corresponding genes which have a similar action. The symbols for these genes, proposed here, are: lm 1A and lm2A for race A, lm 1B and lm2B for race B. Certain backcrosses of hybrid females were found to produce no or very few offspring.· An example of certain mass matings will illustrate this: 30 males of a wild race A strain (Boise I) were mated to 20 hybrid females with a vermilion sepia X-chromosome from the B father, an eosin vermilion Xchromosome from the A mother. Since the hybrid females were homozygous for vermilion, all normal male offspring should have carried that gene. The total offspring of these 20 females consisted of 47 wild-type TABLE 1 THE X-CHROMOSOME OF RACE B beaded scutellar yellow Notch white lm1B* singed vermilion Filiform lethal plexus dusky forked 0.0 24.9 24.9 31. 3 32 .7 34.5 34.6 34.8 35 .3 35.8 rose bubble fused compressed ascute scarlet lm2B* sepia radius rudimentary javelin short 38.9 53.3 54.0 55.7 76.1 105.8 128.2 133.0 138 .6 * The position of the interracial le,thals is uncertain. females and 15 wild-type males. These patroclinous males could not have inherited their X-chromosome from the mother, but must have resulted from the union of an X-bearing sperm and a no-X egg. This no-X egg can arise through primary non-disjunction or, as seems more likely, through crossing-over within the inversions, a four strand double crossover producing only dicentric and acentric X-chromosomes. Gottschewski pointed out that the lethals are not completely dominant in the backcross females; in many cases "rare survivors" appear. The 47 females from the above cros·s belong in that category. When the backcross males were we v (A) or v se (B), no offspring at all were produced; this is expected, since the XO males too, in this case, would inherit the lethals. In a third type of cross: a hybrid female free of the VOL. 27, 1941 GENETICS: K. MAMPELL 339 lethals, backcrossed to a male carrying them should produce males and no or few females; the dat~ bear out this expectation. From these typical examples it may be concluded that both the we v (A) X and the v se (B) X must have carried lethal genes which were completely innocuous intraracially; also, they did not affect the F1 hybrid generation where they would still be in an environment of race A or race B cytoplasm. It is the hybrid cytoplasm, then, which allows these genes to act as lethals. Two genes must be concerned in the production of the lethal effect as shown previously by Gottschewski. A race A fly which is known to carry the lethals 'a nd which has the proper marker genes in the X is mated to a race B fly which is known to be free of the lethals. The resulting hybrid female, when backcrossed, gives a majority of wild-type sons. Those sons carrying any race A marker genes prove to be the result of crossing-over with the race B X-chromosome. No sons appear which have intact the section between the white region and the XR inverted region. Those males survive in which the two regions have become separated by crossingover. Since crossovers in both directions survive, lm 1A in the white region and lm 2A in the XR inverted region must be together to produce the lethal effect; individually they no longer act as lethals. The reciprocal mating shows a corresponding situation in race B; lm 1B and lm 2B lie in the same respective regions and act in the same manner as the race A lethals. The maternal effect by which these interracial lethals act was investigated by the experiment shown in table 2. A race B wild-type male is mated to a race A female which carries in the X -chromosome an inversion across the spindle attachment region associated with a dominant effect, short scutellum (Ss); this X-chromosome is lethal to the male. The autosomes are marked as follows: chromosome II carries Bare (Ba) in an x-ray induced inversion; chromosome III carries Blade (Bl) in standard sequence; chromosome IV carries Curly (Cy) in an x-ray induced inversion; chromosome V cannot be marked. In the hybrid female crossing-over should be suppressed effectively and the identity of racial origin of each chromosome be preserved, since in addition to the inversions mentioned above, there are present two interracial inversions in X, one in II. Crossing-over is suppressed least effectively in III, since some crossing-over does occur between Standard and Klamath, the common race B sequence. Ss Ba Bl Cy hybrid females are backcrossed to we v males of race A; the we v strain carries the race A interracial lethals. From this .cross only few female offspring are obtained; they consist of the 16 different phenotypes given in table 2. Each phenotype individually is backcrossed to we v. All classes with the Ss marker are fertile; all classes without the Ss marker are sterile. All the fertile females are homozygous for the race A X chromosome; all sterile females are racially heterozygous for X. The race A interracial lethals act only in cytoplasm (eggs) produced by a female 1 Cy + + Bl + + Ba + + + Ss Ss Ba Bl Bl + + + =f = s =f = s =f = s =f = s =f + .= s + =f Cy = s + =f Cy = s + =f Cy = s Cy + Ss Ss + Ss + + Ss Ss + v dy (B) ( Ss + Ba + Ba + Ba Ba + + Ba ~ +(B) ~ v dy (B) Cf' X ++++ +(A) Ba Ba Ss Bl + + + + Ss + / +(A) Bl +(A) Ba w' v (A) / Bl + + Bl Bl Bl + + Bl +(B) Cy Cy = *(s) + = f(1 3) + =f(S) Cy = *(s) + = f(13) Cy = s(.5) + = *(j) Cy = s(3) +(B) + + + + + = s(O) = f(ll) Cy Ss Ss Cy = s(O) + + Ba Bl Bl Ba + + + + = *(j) Cy Ba Ss Ss Ba Cy = s(l) + + Bl Bl + + + + + + =f(6) Bl Ss Ss Bl + + + + = *(j) Cy Ba Cy = s(O) Ba + + + + In the 16 classes ·of backcross females all marked chromosomes are race A, all + chromosomes are race B . In thew' v backcross, s = sterile, f = fertile. In the v dy backcross, s = no or few v dy females produced, f = v dy females produced, * = class not obtained, symbol in parenthesis refers to expectation. Numbers in parenthesis indicate average v dy 9 offspring p er mother. w' v (A) Cf' X ( TABLE 2 +(B) Cf' x Ss Ba Bl Cy (A) 9 \. w• v (A) Cf' x Ss Ba Bl Cy 9 xv dy (B) Cf' <:Al tf>. r:n i"' ~ () 0 ~ t-. ~ ~ ~ ~ 0 :::i ~ g; o VOL. 27, 1941 GENETICS: IC MAMPELL 341 racially heterozygous for the X-chromosome. The autosomes have little or no influence. When Ss Ba Bl Cy hybrid females are backcrossed to v dy (B) males which carry the race B interracial lethals, a somewhat different picture presents itself. Of the 16 different phenotypes only eleven have been obtained; all of these produce some offspring. In deciding which chromosome combinations are responsible for the lethality of the lm1B lm2B complex the male offspring are neglected, first, because the viability is decreased by .certain Y-chromosomes, 5 an uncontrolled factor, and, second, because the results would be obscured by independent sex-linked lethals whose mutation rate is increased by hybridization.6 The number of v dy female offspring per mother is used for the determination of the action of the interracial lethals, since only this kind of progeny is comparable in all classes. In table 2 it can be seen that none of the classes with the Cy marker show as many v dy females per mother as any of the classes without the Cy marker. It seems, therefore, that the race B interracial lethals act in cytoplasm (eggs) produced by a female racially heterozygous for chromosome IV. It must be pointed out, however, that the results are not nearly as clear as those for the race A lethals; caution must be used in accepting chromosome IV as the only one responsible, until further evidence is obtained. With these interracial lethals, dominant in hybrid cytoplasm, an effective isolating mechanism is accomplished, since no or very few offspring are obtained from hybrid females which are backcrossed to any kind of males. The individuals carrying the lethals die in the egg or early larval stage. Whether or riot this mechanism in fact is employed in nature will become evident from an analysis of wild populations which is now in progress. Summary.-1. The break-points of the XR interracial inversion lie to the right of compressed and to the left of javelin, respectively. 2. The genetic isolation of race A and race B has progressed an additional step in the hybrids of certain strains where the females produce no or very few offspring. This is accomplished through interracial lethal genes connected with a maternal effect. D. E., Z. i. A. V., 52, 287-3 17 (1929). G., Ibid., 78, 338-398 (1940). 3 Sturtevant, A. H., and C. C. Tan, Jour. Genet., 34, 415-432 (1937). 4 Beers, C. V., Genetics, 22, 577- 586 (1937). 5 Sturtevant, A.H., Proc. Nat. Acad. Sci., 23, 360-362 (1937). 6 Sturtevant, A. H ., Ibid., 25, 308-310 (1939). 1 Lancefield, 2 Gottschewski, CHAPTER I I Jiig,h ~ ~re_g;ueno,y- .1,g ])rosophila pseudoobscura ~ Race di• by Klaus Ma.mpell Wm., G.. mcrclthof:f Laboratories of the JH ologioal Sciences , Cali.for:nia Ins-titute of Teohnology o Variations :i.n the mutation rate may be due to a number of factors in the external or internal environment of the orgQnism . We can control many of these factors ; unknown and a ppear spontaneously ; time and then disappear again . others are they may p~:rsist for some In some cases this may be due to a selec ti on of proper g~;nic mo<lifiers of the mutation rate . Modifying genes have been macle responsi'hle for the control of. the mutation rate , since the mutation rate can be altered by hybridization ; obsoura1 . such a case has b een desarif)ed :for D ~ pseudo- several investigators have observed cases of high mutation frequencies in Do :melanoga~~ter 2 ' 3 ' 4 ; the same nature are more doubtful ii othor cases o:f D.. pseud o obscn.u·a ~ race B, has furnished another example exhilli ting the sudden rise of mutation frequency t to be recordeid in this paper 4! The phenomenon first appeared in the progeny of a pair ma:ting , where :four sex- 1 inked mutants were observed " On ou·t ... crossing :Elies from tl:is culture , mutations occurred again to ian u.noxpe cted degree . is quite variable o The number of mu.ta"'cions per pair mating Some of the most strilcinc cul tu.res are recorded in Table l o only on.e p~:...ir mating . In each case we deal with the progeny of ~l:he average yi e ld per culture is 242 fliesj this ntnnber is based on counts of the progenies of 758 pair ' _matings chosen at ranclom ; ev en i n t he largest pr<>geni es t her e Table lo l of mutations ori g inal bd, y , ri , (spread wing e J 4 1045 v , Aw , I, , (Abnormal) 4 1176 bdot , dx, dy(m), Sc 4 1671 dy(m), 1 , Vg , (AbnOlJMllal) 4 1691 bd, 1697 bd , set, com, asc , Sc, (Houg h e y e) 6 2654 bd, dy{m), bbl. Aw, (rough eyeJ 5 2676 s ct , w , sn ( f) , Aw , t t 5 2708 dy{m) , tt, Tho , (Rough eye) 4 2936 bd' set, 1lig' (Rough (iye) 4 2971 bd' com, RA t 5 a177 dy( m) , set ~ tt, Cy, (wing-mosaic) ~)b . Vg com, a sc, jv' dO'li'lJ, ( ro n_p:h eye ) 5 6 2. are less than 500 flies per bottle. The mutations evidently may occur at an.y time <luring the develo.pment of the germ line or of the somatic line. Mutated individuals of a certain type may occur in nur11bers varying from one to fifty percent of the offspring of a pair mating; in somatic mutations patch.es of varying sizes have been observed. There seems to be no preference as to the time of mutation. The numbers of mutations recorded in this paper re- fer to separate occur:r.ences of muta tion~ not to numbers of indivi du.als showing new mutant characters. In our study only nvisil1le" sex-linked recessives are used as indicators of the high mutation rate, becausE:: they are detected in the irmnedtate offspring of a rate. fema1E~ with the abnormal Autosomal recessivos should not be u sed t because the p»obability of their detection depends on the mating system used" and even then it is not ce rta in at what time and in which indivi dual the,y have arisen. Dominants are d.isrege.rded to confusion with another mechanism which is responsible for a,1'/0j.d inducing mutations of the defiency type; later. we shall refer to this Alt;hough the high mutation rate is observed in ma.les al so. and_ al though dominants and t:mtosoms,l recessives tire produced by the present mechanism, it is for the above reasons that we restrict our observations to visible sex-linked recessives arising in f emales. The most likely cause of a phenomEmon of this nature is a genie one. ~gene It has proved dif :f.icult, however, to demonstrate that is responsible for the increase in the mutation rate. We deal here with a quantitative difference between the normal and. 3. the mutating strains; therefore, large numbers are required to make t he data meaningful. The normal spontaneous mutation rate for visible sex-linked recessives in Drosophila pseudoobscura ~ race B, ls 0 -.01%. ~Phis nu.'1lber is basecl on the occurrence of four se::ii.:... linkecl recessives in 46,804 :flies, from cultures preceding the time of the first observation. of the high mutation ra.te t plus those from the pure Seattle stra:i.n. In d.etermining the abnormal mutation rate we shall cons:td.er only those cul tu.res v11hich have given at le.a,st one .s ex-linked recessive. since this is the only indication of the presence of the nM:u.tator" gene in the mother. Thus, we obtain a mutation rate per c.ml tu.re showing :mutations; rather than a mu.tat .ion rate per total number of flies. The higher the total mutation rate, the greater will be the coir.1.cidence of several sex-linked recessives in one culture, :remembering that we always deal with the of:fspring of one female per culture and that the ~,verage number of offspring is the same for -wild type females and Mut&itor female With a .n ormal mutat,i.on rate of 0 .01'1o for sex... linked r e cessives it is obv-ious th.at two mutations would occur by cha.nee in the same culture so rarely that the mutation rate per culture showing mutations is p ractically i.oo. Thus , from Seattle fema.J.es seven mutations were obtained in seven cultures (rate: 1.00); from Morro females five mutations were obtained in five cultures (rate ::: i .oo). But from selected females 160 mute:tions were obtained in 102 cultures (rate : 1.57). The coinciclance of mutations is directly related to the total mutation rate, so 4. that by knowing one we cs.11 determi ne the other . Due to the small numbers , whi ch are always a disadva ntage in a study of this sort , the err ors may be considerable ; the calcul ations az•e also based on the assu..mption of a constant numbe:r.· of flies per bottle e which further decreases the accuracy of the result . Therefore, the rates which t?ill l1e presented must be taken as approximations; but we are interested i.n getting some idea of the mutat ion frequency in flies heterozygous or homozygous for the Mu.ta.tor gene. servative ~ In genera.1 , the estimates are probably con• ' since the normal spontaneous mutation rate is in- cluded · in the abnormal re,t e , which would tend to lower the coinc idence. Females resulting :from a mating of e. selocted fly to wild type , regardless of the direction in which the cross is made , give a r ate of 1.47 (100 mute.tions in 68 bottl es gi·ving mutations This i s gene . the~ rate obtained from fl:ies heterozygous for the Mutator Asswning a Poisson distribution, that is , e, population homogeneous :ror e. fac tor increasing the mutat ion rate, a result like this would be obtained. , if the rate of mtttat i. ons per t ota l 11,~hnbe r of fli es were 0 . 34}S. Thus , the norms.1 spontaneous P.'liiiiltati on process is increased thirty- four M.mes with one dose o+' the M1rtator gene . If flies can be homozygous for the Mutat or gene . we might ex:i;ieot a. higher mu.ta ti on rate from females that hav e resulted f rom a ma.t ing of both selected male s.nd fems.le ; or, conversely, only if flies resulting f:i:·om such a mating g i ve more mutatio ns, can we know that h omozygous flies a r e produced and that they h~ve a higher mutation rate than the heterozygous flies. Flfmal es resulti ng from such a ma:t i ng , give a r ate of 1. 76 (60 5. mutations in 34 cultures giving mutations). That is the rate obtained from females both of wh " se parents were probably heterozygous for the Mutator gene. From that mat ing we should expect one fourth of the offspring to be homozygous for the :Mutator gene one half heterozygous and one fourth wild type. Let m be the expected mean frequency of occurrence of a mutation in the population of subsamples . Since m is known to be equal to .0034 x 242 for cultures resulting from heterozygous females, and ~qual to .0001 x 242 for cul tu.res 1~esul ting from homozygous wild type :females, the relative frequencies of subsamples with one mutation, two mutations, etc. are known for these two groups. Let k be the number of cul tu.res resulting from homozygous M:utator females; then there are k m em cultures with one mutation in this group, 2k{.3615 ) cultures with one mutation when the mother was heterozygous, and k( .0234) cultures with one mutation when the mother was wild type; similarly for cultu..rc-:s with two, three, etc. nmta.tions. We equate the theo ... retical total of cultures with one mutat ion to the observed number with one mutation, a.nd a.o likewise for the cultures with two. three, etc. mutations; we obtain the following six equations; k ~ + 2k( .3615) + k( .0234) m3 + .0960) + -- 2 2 k(.!!L + .29r/'9) 2em 7 k(·6em 2 m4 k( 24em .0168) = 21 6. m5 tC 120 em + . 0028) = l k(m6 720e1II + .0004) =l The number of mutations in the homozygous group is equal to km, in the heterozygous group it is equal to 2k(242) ( .0034), and in the wild type group it is equal to k't242)(.0001); number of mutations obtaine& was 60; km+ 2k(242)( . 0034) +k(242)( .0001) the total therefore, we find = 60. or, k:• . 60 • We I.6'1-+ m can now determine the most probable value of m in the usual manner 5 • We find m = 1.703. Thus, the mutation rate of females homozygous for the Mutator is 0.70~, approximately twice that 0£ the heterozygotes . with the dosage. This is a linear increase In this connection, it is interesting that the gene dotted in maiz.e l-':;:·c d.1rnes ::::n exponential increase with the (losage as concerns mutations of the ''an gene 6 • The offspring of heterozygous flies should have a total mutation rate o:f 0 .17% if the e.bove oalcula.tions are correct, since now half the flies are heterozygous and half are wild type; therefore the mutation rate should be one half of O.34~f . we actually obtain 7 mutations in 4679 flies which is a rate of 0.15'% ; this is in good agreement with the expectation. The crosses designed. to locate the Muta.tor gene are shown in Table 2. All the females shown in the table should be heter- ozygous for the Mutator gene except those with the wild type chromosome whose homologue is responsible for carrying the Muta.tor gene; from these females we sho ul d expect s, rate of 1.00 whereas all the others should give an average o :e 1.47. Thus, chromosome II would se em to be the carr ier of the M:utator gene. Table 2. selected Stubble(II) or Scute(II~Xl!orro or Seattle ~ St ubble or Scut e or~&f :Morro or r10attle ~~ !liorro or Seat t~ # of muta.tions in culture~~ showing mutations I selected II wild_ type II selected III wild type III selected IV wild type IV rate 1.64 i.oo 1 I I~~ J 6 4 1.50 15 ,1D 8 1.86 11 1.36 This, hmm ever , must be taken cum grano. sal is; not only are the numbers uncomfortably small fo.r this kind of a test, but the a.ot chromosome which cannot be marked was not controlled. The con- clusion, therefore, is tentative. Table 3 shows the me,ps of the chromosomes; the map of the X-chromosome is b~i sea. partially on the one published previously? • Ifot all of the mutants shown on the maps can be assumed to have been induced by the Mutator gene here under discussion; there ie an other mechanism at work in the same cultures which induces an extraordinarily high number of mutations of the deficiency type , mainl0r Smoky, Notch, and i H.nutes. 1 Son1et;ime s it :i.s diffi cu.1 t to tell the two a.part• be- cause in some cases they have nearly the sa.me effect• Descrip- tion of the other mechanism will h~ .... ~ to wait until a later date; it is believed that the se:z:... linlrnd and. autosomal reces- sives and rnan..v of the a_ominants have been induced by the Mutator gene here under consideration. Since tests for allelism were not g enerally a pplied, the number of occurrenc<rn of a certa.i.n mutation can be given only for those loci where identification bJr phenotype is not likely ·to le.ad to confusion with other mutants. The asterisl{S ind.icate that the mutation has oecurI'ed at lea s t once 9 but the natu...re of the mutant does not permit identific&.tion b;>r insp ection. Thus, a.usky and miniature cannot be a.istinguished phenotypically; although many mutations of this type have appeared, it cannot be stat e d how i:1any of each. The ba,r signifies that th e mutant has not b een observed. The mutants were given the names of the D, melanogaster mutants which they resemble most closely . Some cannot readily be Table 3e IIo chromosome X-chromOf30In0 o. o lozenge (lz) almond.ex (amx) beaded (bd) 3.6 3lol 1viinute4 iI~14) Hairless · (HA} 35.5 54.1 61.,9 61Q9 scutella.r (set } yellow (y) prune (pn) deltex {d:ir) 64 .. 7 !fotch {l~ ) white hvJ lm1l3 singed (sn} letha~ ·Jpl1001s ( lP) dusky (dy) minia ture {m) forked (f) rose {n3J tilt (tt} bubble (bbl~ fuse dl (ful) compressed (oom) ascute ( as<B) l~ bno1~mal . wing (Aw) Curvoid (Our) short is) radius incompletus (ri) shiny {shi) tiny bristle (tb) c on<:lenaed ( e o?ll.} rou.p.:hex ( rux) notchy (ny) tlowny ( dow) diminutive (dm} small bristle (sbr} ruby (rb) approximated (app} twisted. (tw) 2 6.,1 10 li4~ - 39 o2 42,,1 6 cinnabar (cnJ 54.7 2 3 24 3 Tuiinu t e8 Oa8 \ Minutel4 (M:1 4 ) :5rl Mutator CM:uJ ? • o.o Abbreviated (AbbJ IJobe {L) 29.2 7306 73.6 74.1 Scnte (Sc) Vestigial (Vg) 50.5 50 .5 9B .,9 10 176.2 l Bl "8 7 >t .)): (],15) Minute6 (11.1:6) Minute7 (M7} MinutelO (MlO) Mi:nute12 (M12) Minute 15 (:Ml5) * * * IV~ *3 * * * * * * * .,i. ··- * * * * * * * * * * * orange~( or} Min11te~) *5 106.,2 1.19o3 * * ~~ gap (gp) 92.,3 ~)9 . 8 * * III. chromosome 73,.5 scarlet f st) lm2B . 149,.0 sepi a (se} radi us ru.d:i.mentary (rr)l7l.4 javelin ( jv) 40 5.1 73 .. 1 77,.2 83 .7 91.6 • o.o Hairle~:isE (HE) bi thorax ('bxJ glass (gl) 71.0 72.8 '72.9 • 1 J_on . {v ) verm:i._ Filifo:r.m (Ipf) :&1 usedr.. ( Fu 2 ) * 50 * Smoky (Sm) 'l'horax ( 'l'ho) Stubble (Sb} chromosome Minute3 (M3) o. o Scutellum diminished (Sd)32.9 BairlessB (HB) 57.2 * Enhancer-Rlass fE- gl) * * * * * ** Iviinute 2 (M2) ~·· ;vnnu t e 9 (]~9) vJ. * Curly ( Cy:r) Dt:.i,chsoid (Dsa.} net (net) daohs (d) J, arge wing cells (Lwc) Minutel (:Ml) Minut ell (Mll) Minutel3 (:M13) 69.7 * * * * 8. compared. RairlessA i s a dominant, lethal to the male. Both prune alleles &.l\lliays show a distinct mottling effect. The del tex alleles are fepiP.J..e sterile and semilethal to the male.. Abnormal wing has large wing cells giving the wing a. coarse appearance ; delta or net type of venation is common; it is usually lethal to the .male . In shiny the chitin has a polished ap pearance. Thorax shows a longitudinal mea_ian groove of the thorax as the most striking characteristic. homozygous; RairlessJ3; :rarely survives when i.n the homozygous state it shows e:ittreme aristo- pedia effects; the heterozygous flies sometimes have the aristae reduced or absent; the legs also are sometimes reduced. Vestigial takes off some. bristles a l ong the costa when heterozygous . Scmtellu.m diminished affects the size of the scrutie1lum which may be practically absent ln extl?eme cases. heterozygous HairlessE. the eyes. HairlessB resemr)les Dachsoid shortens the legs and roughens I,arge wing .cells resembles Abnormal '1Vimg in its I!l.nhancer ... glass allows heterozygous gl ass to phenotypic effect. exp re ~.:rn a. rough eye• In addition to the mutan t s shown in the maps there has occurred a large number of mutants which were either inviable or steriie, lost or discarded before being mapped . or whi ch have not yet been mapped . The total number of observed mutations is ap- proximately two thousand. Aside from the evolutionary significande of Mutator genes it is interesting to specu3'ate about their mode of action . From the nature of the matati ons which are induc ed by this mechanism 'it seems obvious that we are dealing with true point mutations • .· Muta tor c:·enes probabl~r act through a chemical medium. Chemicals are the least satisfactory agents of the more important ones which have been investigated as to their fE.teul ty of inducing mutations.. It is possible that Uutat or genes interfere with the prope r reprocluction of genes and that by learning the nature o:f this mechanism we may know more abou.t the chemistry of gene re nrodu.ction.. Mutator genes re.ther than producing c ertain sub- stances might be responsible for t he lack of a substance necessary for growth and r0production of genes in general . There is no good chance at this time to learn much a.bout Hutator genes exce:ot the result of their act.ivit,y, unless the su'bstance invol ved were in the na tu.re of a hormone; this is unlikely, since intranuclear procesBes are concerned., f~ture It must b0 le:ft to research further to illuminate this problem. The author is indebted to Professor A. B..,Sturtevant for ~:iugge st:ions and criticism and to Mrs .. Harrie t Bonner for the I application of the n1athE'-::matical treatment to the a.ata. Summary. 1. Drosophila _pscudoobscura, race B, has furnishea_ certain stocks with a,n abnormally high mutation rate. 2. The mutations m.-"ly occur at an,y time during the develop- tnent of the germ line or of the somatic line. 3. The phenomenon is due to a dominant ":Mutatorn gene wh ich lhcreases the normal spontaneous re.te about thirty-four times ;,~h,en heterozygous~ and about seventy times when homozygous; this represents a linear inorease in the mutation rate with t he ci..osage of the Mu.ta tor gene" 4,. The Mutat or probabl,it is linked to the secoml chromosome. 5. A.total of approximately two thousand mutati ons was observed ; some of t he mutations were located. 10. 0 turt· eva~ + • l Ct u ' J .. A,.A H G . i;. • ' v ~ ~• .;;, • J.1 . ... 0 u tJ 0 Q~, G v • ~-508 - 310 (1939) . 2nemer e c ~ M ., Genetics 22: 46 9-47 8 ( 193?) # 3Tiniakov.G o, C ~ R.Doklady A c ad . S oi.U~ R . S G S c 22: 609-61 2 (19 39). 4r~eel,J .. v., Genetics 27 : Bl 9 - 5756 (1 942 ). 5Whi ttaker a nd Robinson , The CalculuG of obso:r.vati:ms, n1ao kie & son : 214 (19 24). CHAPTER III A DEFICIENCY INDUCER IN DROSOPHILA P,SEUDOOBSCURA, RACE B by Klaus Mampell Wm·G·Kerckhoff Laboratories of the Biological Sciences, California Institute of Technology INTRODUCTION Recently a case was described where a Mutator was responsible for a great increase in the appearance of point mutations (Mampell 1943). In the present investi- gation a gene induces mutations of the deficiency type such as Smoky, Notch and Minutes; a high percentage of mosaics is common. In Drosophila and Zea genes have been described which by various mechanisms induce gross mutations. The present phenomenon is unlike any of the others, although many of the effects are the same. instance: For The "sticky" gene in maize alters the cell physiology; mitotic and meiotic abnormalities are com- mon; both chromosome aberrations and mutations are induced (Beadle 1937). Chromosome elimination occurs in the eggs laid by D· simulans females homozygous for the claret gene (Sturtevant 1929)· Minute-n in D· melano- gaster produces patches in which the Minute carrying X-chromosome is frequently eliminated (Bridges 1925a). 1 2 Somatic crossing=over leading to small mosaic areas is accentuated by certain minutes (Stern, 1936). Mosaics of the haploid-diploid type were observed in certain stocks of D· pseudoobscura, race A (Crew and Lruny 1938)· The mechanism by which the "Deficiency Inducer" acts is still obscure, but some of the results of its activity can now be presented. METHOD:3 As with the Mutator stocks, in the present study we deal only with cultures that are the result of a pair mating. To avoid confusion with the Mutator, only Minutes, '.Smoky and Notch and all kinds of mosaics are taken as indicators of the presence of the Defi= ciency Inducer. All other kinds of mutant individuals are neglected. This is necessary, because the Mutator and the Deficiency Inducer must be supposed frequently to exist in the same individual, but we are able to tell them apart by the kind of change they induce. Most of the mosaics appear to be bilateral; it is not necessary, however, t;o assume that other mosaics are less common. Bilateral mosaics are more easily detected, and the flies were not carefully examined for possible small patches of mutated tissu e.. Practically all the Minute mosaics re= corded were mutated at least for half the thorax. Other patterns have been observed, but are not usually recordede THE APPEARANCE OF THE PHENOMENON The Deficiency Inducer was found in the Seattle strain which had been suspected for some time of produc= ing an unusually high number of Minutes a.nd various mosaicse The number of mutations per pair mating is variable accord= ing to the genetic constitution of the individuals and the chance coincidence of the random distribution. ,'Blome of the cultures with a particularly high frequency are recorded in Table l· In each case we deal with the progeny of one pair mating. The average yield per culture is 235 flies; this number is based on counts of the progenies of 851 pair matings chosen at random. Even in the largest pro= genies there are less than 500 flies per culture~ This is not t~e, however, for lines with particularly high frequencies. There, the average number of flies per culture goes down to 96· This number is based on 41 pair matings yielding 3920 flies in a line considered homozygous for the Deficiency Inducere As will be ex.plained later, it is probable that an appreciable number of dominant lethals are induced. This may account for the lower yield in the latter series. Only cultures with five or more mutations have been recorded in Table l· Those with less than five are, of course, much more nuraerous. The numbers must be taken TABLE l• Type of' mutation Culture No. of' No. mutations M Sm N other Minute 'Smoky Notch other --- dominants mosaic mosaic mosaic mosaics 1079 1270 1342 1415 1453 1457 1593 2070 2221 2228 2238 2528 a28'7 a666 a67'7 .a683 a.758 b 85 b 92 c '71 cll7 c242 c245 c253 el43 e21'7 + - + + + - 6 1 8 7 5 4 9 3 12 15 2 1 2 5 4 2 2 2 3 + - + + ... 4 6 1 2 4 4 2 + = + - 3 += + + + + + + + + + - + - 1 2 1 3 + 1 3 4 3 5 4 4 1 2 2 4 4 6 9 5 6 5 5 5 5 8 6 5 1 2 2 1 3 2 2 2 4 2 l 5 5 2 4 3 1 l 2 6 8 7 8 5 11 5 6 '7 at least one individual of that nature observed. = no individual of that nature observed. Nu:mbers ref'er to number of mutations, not nUlllber of individuals showing new mutant characters, although in the case of mosaics, the two coincide. 4 as a minimtUn, since only clear cases are included. There may have been different Minutes in some cultures, but since they are usually difficult to tell apart phenotypically, only one Minute is recorded; in the case of the mosaics, those with missing parts of the anatomy were neglected, although they probably represent dominant lethal mutations. There is no doubt that the mosaics represent real mutations, since in some cases they affect the germ line and are then inherited as single factors. Although no cytological investigation has been made, it is probable that we usually deal with deficiencies, since Notch, Smoky (Delta of D· melanogaster) and Minutes are commonly found to be due to deficiencies. Whether these represent actual deficiencies or an inactivation of a series of genes is unknown, but the former is more probable• In order to explain the deficiency type of mutation, we might postulate faulty chromosome division as the cause for their origin· Incomplete reduplication of the chromosome would lead to a large deficiency probably lethal to the individual or to a smaller one which may permit the individual to survive. With this in mind we can understand one reason for the predominance of mosaics. The mutation evidently occurs at any time during the development. 5 Mutated individuals of a certain type may occur in varying numbers in the off spring of a pair mating; there seems to be no preference as to the time of mutation in the germ line or the somatic line since mutated patches of varying sizes have been observed. One of the reasons for the predominance of mosaics is the possibility that, due to the gross kind of mutation, those individuals survive more often which have some wild type tissue. Mo st of the gross mutations are lethal to the individual, and though they oc cur at any time, they are detected only when they occur late enough to affect only part of the fly. 'Ihus, dominant lethals are sometimes detected when they happen to be linked to a dominant marker gene and when they arise early in the development of the germ line. On out- crossing such an· individual to an unrelated fly the ratio of the dominant marker to wild type would be upset; such a case has been observed at least once. A male heterozygous for Stubble was outcrossed to a wild type female· There were seven Stubble flies and 358 wild type flies in the off spring; the ,s tubble flies had tumors and were weak. Flies frequently emerge with various parts of their anatomy missing. This could be interpreted as the result of the action of dominant lethals arising in somatic tissue and accounting for the missing parts usually of the thorax such as legs, wings 6 and halteres~ Obviously, if the dominant lethal occurred somewhat earlier, the whole fly would be missing, and t he mutation would not be discovered. The second reason for the apparent predominance of ! mosaics is the fact that only one Minute was recorded per culture regardless of the actual number of Minute : indivi= duals appearing in that culture. Since the Minutes are usually much alike phenotypically, no attempt was made to distinguish between t hem. Only through linkage tests would it be safe to conclude how many different Minutes there may have been in one culture. This was not attemptede There are only three phenotypes which were used as an indication of the Deficiency Inducer: Minutess Notch, Smoky and In a line with a particularly high rate we should find some cultures with more than one of the three . phenotypes represented. ~nee this is the case (a677 and a683 in Table 1), it is safe to assume that in those cultures we sometimes deal with different Minutes rather than with several individuals showing the same Minute. A third reason can be given. As will be discussed later, it is possible that cytoplasm which has been pre= conditioned by the Deficiency Inducer will induce deficienci es. If this is so, homozygous wild type individuals may yet have deficiencies induced by the action of the cytoplasm, but all of these individuals would be mosaics, since the deficienQies could arise only somatieally by this mechanism. 7 Certain Minute mosaics serve as an indication that deficiencies of varying extent are produced. When a Stubble female was outcrossed to a wild type male, a mosaic occurred in the offspring which was Minute for half the thorax without indication of ~tubble, whereas the unrnutated half was Stubble. This is interpreted as a deficiency including the Stubble locus and by removing the gene, .it would also remove its phenotypic eff'ect. In a similar case, when a Stubble HairlessE female was outcrossed to a wild type male, a mosaic female in the off spring was Stubble HairlessE on the right side whereas the mutated left had an abnormal eye and wing, but was wild type for bristles. This probably means that a deficiency arose long enough to include both the Stubble and HairlessE loci, thus removing their phenotypic ef= fects. Similarly, in two instances, a Stubble female produced a female whose one half was Stubble, while the other half was wild type for bristles, but had a smaller eye and wing. The latter three cases may also be ex- plained on the same basis as the two following ones, namely: haploidy for part of the tissue. There are two examples for this kind of mosaic. In one cases a wild type male was mated to a female homozygous for seven sex-linked mutant genes: rs st se s. bdct set dy In the progeny, one mosaic was wild type female on one side and bdCt set dy rs st se s female on 8 the other side$ Similarly, a bdct set dy rs st se s male was mated to a female heterozygous for wild type and bdCt set dy rs st se s. In the progeny a mosaic occurred which was wild type female on the right half and bdct set dy rs st se s female on the left half. ~ince the side showing t h e sex-linked recessives was female in both cases, it could not have arisen merely by elimination of an X-chromo= some, for then, that part would have been male. Although in both cases other explanations are possible, it could both times be a matter of an entire chromosome set failing to divide~ Then the part revealing the sex=linked reces- sives would be completely haploid and, therefore, normal female• In the first case the male set, in the second case the female set could have lagged one division. Haploid-diploid mosaics have been described also in D. melan- . ogaster where they were first observed (Bridges 1925b), and in D· miranda·D· pseudoobscura hybrids (MacK.night, 1937) Gynandromorphism has also been observed. It could likewise be explained on the basis of an entire chromosome failing to divide. THE CAUSE OF THE PHENOMENON The most likely cause of a phenomenon of this nature is a genie one, but it has proved difficult, so far, to make a gene responsible for the phenomenon. However, unless clear evidence to the contrary is obtained, we shall take the genie cause for grantede 9 Little is known about the location of the Deficiency Inducer gene.. .S Q far, the only cJ::iromosomes which can be excluded are Y, X and II; there is then purely negative evidence. When Stubble is mated to Morro, and the 2tubble sons are again outcrossed to Morro females, the wild type offspring are homozygous for wild type second chr omosomes. On outcrossing these flies to Morro again, fourteen muta= tions are obtained in nine cultures showing mutations. Therefore, chromosome II is excluded, and from the nature of the cross it is evident that at the same time we exclude Y and X• THE EXPRESSION OF THE DEFIC1ENCY INDUCER In race B the normal spontaneous mutation rate for Minutes, :Smoky, Notch and various mosaics is so low that . no percentage can be given. In 30819 flies from cultures that had never been crossed to the ~eattle strain, only one Minute has been observed. When a Minute is observed in a culture, this is taken as an indication of the pres= ence of the Deficiency Inducer in the parent. We are in= terested in knowing the mutation rate of males and females heterozygous or homozygous for the Deficiency Inducer• In order to determine the percentage of mosaics among flies heterozygous for the Deficiency Inducer, we count the number of mosaics in cultures that have a Minute, :&ioky or Notch, and that have resulted from a mating of a selected fly to 10 wild type (Morro in this case). The appearance of one of these mutants indicates that the selected parent carried t h e Deficiency Inducer. We know that this parent is hetero= zygous for t he Deficiency Inducer , becau se it has resulted from a mating of a selected fly outcrossed to ~~rro. Mosaics in this case are supposed to have resu lted from a mutation in the zygote at the earlies t. From heterozygous male out~ crosses nineteen cultures with Minute s were obtained. Multi- plying this number by the average number of off spr i ng per culture we obtain a total of 4465 flies half of wh ich will carry the Deficien cy Inducer. Among these 223 3 heterozygous Deficiency Inducer flies there are three mosaics or o.13%· From the same kind of female outcross, we obtain fourteen such cultures or 3290 flies half of wh ich are heterozygous . There are here eight mosaics in 1645 heterozygous Deficiency Inducer flies or ~ rate of o.49%· Th e ch i square test shows such a difference to oc cur by random sampling with a proba= bility of about 5%· There are thus almost four times as many mosaics when the heterozygous parent was female than t h ere are with the male as the heterozygous parent. We deal t hen probably with a maternal effect , that is , the Deficiency Inducer gene is more effective in cytoplasm that has been preconditioned by the Deficiency Inducer gene than in wild t ype cytoplasm. From a certain mating (c117 in Table 1) seven mosaics were obtained in 26 5 flies. This culture may be homozygous 11 for the Deficiency Inducer genes established from this culture~ ,Bome inbred lines were Among a total of 3920 descendants there were 62 mosaics or a rate of 1·58%· If it is true that a maternal effect is connected with the Deficiency Inducer , we should expect the following rates: one dose in wild t ype cytoplasm, (this gives a rate of 0 . 13%· as was learned from the male outcross), no Deficiency ~nducer, one dose and two doses of it in heterozygous cyto = plasm; no Deficiency Inducer, one dose and two doses of it in homozygous cytoplasm. All of t hose we can tell onl y by carefully marking the Deficiency Inducer s This has not been ac complished so far ® We do know that the combination of no Deficiency Inducer and one dose of it in heterozygous cytoplasm gives a rate of Q. 49% , but we do no t know which proportion of t his rate is repre sented in either group. We now determine the relati ve numbers of male and female mosaics . A total of 459 mosaics i s considered her~. Of t hose, 219 ar e males and 240 are females. Among t he females there are 12 Notch mosaics wh ich cannot oc cur in males, s i nce Notch is lethal to the male.. subtracting t h is number from the total number of female mosaics , we get 228, or a nearly equal number of male and female mosaic s.. Since there usually is a slight excess of fe = males in the cultures , the numbers are certainly not sig= nificantly different. We should , however, expect a higher number of Minute mosaics in female s than in males, since 12 a sex=linked Minute would probably be lethal to the maleo Indeed, there are 95 male Minute mosaics and 135 female Minute mosaics, yet, the total number of mosai c s is approx= imately the same in the two sexeso This might mean that the Deficiency Inducer affects the males more readily than the femalese They would have an excess of mutations in the autosomes to make up for the excess of sex=linked dominants the females can revealo Thus, there are fifteen male :&noky mosaics and eleven female Snoky mosaics, but, since the numbers are small, the difference may not be significants The other possibility is that the Deficiency Inducer is responsible for sex=linked recessives which would show in males, but, of course, not in female mosaicse This is the more probable explanation, especially since mosaicism for a condition strongly suggesting beaded has been observed . ten times in males~ but never in females® Also, mosaicism for conditions indi cating such sex-linked recessives as dusky, tilt, white and yellow has occurred in males~ but never in females® This would mean that recessives are induced as well, but it may be that they are of a different nature than ordinary recessives$ Perhaps we deal here with very small deficiencies including just one locus$ If we assumep then, that the mutability affects both sexes equally, we must come to the conclusion that the mechanism induces an approximately equal number of reces~ sives and dominants, but since dominants seem to be in the 13 minority in the ordinary spontaneous mutation process, it would induce more dominants proportionally• This is con= trary to the observations with the Mutator, which also induces recessives and dominants, but seems to do so preserving the proportions of the normal spontaneous mutation process. THE DISTRIBUTION OF THE MUTATION:S The Minutes seem to be distributed at randoms Only the X=chromosome shows relatively fewer Minutes, but this can be explained on purely technical grounds. Since Minute in the X is almost always lethal to the male, we would automatically discriminate against sex-linked Minutes by employing the technique usually used in allocating mutants to chromosomes, namely, male outcrosses. In a random sample of 85 new Minutes, 34 were males and 51 were females. The excess of females indicates that sex-linked Minutes occur wit h proportionate frequency. rived from the mosaics. ~iency The same conclusion was de = We assume, then, that the Defi- Inducer affects the chromosomes at randome Table 2 shows &loky, Notch and the Minutes as far as they have been located. DIOOU$SION If our assumption that the cytoplasm plays a role in the production of these mutations is correct, it would seem that we may deal with a greater viscosity of the cytoplasm TABLE 2 x- chromosome Minute4 35.5 Notch Minutel7 Minutel9 II· chromosome :smoky (25)* O.. o MinuteB Minutel4 Minutel6 Minutel8 III· chromosome IV. chrome some Minute5 Minute6 Minute? MinutelO Mi nute12 Minutel5 Minute20 Minute3 o.o Minute1 Minute 2 Minute9 Minutell Minutel3 * (48 ).i,~ Numbers in parenthesis refer to number of occurrences. 14 which t hus would tend to break the chromo somes. In that case it wou ld not be explained why we do not observe in= versions and translocations. The same would hold, if we were dealing with a. spindle abnormality. Here, likewise, we should expect aberrations of all kinds. It is necessary to make a number of crossing-over tests in order to discover inversion s genetically , but translocations should be dis= covered readi~l· In the great majority of male outcrosses t h e autosomes were marked with dominants. Any transloca- tion affecting t hem would have been discovered. There is only one single case suggesting a translocation, and that one is between homo logous chromosomes. A male which was Smoky over Stubble was outcrossed to wild type. In the off spring, besides the expected :&noky and Stubble flies, Smoky Stubble and recessive :sltubble occurred as well· If . t h e breakpoints were close and in the vicinity of the Stubble locus, we might get a small duplication in one chromosome and a small deficiency in the other . This is t h e only suggestion of anything resembling a translocation. It is because of the fact that apparent deficiencies are the only abnormalities produced t hat the author prefers t h e hypothesis t hat a failure for the chromosomes to be perfectly reduplicated accounts for the deficiencies0 In t he discussion of the Mutator gene it was postulated that the Mutator may be responssible for the lack of a substance necessary for the proper growth and reproduction of genese 15 I t is indeed intriguing to think that the Deficiency Inducer might be responsible for the lack of a substance necessary for the proper growth and reproduction of the connecting fibers linking the genes togethere We could then understand why it could induce mutations of a much grosser nature than the Mutato~ s ince the area involved could be much more extensive ~ Anything from one locus to the entire chromosome set could be and evidently is affectede Together these two genes would control one of t he most original and vital processe s that are the property of chromosomes and genes. SUMMARY l· A gene is described which induces mutations of the deficiency type, such as Notch, Smoky and all kinds of Minutes and mosaicse 2. The "Deficiency Inducer" acts in males and females, but a maternal effect seems to be connected with the difference in rates, since heterozygous females pro= duce more mosaics than heterozygous malese 3. The Deficiency Inducer is not in Y, X, or II· 4. The mutations are distributed in random fashion over the chromosomes • 5e The possibility that the Deficiency Inducer has to do with chromosome growth and reproduction is dis= cussed .. 16 LI TERATURE Beadle, F· w., 1937. Chromosome Aberration and Gene Mutation in Sticky Chromo some Plants of Zea mayse Cytologia, Fujii Jubelee Volume: Bridges, c. B., 1925a 43=56 Elimination of chromosomes due to a mutant (Minute-n) in Drosophila melanogaster. Proc. N a t. Acad~ Sci. 11:701-706. Bridges, c. B. , 1925b. Haploidy in Drosophila melanogaster. Proc. Nat. Acad. Sci. 11:706=710. Crew, F· A· E· and R· Lamy, 1938 . pseudoobscura. J. Genet. MacKnight, Re H ., 1937. Mo saicism in Drosophila 37:211~228. Haploid Ovarian Tissue in Droso- philao Proc. Nat$ Acad. Sci. 23:440=443. . Mampell, K., 1943e Hi gh Mutation Frequency in Drosophila pseudoob scura, Race B· Proc. Nat. Acad. Sci. 29 : Stern, c., 1936. '&>matic crossing-over and segregation in Drosophila melanogas t er. Sturtevant, A· Ho , 1929. Drosophi la simulans: The claret mutant type of a study o:f chromosome elimina- tion and of cell-lineage $ 135:323=356 .. Genetics 21:625~730e Zeitschr.f swiss mZool. CONCLUSION Every research must remain, more or less, incomplete. There is practically no end to the questions that have a bearing on any problem. to answer all questions? use his own judgment. Is it necessary Again, the investigator must It is as important to know where to stop as it is to know how to proceed with t he res.e arch. It is easy to lose sight of the real aim and to follow blindly the many alleys which lead nowhere in particular. Often it is helpful to terminate the research temporarily, and to view it from a distance. Then, the big question-marks are still visible whereas all the innumerable little ones have faded away. It is these big questions which we must aim to solve. In the case of the "female sterility" the genetic analysis could be extended, but the important question: 11 Is this mechanism employed in wild populations?", demands an answer in preference to all others. It is necessary to test populations of race A and race B from regions where they overlap and from those where they never come in contact for the frequency of the 11 1m 11 genes. We know now how the genetic isolation of these species can be accomplished. After the study of wild populations we shall know whether some populations have i already become so isolated. This research may profitably be continued in the future. The 11 high mutation freque ncy" research requires for its continuance the maintenance of the stocks, which in itself constitutes a problem. The few investigators who have worked on problems of a similar nature can appreciate the difficulty. The inevitable fate of these stocks has been t heir loss. Studies of this sort are carried on at a great disadvantage to the investigator• He never knows whether the next step can be reached, for the phenomenon might have vanished in the meantime. Mutator genes are notorious in this respect, and the author does not know at this time, if the Mutator still exists in the cultures. Many times before the phenomenon had apparently disappeared, only to reappear periodically. Thus, the uncertainty makes the research extremely difficult• However, it is still planned to solve many of t he important problems connected with t his work. We now know the effect of the Mutator. It is doubtful that the most important question: "How does the Mutator act?", can be answered in the near future., With the " Deficiency Inducer" the same difficulties exist. Work with this problem is being pursued actively at this time. It must be pointed out again that a non= genie explanation has not been excluded. ii There are reasons for suspecting an infectious agent as the direct or indirect cause of the deficiencies.. Epidemic-like waves of deficiencies of the same nature as those described in Chapter III, have persistently appeared in other stocks of race B, in certain stocks of race A, and in D· melanogaster.. "gene", 11 If we substitute 11 organism 11 for mild and heavy dose'' for "heterozygous and homozygoustt, the facts could be explained just as consistently on that basis. The possibility of a contagious disease is not too fantastic to be at least considered~ If one can show that the cause is linked to a definite chromo= some, it can safely be called a genee now to establish this. Work is in progress The solution may be expected soon• In the meantime, the Deficiency Inducer will tentatively be considered a ·gene. The next important phase of the work is the cytological investigation. Not only will this throw some light on the nature of the "deficiencies" that are induced, but we may also find out whether or not the cytoplasm per se is affected. question: 11 This leads to the important How does the Deficiency Inducer act?" As with the Mutator, the answer should not be expected soon, if it can be given at alle In both cases transplantation and injection experiments are planned to test for the possibility of diffusible substances; particularly in the case iii of the Deficiency Inducer where a cytoplasmic effect seems to exist, , those experiments may yield some resultse It was pointed out in the foreword that new problems, completely independent of those discussed~ arise constantly. future. Some of these may be preferred in the The author does not want to predict which problem will be solved next. iv