MARINE ECOLOGY PROGRESS SERIES Vol. 4: 195-202, 1981 - Published February 27 Mar. Ecol. Prog. Ser.

Relationship Between Breeding Systems and Resistance to Mercury in Euplotes crassus (Ciliophora: Hypotrichida)'

Department of Zoology, University of Toronto, Toronto. Ontario, Canada M5S 1Al and Istituto di Zoologia, Universita di Pisa, 56100 Pisa, Italy

ABSTRACT: A toxicity bioassay was used to measure the median tolerance limit to mercuric ion of different stocks of the marine hypotrichous Euplotes crassus. The stocks were representatives of local populations diversified in their breeding systems (inbreeding or outbreeding). The results show a greater tolerance to mercury in outbreeders Genetic analysis of interstock tolerance-differences indicates no simple genetic basis. An interpretation is presented which suggests that the different degrees of mercury tolerance revealed by the representatives of inbreeding and outbreedlng populations could depend on their peculiar genetic organizations. Such peculiar genetic organizations might be the product of the distinct ecogenetic strategies followed by the populations to which the stocks belong.

INTRODUCTION tant factor determining the degree of resistance to environmental stress (Nyberg, 1974). Excess toxic substances are often introduced into The object of the present study was to establish ecosystems as by-products of industrial processes. The whether a relationship exists between breeding sys- sudden changes in concentration of toxicants usually tems and resistance to severe stress caused by mercury. subject organisms to severe stress. There is ample As is well known, mercury plays an important role in experimental evidence that different species - as well water pollution. as different populations belonging to the same species The cosmopolitan marine ciliate Euplotes crassus - -may vary, sometimes widely, as far as the tolerance to recently found to comprise populations committed to a particular toxic agent is concerned (e.g. reviews by either inbreeding or outbreeding (Luporini and Dini, Ford, 1971 and Cairns, 1974). Once these observations 1977) - was used as a model system. The overall become established, the investigation of the mechan- morphological features of the stocks used in this isms responsible for the differential susceptibility of investigation conform fairly well to the general the organisms, and the factors affecting their ability to description of the classical morphospecies E. crassus. withstand varied toxicant concentrations in the envi- Yet, these stocks consist of at least two sets of mating ronment, become the major tasks for the biologist. The types between which gene flow does not occur; this system of breeding (the degree of inbreeding or out- indicates that they constitute two distinct biological breeding) followed by an organism may be an impor- species. However, the fact that related biological species cannot yet be identified reliably leads me to - group both aforementioned species in the same classi- This work was partially accomphshed at the Ramsay cal, typologically-defined species - E. crassus. The Wright Zoological Laboratories, University of Toronto, observations to be reported indicate that mercury is Canada, with the support of a NATO Fellowship awarded to F. Dini through the Italian National Research Council (CNR) highly toxic, even in trace amounts, and indicate a " Present address: Instituto di Zoologia, Universita di higher tolerance to mercury in outbreeding than Pisa, Via A. Volta4, 56100 Pisa, Italy inbreeding populations of E. crassus.

O Inter-Research 196 Mar. Ecol. Prog. Ser. 4: 195-202, 1981

MATERIALS AND METHODS gation was attained at about the same time, following an immature period which was short compared with Natural Sources the immature period of the A stocks (Luporini and Dini, 1977; unpubl. observations). The duration of the Of the 5 stocks of Euplotes crassusused, GI, GvIl,and other stages of the life cycle was also relatively short, G,, were derived from collections at the same locality, and so was the total life cycle (Dini and Bertini, 1978; whereas PB, and ST were collected from different unpubl, observations). Moreover, several indications locations, both geographically very far from the former. led Luporini and Dini (1977) to argue that the occur- The locations from which the stocks were collected are rence of conjugation was largely a laboratory listed in Table 1. phenomenon, suggesting that autogamy was the more

Table 1. Euplotes crassus. Geographic location, breeding pattern, stock designation, and mating relations among the stocks used. Percentage of mating was scored by observation of living cells under a stereomicroscope (20X); + + + + roughly 70 % of the cells engaged in pairing; ++ roughly 20 %; + roughly 10 %; - 0 %

Geographic location Breeding pattern Mating relations Stock G, Gwl GI1 PBI ST

Gesira/Somalia (Indian Ocean) Inbreeding GI ++ + + { G",, ++ + Gesira/Sornalla (Indian Ocean) ++++ - Piombino/Italy (Tyrrhenian Sea) Outbreeding - S. Terenzo/Italy (Ligurian Sea)

Breeding Features likely form of fertilization pursued by A+ stocks in nature. All these features, plus the considerable non- The Stocks G,,, PB,, and ST (hereafter referred to as viability following conjugation (Luporini and Dini, 'non-autogamous, A-, stocks') reproduced sexually by 1977), point toward an extremely inbreeding way of conjugation only (a process of reciprocal fertilization life (Sonneborn, 1957; Nanney, 1980). Of interest to occurring between temporarily paired cells ordinarily this investigation are the genetic consequences of of complementary mating type). They were chosen as autogamy in Euplotes crassus. When the 2 unlinked the best characterized, healthy-growing representa- loci a (autogamy) and mt (mating type) were consid- tives of local populations whose overall features speak ered, the appearance of recombinants in heterozygous for highly outbreeding patterns (Sonneborn. 1957; A+ stocks after having passed through many suc- Nanney, 1980): long periods of sexual immaturity, cessive autogamies was an extremely rare event maturity, senility, and a long life cycle; cross conjuga- (Luporini and Dini, 1977; Dini and Luporini, 1980). tion (interclonal conjugation) during maturity, but Hence it can be inferred that the A+ stocks largely cross conjugation and (intraclonal conjugation maintained their original genotypes characterized by a due to assortment of the alternative phenotypes of certain degree of heterozygosity (Luporini and Dini, heterozygotes at the mt locus) during senility; a system 1977) - an inference confirmed by the present investi- of multiple interbreeding mating types controlled by gation, at least with respect to Stock Gvrl. one-locus multiple alleles (unpubl. observations) according to the model proposed by Heckmann (1963). All these features, except selfing, concur to favour Breeding Relations mating between unrelated partners in nature. If a suit- able unrelated partner is not found, eventually selfing GI, GVll,Gill and PB, were conspecific stocks each may occur; it serves to initiate a new life cycle. In the belonging to a different mating type. This statement absence of fertilization death would occur. was based on the results of mating tests (Table 1) and The Stocks G, and Gml (referred to as 'autogamous, breeding analyses (taking as marker genes those A+, stocks' in this paper) were representatives of popu- determining autogamy (a) and/or mating type (mt)) lation~in which fertilization occurred either by conju- which indicate that all stocks were potentially able to gation or by autogamy (a process of internal fertiliza- exchange genes with each other to a greater or lesser tion occurring in single, i, e. unpaired cells and repre- degree (Luporini and Dini, 1977, unpubl. results). In senting the closest form of inbreeding; Luporini and combinations between the AC(GIand Gvll)and A-(GIl Dini, 1977). The ability to undergo autogamy or conju- and PB,) stocks feeble mating reactions, several D~ni:Breeding systems and relsistance to mercury in Euplotes 197

induced homotypic pairs (between cells of the same mixing Stock GI1 with a long-maintained laboratory mating type), and low viability rates after conjugation stock belonging to a different biological species and usually occurred, particularly when the PB, stock was marked by small cell size. Only vigorously growing involved. The inherent low mating reactivity of the A+ clones were retained, and they were maintained in stocks (Luporini and Dini, 1977), and a nucleo-cyto- active log-phase growth by being fed excessive food at plasmic or genomic incompatibility between the two 20"-21 "C. Under such conditions, the generation time types of 'stock might be the major causes of the for all the clones retained was approximately 14 h. phenomena observed. Even though autogamous and Because the generation times were closely similar, it non-autogamous populations displayed signs of repro- could be argued that clones from A+ and A- stocks ductive isolation from each other, the tests for gene were not populations of synchronously aging cells, flow indicated that the threshold of an irreversible since the two types of stock differed in their life spans. evolutionary divergence had not yet been passed. However, this difference may be considered immate- Mating never occurred when the A- Stock ST was rial because all clones were used before they overcame repeatedly mixed with any of the foregoing set of the sexual immaturity period. Moreover, Nyberg stocks. Yet, it was able to mate with other stocks in (1978) showed that older cells of the ciliate Para- mixtures performed on the same day. On the basis of mecium tetraurelia are not necessarily more suscept- the mating tests, it may be argued that Stock ST ible to environmental stress, as is often assumed. belongs to a different biological species. The specimens employed in the experiments were also synchronized as to their cell-cycle stage. Since the experimental design was to test clones from the diffe- Cultivation Methods rent stocks together on the same day, a group of cells in the late stage of division (easily recognized by the The standard procedure used in culturing stocks has familiar peanut profile) was picked out of a log-phase already been described (Luporini and Dini, 1977). Fol- clonal culture of each stock with a braking pipette and lowing Heckmann (1963), stocks as well as their clonal separately transferred into fresh culture medium cultures, were grown in Erd-Schreiber seawater diluted by the addition of an equal volume of seawater. medium inoculated with the alga Dunaliella salina, The selection of each group of staged cells required a except that artificial seawater was used. The artificial few minutes and was accomplished l h before its seawater was prepared by dissolving Marine-Tropic employment in the experiment. This time span was New salt (Euraquarium s.p.a., Bologna) in slightly long enough to permit the morphogenetic events acidic, double-distilled water. It was always filtered which culminate in successful cell division. The order and pasteurized at 85 "C before use. Seawater parame- of the experimental employment of the different cell ters during the study were: pH = 8.2 -t 0.1; salinity groups followed the order of their isolation. In this way, = 31.4 + 0.2 %OS,dissolved oxygen from 7.1 to 8.5 mg owing to the similarity of the generation times, tested 1-1. specimens from the same as well as from different clones were all synchronized in the early stage of the cell cycle. Synchronization of Experimental Cells Bioassay Procedure In order to perform experiments with cells of similar age from the same as well as from different stocks, new Mercuric chloride of analytical quality (Fisher Scien- life-cycle stages of all the stocks used were contem- tific Company) was used to prepare a 10-3 M solution poraneously initiated by autogamy or conjugation. in artificial seawater. The procedure employed by (Age in is usually defined in terms of the Nyberg (1974, 1975) for heavy metal tolerance tests in number of divisions the organism has undergone since freshwater ciliates was followed with several modifica- its last sexual process.) Clones of both A+ stocks were tions. An appropriate amount of 10-3 M HgC12 was established from exautogamous individuals taken from diluted with seawater to produce the test solution of the original stock cultures. A- stocks were cloned the highest concentration used (1.26 W). The test sol- starting from exconjugant cells. They derived from utions of lower concentration were made by consecu- selfing pairs spontaneously occurring in the original tively diluting the next higher concentration with PB, and ST stock cultures (heterozygous at the mt 20.9 % seawater. In this way, the concentration range locus). Selfing did not occur in Stock G,, because it applied included a series of steps in concentration was homozygous at the mtlocus. In this case, following equally spaced on the log scale, in particular: roughly the procedure employed by Nobili (1967), exconjug- 1.26, 1.0, 0.80, 0.63, 0.40, 0.32, 0.25, 0.20, 0.16, and ants were obtained from homotypic pairs induced by 0.13 pM mercuric ion. (One micromole 1-l equals 0.2 198 Mar. Ecol. Prog. Ser. 4: 195-202, 1981

mg 1-l Hg+ +). For each stock tested, a different set of There was variability in survival responses of the cells consecutive steps in concentration of the foregoing to a given mercury concentration among the runs of the series was ordinarily used. In all cases the set same clone. Since such variability appeared to be employed covered the whole range of kills from zero to distributed normally I proceeded to pool the data 100 %. The actual concentration of the metal ion in the obtained at each concentration in each run over all solutions was not verified. Thus, it is only the 'nominal' runs. From the pooled survival responses at various rather than actual concentration. In view of the volatili- concentrations the 24-h median tolerance limit of the zation and surface adsorption phenomena occurring clone was estimated by probit analysis following Fin- with respect to mercury (Glickstein, 1979), the above- ney (1964). The median tolerance limit is an estimate mentioned concentrations represent overestimations. of the median lethal concentration (LC33), that is, the The pH of the test solutions did not appreciatively mercury concentration eliciting mortality in 50 % of differ from that of the seawater before addition of the the test organisms. metal ion. Just before the series of clones from different stocks were run through a mercury tolerance bioassay, new test solutions were prepared using the same beak- RESULTS ers as containers. The same series of clones was run various times on successive days. Mercury Tolerance of the Stocks In testing the resistance to mercury three single specimens showing normal swimming behaviour were The median tolerance limits of the clones bioassayed picked up with a braking pipette from a group of for each of the 5 stocks of Euplotescrassus are reported synchronized cells and transferred to each 0.5 m1 test in Table 2. The values are given in LLM1-I Hg++. In solution dispensed into a well of a three-spot hemi- addition to single clone data, Table 2 also gives the spheric depression slide. (Throughout this investiga- mean and standard deviation for each stock over all tion, the same wells for the same test solutions were clones. There were differences in mercury tolerance used.) Three replicates of each test concentration were among clones from the same stock. The comparisons always made, so that the toxic effect of mercury was between the variances of the stocks indicate, however, assessed on a total of 9 specimens for each concentra- that the variability inside the stocks is homogeneous: tion at each run of each clone. The specimens were the comparison between Stocks PB, and G, (which transferred into the test solutions starting with the showed the highest and lowest variance, respectively) controls and continuing towards the highest concentra- indicates that no significant difference exists (F3,2 = tions. (Each control well contained 0.5 m1 seawater 4.69, P > 0.05). The data of Table 2 also indicate that identical to that in the experimental wells, except that there was variability in the stock-averaged median it lacked the metal ion.) The amount of medium carried tolerance limits. The results of the one-way analysis of over with the cells at each transfer was less than one variance indicates that such variability is highly sig- hundredth of the volume in a depression (0.5 ml). The nificant (F, 13 = 151.34, P < 0.01). Tukey's test (Lison, resulting dilution of the test solutions was, however, 1961) was used to determine where significant differ- counterbalanced by the medium evaporation (about ences between stocks occurred. The results of statisti- 4 %) occurring during the exposure time of cells to cal comparisons indicate that there is no difference in mercury. The 3 depression slides containing the cells tolerance between the A+ Stocks G, and GvlI (from were incubated inside moist chambers at 20°-21 "C in inbreeding populations) but highly significant differ- the dark. After 24 5 l h the wells were checked and ences (P < 0.01) between these stocks and the A- scored for surviving cells. Cells unable to swim or Stocks ST, G,,, and PBI (from outbreeding popula- creep on the bottom of the well were regarded dead. tion~),whlch were about twice as tolerant to mercury

Table 2. Euplotes crassus.Mercury tolerance (LCSoin pM Hg++) of the clones bioassayed for each stock. Stocks designated by different symbols demonstrate significant differences in mercury tolerance, whereas no significant differences occur between

stocks designated by the same symbol. One pM 1-I equals 0.2 mg 1-IHg++

Clones Stocks

.G "11 'ST

1 0.257 0.264 0.520 0.674 0.665 2 0.236 0.301 0.502 0.615 0.636 3 0 272 0.269 0.570 0.680 0.7 10 4 0.298 0.630 0.621 Mean + S.D. 0.255 + 0.018 0.283 + 0.020 0.531 -t 0.035 0.650 f 0.032 0.658 + 0.039 Dini: Breeding systems and resistance to mercury in Euplotes 199

as the former. Among A- stocks, species-species dif- mating type. There is complete dominance among the ferences in mercury tolerance appear to occur. In fact, alleles (in our case mp.. . > mR. . . > mtl, see below), while no difference in tolerance is detectable between so that in each diploid cell carrying a combination of the conspecific Stocks G,, and PB,, statistically sig- two alleles at a time, only one of them is phenotypi- nificant differences (P < 0.05) exist between these cally expressed (Heckmann, 1964). The single locus a stocks and Stock ST belonging to a different biological with a pair of alleles (a+,a-) controls the ability to species. This could reflect the existence of actual dif- undergo autogamy; the dominant allele a+ permits the ferences in the outbreeding degrees of the two biologi- expression of the autogamy trait (Dini and Luporini, cal species. However, the variability recorded amongst 1980). Forty-five pairs were isolated from the GvrlXG,] A- stocks does not challenge the overall consistency of mixture, but only 41 of the possible 90 F, exconjugant the relationship of greater tolerance to mercury in clones, or 45 %, survived. Ten of these F, clones, which outbreeders. grew well with a similar fission rate, were selected and bioassayed during the sexual immaturity period. At variance with the procedure for the parental stocks, the Cross-Breeding Analysis set of consecutive steps in mercury concentration used usually included 3 lower concentrations (specifically: Since gene flow occurred between A+ and A- 0.10, 0.08, and 0.06 pM mercuric ion) beyond the last stocks, I sought to gain insight into the control system step of the series reported in the Materials and of the mercury tolerance differences observed. A cross Methods section. was made between the low-resistance (l.-r.) A+ Stock As I remarked earlier, homotypic pairing may occur Gwl and the high-resistance (h.-r.) A- Stock G,,. These in the mixture between Stocks G,,,, and G,,. However, stocks were selected because of their better mating the genetic markers in the two stocks made it possible reaction and higher viability rate after conjugation. to distinguish between clones from homotypic and The newly established clones - one from each stock - heterotypic pairs. When the 10 F, clones bioassayed tested for mercury tolerance were involved in the cross. reached sexual maturity, they were analysed for both Their phenotypes and genotypes at the unlinked mt mating type and autogamy trait. Four clones expressed (mating type) and a (autogamy) loci, proved to be the mating type P-W but no autogamy, whereas the other 6 same as the original stocks: Gvll,mating type P-y (m@/ clones displayed both traits, that is, their phenotype m@),A+ (a+/a-); G,,, mating type P-W (mP"/mP), A- was P-W,A+. (The F,clones not bioassayed showed the (a-/a-). (In Euplotes crassus, the mating type trait is same segregation pattern.) On the basis of the known under direct genetic control at one locus with a series genotypes of the parents, this is prima facie evidence of multiple alleles, each allele determining a different that the allele mP" occupies a higher rank in the domi-

Fig. 1. Euplotes crassus. Frequen- cy distributions of mercury toler- ance in pM Hg+ +) in F, and B, progenies. Arrows inside his- tograms indicate 24-h median to- lerance limits of the F, and B1 progeny clones bioassayed. In ad- dition, the mean and standard de- viation of the clone median toler- ance limits for each progeny and parental stock is given. Lines showing k 2 a have been drawn to show the approximate limits to the distribution of either pro- genies or parents Mar. Ecol. Prog. Ser. 4: 195-202, 1981

nance ladder than the mP allele, and suggests that the dered overall, the B, progeny tolerance levels showed first 4 clones may be derived from homotypic pairing, a distribution similar to the F,; that is, the backcross to whereas the last 6 are surely the outcome of cross- the 1.-r., parental Stock Gvrr did not result in any fertilization which occurred in heterotypic pairs (from tolerance difference with respect to the F, progeny pairing between cells of Gm, and G,,). Accordingly, (Fig. 1). After reaching sexual maturity, all the 30 B, they had to carry the genotype mP"/mP or mP/mtl, a+/ clones retained were analysed for both mating-type a-. Only these 6 F, clones were further considered. and autogamy trait. They fell into 4 phenotypic classes: They belonged to 3 synclones, i.e., they derived from 3 P-W, A+; P-y, A+; P-W, A-; P-y, A-. The respective pairs in which both partners survived after conjuga- frequencies observed, 13:8:1:6; are in agreement tion. This means that the 2 clones of each synclone (x2=4.76; 0.2 > P > 0.1) with a 3:3:1:1 ratio; two clones were of different cytoplasmic parentage, one deriving were unclassified. Given the genotypes of the Stocks from the l.-r. Gml, the other from h.-r. G,,. Under these Gwl and G,, and the order of dominance of the mt circumstances, one should expect each synclone to alleles reported above, these are the expected results show the same parental distinctions when these last when a true hybrid F, clone (mating type P-W [mP/ are inherited via cytoplasm. The frequency distribu- mP], A+ [a+/aP]) is backcrossed with the parental tion of the 24-h median tolerance limits estimated for Stock GWl(mating type P-y [mPlmrJ],A+ [a+/a-I). This the F, clones are reported in Figure 1, along with the indicates that cross-fertilization had indeed taken mean and standard deviation for the F, over all clones. place at backcross. Unfortunately, the low mating reac- The F, average median tolerance limit was lower than tivity and low survival after sexual reorganization pre- that of the l.-r., parental Stock GWl.As shown in Figure cluded using B, clones in further breeding experi- 1, F, clones of all synclones became alike with respect ments. to mercury tolerance despite their cytoplasmic descent In spite of its incompleteness, the cross-breeding from differentiated parents. Since synclones derived analysis is evidence that the mercury tolerance differ- from conjugating partners between which a massive ences existing between the Stocks A+ GvIl and A- Gll cytoplasmic exchange did not apparently occur, a cyto- are (1) determined by nuclear differences, and have (2) plasmic inheritance of the trait in question appears no simple genetic basis. inconsequential. Yet, the disappearance of the paren- tal distinctions, and the acquisition of peculiar charac- teristics by the F, clones in the presence of cross- DISCUSSION fertilization point towards a significant role of the nucleus in the maintenance of the parental differences The major conclusion of this study is that non-auto- in mercury tolerance. Finally, it may be noteworthy gamous (A-) stocks of Euplotes crassus, representa- that the 4 F, clones, which were not considered tives of populations following an outbreeding strategy, because they might be from homotypic pairing, show a higher resistance to mercury than autogamous behaved similarly to the 6 true hybrid clones with (A+) stocks from inbreeding populations. In this con- respect to mercury tolerance. This indicates that all the text, consider the results obtained by Persoone and 10 F, clones bioassayed were the outcome of cross- Uyttersprot (1975). These authors investigated the fertilization. effects of different mercury concentrations on a North The mating type uniformity of the F1 progeny pre- European A- multiclonal stock of E. vannus (a typolog- sented a constraint on further cross-breeding analysis. ically-defined species whose separateness from the The results to be presented concern the sole type of closely related morphospecies E. crassus is based on cross possible to make, that is, the backcross of a true historical grounds only; Genermont et al., 1976; Gates, hybrid F, clone (mating type P-W, A+) to the 1.-r., 1978). They found that all specimens survived in parental Stock GvlI (mating type P-y, A+). Among the 0.1 mg 1-I Hg++ (=OS0 m).This concentration cor- true hybrids, the F, clone showing the highest toler- responds to the lowest mercury concentration which ance level was chosen. Owing to the low mating reac- killed all the test organisms in my A+ stocks, whereas tion occurring whenever A+ animals are involved in it is close to the highest concentration (0.07 mg 1-I the mixtures (Luporini and Dini, 1977), only 28 pairs Hg+ + = 0.32 pM) killing no test organism in the A- were isolated and 30 of the possible 56 B1 exconjugant stocks G,, and PB,. In the light of these considerations, clones, or 53 %, survived. All of the 30 B, progeny it is a reasonable inference that the A-, multiclonal clones were retained, but only 10 healthily growing stock tested by Persoone and Uyttersprot had to show a clones with similar fission rates were scored as to mercury tolerance level significantly higher than that mercury tolerance during the sexual immaturity of At stocks. Consequently, my conclusion of the rela- period. There were 3 synclones, plus 4 additional tionship of greater tolerance to mercury in A- stocks is clones whose partners failed to produce a clone. Consi- strengthened. Dini: Breeding systems and resi stance to mercury in Euplotes 201

Adaptation to concentrations of metals which are tolerance may result. Since the higher resistance of the normally toxic is a well-documented phenomenon in A- stocks is assumed to be due to a peculiar, internal various organisms (e.g. the reviews by Ashida, 1965 genetic organization as a consequence of an outbreed- and Bradshaw, 1971). Among ciliates, Tingle et al. ing strategy, it may be inferred that their resistance to (1973) obtained cytological indications for adaptation mercuric ion is only an aspect of a more general prop- to mercuric chloride in the ciliate Tetrahymena erty which allows outbreeders to better withstand pyriforrnis. It might thus be thought that the higher environmental stress. Indeed, the similar responses to resistance of the A- stocks of Euplotes is of adapta- mercury shown by A- stocks of different biological tional origin. Such supposition appears to be unlikely species (Tables 1 and 2), and their common ability to since, except perhaps for the Stock PB,, the other A- resist lower temperature and pH than A+ stocks stocks were collected from locations which cannot (unpubl. observations) provide some support for this reasdnably be suspected of having high levels of mer- argument. curic ion; this is an assumption partially supported by The existence of a relationship between the breed- the studies of Piro and Rossi (1979). Furthermore, the ing system of a species and its resistance to environ- A- stock GI1 was collected along the Somalian coasts mental stress is not without precedent in the literature. from the same location as the A+ stocks, and it is Nyberg (1974) studying different species of freshwater therefore reasonable to assume that all these stocks ciliates, belonging to aurelia, P, mul- were subjected to the same environmental stress. timicronucleatum, and Tetrahymena pyriformis com- The results of the cross-breeding analysis suggest plexes, noted an impressive consistency of the rela- that the mercury tolerance differences recorded bet- tionship of greater tolerance to various stressing agents ween the A+ G",! and A- G,, stock do not have a in outbreeders. Nyberg's and the present results both simple genetic basis. Two major explanations of the point to a greater ability of outbreeding forms of cili- marked decrease in the degree of resistance shown by ates to cope directly with changing conditions. the F, and B, progenies may be considered. One expla- These findings may be used in making a prediction nation holds that multiple loci, each having an indi- and generalization: environments subjected to sudden vidual share in the total effect, are controlling mercury changes are expected to be principally colonized by tolerance. If this is so, then we would have to assume outbreeding forms. The key to the acceptance of the that (1) the crossed parents were heterozygous at a extrapolation to more complicated organisms is (1) that number of loci, and (2) by chance all the progeny ciliates are facing the same challenges to their survival clones tested carried genotypes constituted by combi- as are other organisms, and (2) the considerable nations of genes specifying low resistance to an extent amount of evidence indicating that ecological larger than the low-resistance (1.-r.) Parent Gvrr responses of Protozoa show many of the same complex Though a scanty number of progeny clones was phenomena considered to be characteristic of the more examined, the overlapping distribution of the F, and B1 advanced organisms. tolerances and their sharp downward shift with respect to the 1.-r. parent, render this explanation doubtful. Acknowledgements. I would like to thank Dr. M. A. Gates As it has been stressed previously (Materials and (University of Toronto), Dr. P. Luporini (Universita di Methods section), there are strong indications of a Camerino), and Dr. R. Nobili (Universita di Pisa) for reading the manuscript and making helpful suggestions. Assistance reduced gene flow between A+ and A- stocks which, and suggestions regarding the statistical elaboration of data indeed, follow different ecogenetic strategies; for in- were provided by Dr. M. Marchi (Universita di Pisa) and Dr. J. stance, at variance with A- stocks, A+ stocks do not J. B. Smith (University of Toronto), to whom I am grateful. I apparently rely on recombinational variety to meet also want to thank Dr. J. Berger (University of Toronto) and Dr. D. Nyberg (University of Illinois) whom I talked to about environmental challenges. This calls for an evolutio- this problem. nary divergence and changes in internal genetic organization of the 2 types of stocks. In the light of these arguments a second type of explanation, perhaps more likely, is set forth: The different degrees of sus- ceptibility to mercuric ion shown by the representa- LITERATURE CITED tives of autogamous (inbreeding) and non-autogamous Ashida, J. (1965). Adaptation of fungi to metal toxicants. A. (outbreeding) populations are the expression of diffe- Rev. Phytopath. 3: 153-174 rent characteristics of their gene complexes as a whole. Bradshaw. A. D. (1971). Plant evolution in extreme environ- Genes in each type of stock form harmonious combina- ments. In: Creed, R. (ed.) Ecological genetics and evolu- tion. Blackwell, Oxford, pp. 38-81 tions because they have become coadapted by natural Cairns, J., Jr. (1974). Protozoans (Protozoa). In: Fuller, J., Hart, selection. In the hybrids, such coadapted gene com- W. (eds) Pollution ecology of freshwater invertebrates. plexes are broken up, and a downward shift of mercury Academic Press, Inc., New York, pp. 1-28 202 Mar. Ecol. Prog. Ser. 4: 195-202, 1981

Dini, F., Bertini, B. (1978).Autogamy-induced effects in aging din) (Ciliata Hypotrichida). Monitore zool. ital. (N.S.) 11: clones of Euplotes crassus (Clliata, Hypotrichida). Boll. 119-154 Zool. 45: 208 Nanney, D. L. (1980). Experimental Cillatology: An introduc- Dini, F., Luporini, P. (1980). Genic control of the autogamy tion to genetic and developmental analysis in ciliates, tralt in the hypotrich ciliate. Euplotes crassus. Genet. Res. John Wiley & Sons, New York 35: 107-1 19 Nobili, R. (1967). Induzione di autoconiugazione in mes- Ford, E. B. (1971). Ecological genetics, Chaprnan and Hall, colanze interspecifiche di Euplotes. (Ciliata Hypot- London. richida). Arch. Zool. Ital. 52: 51-64 Gates. M. A. (1978). Morphometric variation in the hypotrich Nyberg, D. (1974). Breeding systems and resistance to ciliate genus Euplotes. J. Protozool. 25: 338-350 environmental stress in ciliates. Evolution 28: 367-380 Genermont, J., Machelon, V., Tuffrau, M. (1976). DonnCes Nyberg, D. (1975). Genetic analysis of copper resistance in experimentales relatives dans le genre Euplotes (Cilies Paramecium aurelia syngen 4. Genetics 80: 463-473 Hypotriches). Protistologica 12: 239-248 Nyberg, D. (1978). Copper tolerance and segregation distor- Glickstein, N. (1979). The potential loss of dissolved mercury tion in aged Paramecium. Exp. Geront. 13: 431-437 and selenium in marine experimentation. Mar. Pollut. Bull. 10: 157 Persoone, G., Uyttersprot, G. (1975). The influence of organic Heckmann, K. (1963). Paarungssystem und genabhangige and inorganic pollutants on the rate of of a Paarungstypdifferenzierung bei dem hypotrichen Ciliaten marine hypotrichous ciliate: Euplotes vannusMuller. Rev. Intern. OcCanogr. MBd. 37: 125-151 Euplotes vannus 0. F. Muller. Arch. Frotistenk. 106: 393-421 Piro, A., Rossi, G. (1979). Determinazione di cadmio e mer- Heckmann, K. (1964). Experimentelle Untersuchungen an curio nel Mar Ligure. Atti del Congresso P. F. Oceanog- Euplotes crassus I. Paamngssystem, Konjugation und rafia e Fondi Marini, Roma, pp. 1-13 Determination der Paarungstypen. Z. Vererbungsl. 95: Sonneborn, T. M. (1957). Breeding systems, reproductive 114-124 methods, and species problems in protozoa. In: Mayr, E. Lison, L. (1961). Statistica applicata alla Biologia Sperimen- (ed.) The species Problems. Am. Ass. Advrnt Sci. Publ., tale (Ital. Edit.), Casa Editrice Ambrosiana. Milano Washington D.C.. pp. 155-324 Luporini, P,, Dini, F. (1977). The breeding system and the Tingle, L. E., Pavlat, W. A., Cameron, I. L. (1973). Sublethal genetic relationship between autogamous and non-auto- cytotoxic effects of mercury chloride on the ciliate Tet- gamous sympatric populations of Euplotes crassus (Dujar- rahymena pyriformis. J.Protozool. 20: 301-304

This paper was presented by Professor B. Battaglia; it was accepted for printing on December 2, 1980