zyxwvuts zyxw Ethology 81, 332-343 (1989) 0 1989 Paul Parey Scientific Publishers, Berlin and Hamburg ISSN 0179-1613 Commentary Evolutionary Biology Research Group, Department of Environmental and Evolutiona y Biology, University of Liverpool, Liverpool zyxwv Nuptial Feeding in Insects: Mating Effort versus Paternal Investment L. W. SIMMONS & G. A. PARKER This Commentary concerns the relative importance of mating effort and paternal investment in the evolution of nuptial gifts by males to females, in insects. Mating effort is “that proportion of reproductive effort expended in finding a member of the opposite sex or in subduing members of the same sex in order to 1972, and formally defined by LOW1978; see also ALEXANDER mate” (c.f. TRIVERS & BORGIA1979). Paternal investment is “any investment by the parent in an offspring that increases the offspring’s chance of surviving at the cost of the parent’s ability to invest in other offspring” (TRIVERS 1972). Although it does not strictly accord with TRIVERS’ precise definition (WICKLER 1985), we term any increase in a given male’s total surviving progeny by increasing the reproductive output of a given female as paternal investment. This accords with LOW’S(1978) division of reproductive effort into mating and parental effort. LOW’Sparental effort is distinct from TRIVERS’ parental investment; it is that proportion of reproductive effort devoted to the production of progeny as a whole, whereas parental investment is the proportion of parental effort devoted to an individual offspring. Parental effort is the total reproductive effort devoted to parental investment, irrespective of the individual offspring receiving it. Despite TRIVERS’ careful definition of parental investment, he used the term to mean either parental investment or effort as defined above (ALEXANDER & BORGIA1979). Where a male gains via increasing an individual female’s gametic output through the donation of a nuptial gift, he is thus investing paternally. In contrast, if a male gains by increasing the proportion of eggs he fertilizes f r o m a given female or by increased mating opportunities, this will be considered mating effort. U.S. Copyrighr Clearance Center Code Statement: 01 79-1613/89/8104-0332$02.50/0 zy zyxwv Nuptial Feeding in Insects: Mating Effort versus Paternal Investment 333 Both mating effort and paternal investment seem perfectly plausible candidates for either the origin and/or maintenance of nuptial gift donation. A controversy has arisen over which of these selective pressures has exerted the greater influence on the gift donation characteristic (WICKLER1985, 1986; GWYNNE 1986 a; SAKALUK 1986 a). We first distinguish between the evolutionary origin of a characteristic, and its maintenance once fixed in a population. To deduce the origin of a character, we consider the selective forces that allow a mutant gene for a particular characteristic to spread in a population fixed for some ancestral state, prescribed by an alternative allele to the mutation. For the maintenance of a character, we try to estimate why the mutation, now fixed in the population, is stable against invasion by the ancestral allele, or by some new alternative. Selection pressures responsible for the origin of a character need not be equivalent to those responsible for its maintenance, for several reasons: First, fitnesses of the two alternative states (ancestral; new) may be frequency dependent. Second, there may be frequency dependence in that the set of alleles that could invade the ancestral condition may be different from the set of alleles that can invade the new condition. Since the spread of any character is likely to promote a series of changes in others (especially in games of couriteradaptation), the force acting to maintain a character could ultimately bear little relationship to the force through which it originated. This distinction is not new; we stress it because it is important in the present debate. Others writing on the same topic have used different terms: QUINN & SAKALUK’S (1986) ‘primary function’ = origin and ‘effect’ = maintenance; GWYNNE’S (1986a) ‘evolved function’ = origin and ‘consequence’ = maintenance. The idea that nuptial gifts may function to generate more or better progeny is an old one. DARWIN (1871) considered the “expenditure of power” by parents as the limit on their fertility, and it was FISHER(1930) who first defined parental care as “expenditure in the form of nutriment, effort, or exposure to danger, incurred in the production and nurture of the young”. In his study of the Empidinae, DOWNES (1970) argued that prey donation by males probably evolved in the context of female nutrition; females rely entirely on male donations for ovarian maturation. TRIVERS (1972) himself considered the donation of food items by insects as a way in which males invest paternally although, as pointed out by PARKER (1979), gift donation can be paternal investment only if the donating male’s progeny benefit from the gift. The notion that gifts may relate to sexual selection and enhanced mating opportunities goes back to DARWIN (1871), who suggested that by pairing with more “attractive and vigorous males”, females would rear greater numbers of progeny, particularly “if the male defends the female, and aids in providing food for the young”. In one of the first major discussions of gifts in insects, THORNHILL (1976 a) suggested an evolution through intersexual selection, through female choice of males providing larger paternal investments during copulation. ALEXANDER & BORGIA(1979) considered that donations such as food, territories or protection, originally considered by TRIVERS as parental investment, would best be considered mating effort because they evolved to attract and persuade females zyxwv 334 zyxwvu zyxwvu zy L. W. SIMMONS& G. A. PARKER to mate. The use of gifts in female nutrition and egg production would then be incidental to the primary force on males, to fertilize eggs. GWYNNE(1984) suggested that mating effort be divided into promiscuous mating effort (aimed simply at acquiring fertilizations) and non-promiscuous mating effort (which reduces female reproductive costs and/or benefits the offspring while reducing the male’s ability to obtain mates). QUINN & SAKALUK (1986), however, claim that the original definition of parental investment has become distorted by a change in emphasis from the effect of male effort on offspring survival (maintenance), to an emphasis on its origin as propounded by ALEXANDER & BORGIA (1979). QUINN & SAKALUK argued that it is the incidental effects of investment that are critical in determining whether gifts be categorized as mating effort or parental investment. zyx The most detailed consideration of gifts in relation to mating effort is that of WICKLER (1985), who argues that nuptial gifts both originated and are currently maintained through mating effort, rather than paternal investment. H e argues as follows: 1) Sperm competition (PARKER 1970a) acts to increase sperm numbers, so that there is an excess over that required by the female for fertilizing the eggs. 2 ) Females will then become selected to metabolize excess sperm, leaving males evolutionarily “trapped” into maintaining high sperm counts. 3) Selection eventually favours female preference for males with better gifts. This leads to the female choice spiral (FISHER1930), so that males evolve enhanced quantity and quality of gifts. WICKLER argues that because of last male sperm precedence (PARKER 1970 a), males have restricted benefit through paternal investment. The delay between mating and incorporating the gift into the eggs means that the gift will go to the progeny of another male (see also WICKLER & SEIBT 1985); it is hence “pseudo-parental’’ investment. H e therefore argues that mating effort, not paternal investment, has been responsible for the origin and maintenance of nuptial gifts. H e suggests that the male may benefit by increasing female survival if (1) his chances to sire her future offspring also increase, or ( 2 ) “indirectly, if it prevents a future increase in male competition due to a shrinking number of surviving females”. Whilst we fully accept (I) and have examined conditions under which this effect can lead to the origin of gifts (PARKER & SIMMONS 1989), we cannot envisage that ( 2 ) could exert a significant selective pressure on male behaviour in large, randomly-mating populations. A male that gives a gift has so small a chance of meeting the same female again that, in the absence of any direct gain [such as (1) or related effects] the gift should always be used by the male to enhance his own survival. We feel that WICKLER’S (1985) evolutionary pathway for the origin and maintenance of gift donation is plausible, and can see no reason why stages 13-) could not operate commonly in many species. However, it is explicit throughout stages 2) and 3) of this pathway that male-derived resources are being used by females to enhance reproduction. Whether paternal investment can also be important will then depend on: (i) the pattern of allocation of the gift resource to the progeny; and on (ii) the sperm precedence pattern. Indeed, both features have 1985; been considered important by all the contributors to the debate (WICKLER zy zyxw 335 Nuptial Feeding in Insects: Mating Effort versus Paternal Investment GWYNNE1986a; SAKALUK 1986a), and in our theoretical models (PARKER& SIMMONS 1989). SAKALUK (1986 a) criticized WICKLER’S argument against a paternal investment function of male gifts, pointing out that last male sperm precedence is not the case in some species, and when sperm mix in storage, the donating male can fertilize some of the eggs which he helps to produce. The literature on rates of incorporation of gift nutrients into eggs is reviewed in Table 1, which also (if known) gives P2, the proportion of eggs fertilized by the last male to mate & PARKER 1976). Some of these data are very recent andwere not (BOORMAN accessible to WICKLER (1985). The available data now suggest that incorporation can be rapid, for most species within 24-48 hr. Further, in some species, increased egg output occurs over the same time period (e.g. Drosophila spp. STEELE1986; Colzus eurytheme RUTOWSKI et al. 1987; for review see PARKER & SIMMONS 1989). Therefore, even in species with high P2 the nurturant male may fertilize eggs to which he contributes. We envisage that males will be selected to ensure rapid incorporation of their donations into eggs which they can fertilize (PARKER & SIMMONS 1989). We interpret substances in the ejaculate which both stimulate vitellogenesis and induce a refractory period in the mated female, as a means by which males can achieve these fertilizations. In Requenu verticalis, the slow nutrient incorporation is associated with a complete first male advantage at fertilization, a 5-day refractory period, and a mating bias for virgin females (GWYNNE 1986b; GWYNNE1989). This ensures that the nurturant male fertilizes the eggs. Mechanisms which reduce the chance of a second male fertilizing eggs derived from nutrients of the first male are also found in D. mojuvensis (MARKOW 1988). Both WICKLER (1985) and SAKALUK (1986a) claim that a shift in the pattern of sperm precedence may result through female adaptation. WICKLER states that “in zyx zyxw zyxwvutsr Table 1: Comparison of the rate at which male-donated nutrients are incorporated into developing oocytes with the degree of last male sperm precedence (P2) Species Diptera Drosophila melanogaster D. pseudobscura D. subobscura D. mojavensis Lepidoptera Colias eurytheme Plodia interpunctella Orthoptera Requena verticalis PI 0.83-0.99 0.79 Rate of incorporation Reference 24 h 24 h 24-48 h 24-48 h 1, 2, 3 3 4 5, 6 < 24 h 1 0.96 2-8 h 8, 9 7 0 9-13 days 10,ll 1 BOORMAN & PARKER1976; 2 GROMKOet al. 1984; 3 BOWNES& PARTRIDGE 1987; 4 STEELE 1986; 5 MARKOW& ANKNEY1984; 6 MARKOW1988; 7 BOGGS& WATT 1981; 8 BROWER1975; 9 GREENFIELD 1981; 10 BOWENet a]. 1984; 11 GWYNNE1989. 336 zyxwvu zyxwvu L. W. SIMMONS& G. A . PARKER species with mating presents and repeated matings, selection on females favours last mating male’s sperm precedence, encouraging males to continue courting and transfer nutrients”. In contrast SAKALUK claims that nuptial gifts may impose selection on females to favour a mixed sperm utilization pattern; allowing sperm to indulge in the raffle principle (PARKER 1982) “would allow females to reduce the number of sperm transferred by males who provide inadequate nuptial gifts, thereby reducing the number of eggs fertilized by such males, and to reward more generous males by permitting fuller insemination”. Both authors could be accused of committing what we term a “sequence fallacy” - a kind of evolutionary tautology that is not uncommon in sociobiological speculations, and which relates to the “adaptive valley” problem (WRIGHT 1932). Imagine two characters, A and B. Both have some implicit cost in the absence of the other, even if this relates only to a shift from some previous optimum. However, if A is fixed, selection favours the addition of B, and if B is fixed, selection favours addition of A. The fallacy is to assume that AB will automatically evolve, without impediment. Only if both characteristics are above their respective threshold frequencies in the population can selection push both to fixation. If, however, both exist only as rare mutations, selection will act against both A and B rendering an AB state impossible unless and until such time as genetic drift can push both A and B above their respective thresholds. In WICKLER’S case, A could be the last male precedence pattern (envisaged as a female adaptation). This is favoured because it will favour the evolution of B greater courtship persistence and nutrient transfer by males. But, in turn, B requires that A is present at high frequency before it can spread by selection. WICKLER’S argument might be retrieved if he were to envisage that gifts are more likely to evolve where, for some other reason, a last male sperm precedence pattern has already evolved. Alternatively, in some insects, the female simply allows the male to mate while she eats the gift (e.g. THORNHILL 1976b). This leads to increased sperm precedence. In females, selection favours as much nutritive benefit as possible; this generates high sperm precedence because the male has more time to transfer sperm. More matings will now be with non-virgins, since mated females will become receptive. Thus for males, larger gifts will allow increased sperm transfer and increased benefits in sperm competition. SAKALUK’S case is parallel; A might be the mixed sperm usage pattern, and B the gift donation habit of males. The tautology is perhaps here more complete because of the terms “reward” and “reduce” - which imply that it pays a female to contrive an increase or reduction in progeny to a male because this will facilitate evolution of male behaviour in a direction that ultimately benefits female interests. Again, SAKALUK’S arguments might be more appropriate if paternal investment were envisaged more likely to evolve in species which, for some other reason, already showed sperm mixing. O u r own view of the effects of gifts on the female’s sperm usage pattern is that gift donation would greatly favour a change from unreceptivity after mating, to receptivity to most or all males, since gifts offer a real advantage in multiple mating to the female (benefits in the absence of gifts, and depending purely on “good genes” arguments, are the subject of some debate, see PARKER 1984; see zyxwvuts Nuptial Feeding in Insects: Mating Effort versus Paternal Investment 337 KNOWLTON & GREENWELL 1984; EBERHARD 1985 for some opposing views). If the female mates often during her life, rather than seldom, this might result in some changes in her sperm storage organs and sperm usage patterns. For instance, it may be optimal for the female to reduce the size and hence cost of her sperm storage organs. Alternatively, should females benefit from nutrients absorbed from the seminal fluid, selection may increase spermathecal size in order to collect greater quantities of utilizable ejaculates. In several gryllids, the spermatheca is an unusual elastic structure which continues to expand to store all sperm offered (e.g. LOHER& RENCE1978; SAKALUK 198613; SIMMONS 1986, 1987) and is the absorption site for substances which contribute to female fecundity (BENTUR& MATHAD1977; SIMMONS 1988). Such adaptations could lead to sperm mixing because of the large quantities of sperm stored (e.g. SAKALUK 1986b; SIMMONS 1987). In general, however, we feel that the sperm precedence pattern is more likely to be a reflection of male adaptation (PARKER1970a) than of female adaptation (WALKER 1980) because the selective forces acting on males are likely to be much stronger than the selective forces acting on females, although it is not inconceivable that females can capitalize on the sperm precedence pattern (e.g. THORNHILL & ALCOCK1983; SIMMONS 1987) - for a fuller discussion, see PARKER (1984). WICKLER (1985) claims that if selection on females has optimized egg size, females would use male nutrients to increase egg number. H e argues that selection on males would, in contrast, favour increased size of the eggs he actually fertilizes, rather than an increase in the number of eggs he does not. We think that WICKLER’S implicit points (that gifts should not cause a deviation from the female’s optimum towards male interests, and that there could be sexual conflict in the pattern of allocation of the gift) are rather important because they generate a more sophisticated debate. However, our own analyses (PARKER & SIMMONS 1989) do not lead to quite the same conclusions. We find, using a marginal value approach, that small or moderate-sized gifts are likely to cause the female to reduce the gametic output (= egg size x egg number) at the next batch, though she will produce the batch more quickly. In a sense, this is even better for WICKLER’S argument against the role of paternal investment in the origin of gifts; it means that a male mating early in a female’s reproductive cycle might sire fewer, or less fit offspring at the next batch, than if he were to give no gift at all. H e may gain, however, since the gift would bring about a more rapid production of the next batch - which would reduce risks to his future eggs (WICKLER 1985), and may also reduce the risk that his sperm is displaced by another male. Whether these benefits could outweigh the costs to the male (via the female’s shift in optimal gametic output) remains to be estimated. We find that sexual conflict over the allocation of the gift resources could be extensive when gifts are rare; in general when a male’s paternity shows a decline with subsequent batches, he will profit from a more immediate use of the gift than would be optimal for the female. We interpret the fact that male accessory gland fluids commonly act as oviposition stimulants as male manipulation of female reproduction. If gifts are a regular and predictable feature of the female’s reproductive cycle (I gift, 1 mating, per cycle) and if paternity declines in zyxwv zyxwv 338 zyxwvu zyxwvu zyxwv L. W. SIMMONS& G. A. PARKER subsequent batches, the ESS is to have all of the gift used in the batch following mating, assuming nutrients can be transferred as quickly as this - even if sperm precedence is only > 50 %. There should be no obvious sexual conflict over this allocation pattern, so females will not necessarily be expected to show counteradaptations to male stimulants once gift donation becomes fixed. O u r proposed evolutionary pathway is related to WICKLER’S, but takes into account changes in female receptivity, the outcome of sexual conflict, the likely changes in the female’s reproductive allocation during the evolution of gift donation, and the findings of our recent models (PARKER & SIMMONS 1989). First, we differentiate between four types of gift: (1) Prey gifts - the gift is a prey (or other food item, including obvious ritualizatians such as the “empty” prey presents as found in certain empidids) collected by the male. (2) Seminal ggts - the female channels male-derived substances related to sperm production (e.g. sperm, accessory gland fluids, parts of the spermatophore, etc.) into gamete production. (3) Somatic gifts - part of the male’s body is eaten; usually a specialized organ or its products (e.g. metanotal glands, salivary secretions, etc.), but does not typically result in death. (4) Suicide gifts - the female eats most or part of the male during or after (and sometimes even before) copulation, and there in no sign that the male attempts to escape; the death of the male is certain if the female receives the gift (e.g. certain mantids and arachnids). We see no obvious reason why either the origins, or the reasons for maintenance of these different gift types need be identical. In fact, we see a clear distinction between prey and suicide gifts on the one hand, and seminal and somatic gifts on the other. With prey and suicide gifts, a large amount of nutrient resource would have been available at the origin of their development. Seminal and somatic gifts are likely to have evolved by gradual specialization, and hence by gradual specialization from an initially marginal magnitude. We therefore argue that whilst all forms of gift could now be maintained, at least in part, by paternal investment, prey and suicide gifts are much more likely to have had a significant paternal investment origin component than seminal or somatic gifts. Prey gifts may nevertheless have had a mating effort origin. In a species in which at least originally both sexes were predacious, a male might gain extra mating opportunities (e.g. by being able to mate with a mated female that would normally be unreceptive) by conveying a gift which rendered the female passive and hence accessible to sperm transfer, at least during the time she took to eat the gift. This would require that males were able to respond sexually to female cues during the time that they were capturing o r consuming prey items, and that there was some genetic variance in ability to switch motivation between feeding and sex. As a bonus, some parental investment benefit may have accrued to the early gift donating mutants, especially if ancestral females were typically unreceptive after mating, allowing better chances of the gift being used for the mutants’ progeny. DOWNES (1970) envisaged that some Empidinae prey gifts might have originated through parental investment. Imagine a mutant that simultaneously zyxwvut zyxwvu zyxw Nuptial Feeding in Insects: Mating Effort versus Paternal Investment 339 allows novel mating access and increases progeny by paternal investment. The stronger force in its spread would be mating effort rather than parental investment, unless the gift approximately doubles the expected number of progeny to the male via the female. In one sense, the mutant relates entirely to mating effort, since any additional bonus due to parental investment, however large, is not felt unless mating access is gained first (c.f. ALEXANDER & BORGIA1979). A female choice origin for prey gifts also seems plausible. Females that preferred males with prey could have experienced a selective advantage over females that mated randomly; prey donation could thus have originated by a Fisherian process (mating effort). Once the habit had fixed in the population, females might not mate indiscriminately, whenever possible, in order to maximize the total protein intake rate. The fact that females of certain insects still exercise mate choice related to the prey quality (see THORNHILL & ALCOCK1983) probably relates to the optimal diet model of optimal foraging theory. Depending on the interval between successive encounters with males (travel time costs), a gift type would fall out of the optimal set (and should hence be rejected) if its total energy uptake/handling time is lower than the average expected energy uptake rate for “foraging” as a whole (see e.g. KREBS1978). Models of optimal mate choice in conditions of varying mate quality have been proposed by JANETOS (1980) and PARKER (1979, 1983); in the case of nuptial gifts the equivalence between the optimal diet models and the optimal mate choice models becomes an exact one. An alternative origin for prey gifts would be the case where the gift did not create any extra mating opportunities, but simply increased the number of progeny sired by the gift-giving male. There are two possibilities for such an increase. The first relies on increasing the female’s gametic output as a result of the gift. We would term this paternal investment, remembering that our definition of parental investment is wider than that originally given by TRIVERS (see above). The second type of increase relies on a change in sperm precedence pattern as a result of the gift. That nuptial gifts can increase sperm precedence by allowing increased sperm transfer has been demonstrated by THORNHILL (1976 b). We would term this component mating effort. Somatic gifts such as specialized salivary, metanotal, or accessory gland secretions may also have originated by increasing male access to otherwise unreceptive females, rendering them temporarily passive whilst genital contact was obtained. Perhaps female choice was more likely, leading to FISHER’S spiral. Parental investment probably exerted less influence; in the ancestral state the amount of resource transferred would have been marginal. Seminal gifts may have had the origin proposed by WICKLER(1985); i.e. via phagocytosis of sperm, uptake of accessory gland and spermatophore products, etc. This relies on mating effort - products of male competition are capitalized upon by females; males increase efforts to regain advantages. However, we see a difference between the effects of female use of sperm, and female use of plugs. If female consumption of sperm affects each competing ejaculate equally, it is not clear that this will affect the optimal expenditure on sperm, since it will not affect the relative gains of the different males at 340 zyxwvu zyxwvu L. W. SIMMONS& G. A. PARKER fertilization. The optimal expenditure might, however, be affected if the ejaculate of the first male to mate is significantly eroded before the second male mates. For plugs, the effects of erosion by females may be more severe, often more readily leading to the sort of escalation proposed by WICKLER. Firstly, plugs may conflict with female interests; it may pay females to erode plugs quickly for reasons other than just nutrient benefit. This would lead to male increases in plug size, and to the sort of escalation WICKLER envisaged. Females usually possess mechanisms for digesting the spermatophore and associated structures. If these prevent other males from mating, it is easy to envisage a male-female arms race leading to spermatophore/plug enlargement, and extensive nurturant usage by females. We cannot see a clear argument in favour of a paternal investment origin of seminal gifts: the nutrient transfer would have been marginal initially. Paternal investment might, however, be significant in the muintenunce of such gifts, because of the escalation of resources donated and because of selection on males to favour swift usage of the resources. The strongest case for a paternal investment origin may relate to suicide gifts. If females mate only once, and if a male’s prospects of finding a second mate are poor, selection may favour suicide as a form of paternal investment. Such an origin (and maintenance) seems likely where a male is consumed willingly during, or especially after, copulation. For there to be no sexual conflict over this form of cannibalism (i.e. a true suicide gift), increased fitness (or number) of offspring via the cannibalistic mate must exceed the expected value of searching for extra matings (PARKER 1979). However, in many instances, the male attempts to avoid cannibalism (e.g. LISKE& DAVIS 1984), indicating that there is sexual conflict. Where males are eaten before sperm transfer, there must always be sexual conflict. Where there is no paternal care, it will pay females to eat males. MANNING (1966) proposed that somatic gifts and prey gifts evolved as appeasement mechanisms, to prevent the female eating the male. If paternal investment has generally been a less significant component than mating effort in the origin of seminal, somatic, and possibly also prey gifts, this need not necessarily apply to their maintenance. During specialization of gift donation there would generally be: 1. Selection via female choice for increased size of the gift given (for suicide and prey gifts, such effects would be much less extensive), resulting in greater costs to the male. 2. Changes in the female’s foraging pattern. Consider a prey gift. Initially, when gifts were unpredictable, if females could not alter their foraging pattern facultatively in relation to resources accrued, the initial effect of receiving prey may have been to increase female gametic output following gifts, though we envisage that this might be temporary only (PARKER & SIMMONS1989). Females may later adapt by showing a reduction in the next clutch output, which may reduce the advantages to males of prey donation. Significant sexual conflict might exist over how the gift is used. 3. Whether gift donation then went extinct, or became exaggerated, would depend on a balance between the trade-offs in males of not giving gifts (using zyx 341 zyxwvu zyxwvu Nuptial Feeding in Insects: Mating Effort versus Paternal Investment benefit to gain extra survivorship or extra mating activity), and the intensity of female preference for the gift-donating males. 4. Selection for agents in males that manipulate female physiology into making more immediate use of the gift, depending on the sperm precedence pattern (PARKER & SIMMONS 1989). 5 . Later, as gifts became a regular feature of the female reproductive cycle, females may not conflict over how a gift is allocated between the subsequent egg batches. For all cases where P2 > 0.50 (i.6:. most insects) selection on males & SIMMONS1989). There would now be favours immediate gift use (PARKER nothing to prevent male manipulation so that as much of a gift as possible is used immediately after mating. 6. O n e of the most major changes during the evolution of gift donation would concern female receptivity to multiple matings. Many female insects become unreceptive after mating, and remain so until they have used up much of the stored sperm. When gift donation accompanies mating, clearly females should accept most matings. From optimal foraging theory, females should show unreceptivity to male i only when: [energetic gain from i/handling time for i’s gift] < [average overall rate of energy gain from the optimal set of males, inclusive of both handling times and travel times between successive gifts]. In many giftdonating species, females are receptive to most males (e.g. THORNHILL 1976b; SAKALUK 1987). However, some species that are receptive to multiple mating d o not give gifts (e.g. Scutophagu stercouauiu, where it pays females to mate so as to gain a guarding male; PARKER 1970 b), and some species with relatively small gifts become unreceptive after mating (e.g. Drosophilu melunogusteu, GROMKO et al. 1984), presumably because the benefit of the gift does not outweigh the costs of remating. 7. If females accept most males regardless of gift quality, then unless paternal investment is significant, selection would presumably act to reduce costs of gifts to males; mutant males with reduced investment would not suffer rejections, and could use the gift energy to gain extra matings. This might be mimicked in species where the gift has no nutritive value (e.g. certain Empididae, KESSEL1955). If paternal investment is significant, specialized and expensive gifts can readily be maintained, since “cheating” is not favourable to males. Gift reduction can also be prevented if sperm transfer increases with gift size; THORNHILL (1976 b) has shown in Hylobittucus upiculis that the copula duration and hence the P2 achieved by a male increases with the size of prey donated. In summary, we feel that mating effort probably contributed most significantly to the origin of seminal and somatic gift donation, and possibly also prey gifts, but that paternal investment must have contributed most to the origin (and maintenance) of suicidal gifts, and could have been significant in the origin of prey gifts. However, paternal investment may now contribute very significantly to the present maintenance of energetically expensive gifts, especially because species with high paternal effects on progeny fitness will not be susceptible to invasion by “cheating” (gift reducing) mutants. zyxwvuts 342 zyxwvu zyxwvutsrq L. W. SIMMONS & G. A. PARKER Literature Cited zyxwvutsr zyxwvuts ALEXANDER, R. D., & G. BORGIA,1979: O n the origin and basis of the male-female phenomenon. In: Sexual Selection and Reproductive Competition in Insects. (BLUM,M. S., & N. A. BLUM, eds.) Acad. Press, New York, pp. 41 7 - 4 4 0 . & S. B. MATHUD,1977: Mating induced oviposition and egg BENTUR,J. S., K. DAKSHAYONI, production in the crickets Gryllus bimuculutus de Geer and Plebeiogyllus guttiwentrzs Walker. Z. Angew. Entomol. 84, 129-135. BOGGS,C . L., & W. B. WAIT, 1981: Population structure of pierid butterflies IV. Genetic and physiological investment in offspring by male Colzus. Oecologia 50, 3213-324. BOORMAN, E., & G. A. PARKER, 1976: Sperm (ejaculate) competition in Drosophilu melunoguster and the reproductive value of females to males in relation to female age and mating status. Ecol. Entomol. 1, 145-155. BOWEIV,B. J., C. G. CODD, & D . T. GWYNNE,1984: The katydid spermatophore (Orthoptera: Tettigoniidae): male nutritional investment and its fate in the mated female. Aust. J. Zool. 32, 23-31. BOWNES,M., & L. PARTRIDGE, 1987: Transfer of molecules from ejaculate to females in Drosophilu melunoguster and Drosophilu pseudobscuru. J. Insect Physiol. 33, 941-948. BROWER,J. H., 1975: Sperm precedence in the Indian meal moth, Plodzu interpunctellu. Ann. Entomol. SOC.Am. 68, 78-80. DARWIN, C., 1871: The Descent of Man and Selection in Relation to Sex. John Murray, London. DOWNES,J. A., 1970: The feeding and mating behaviour of the specialized Empidinae (Diptera): observations on four species of Rhamphomyiu in the high arctic and a general discussion. Can. Entomol. 102, 769-791. EBERHARD, W. G., 1985: Sexual Selection and Animal Genitalia Harvard Univ. Press, Cambridge. FISHER,R. A., 1930: The Genetic Theory of Natural Selection. Dover Publ. Inc., New York. GREENFIELD, M. D., 1982: The question of paternal investment in Lepidoptera: male-contributed proteins, in Plodzu interpunctellu. J. Invert. Reprod. 5, 323-330. GROMKO,M. H., D. G. GILBERT, & R. C . RICHMOND,1984: Sperm transfer and use in the multiple mating system of Drosophzlu. In: Sperm Competition and the Evolution of Animal Mating Systems. (SMITH,R. L., ed.) Acad. Press, New York, pp. 3 7 2 4 2 7 . D. T., 1984: Male mating effort, confidence of paternity and insect sperm competition. In: GWYNNE, Sperm Competition and the Evolution of Animal Mating Systems. (SMITH,R. L., ed.) Acad. Press, New York, pp. I 17-150. _ - , 1986a: Reply to: Stepfathers in insects and their pseudo-parental investment. Ethology 71, 74-77. --, 1986 b: Courtship feeding in katydids (Orthoptera: Tettigoniidae): investment in offspring or in obtaining fertilizations? Am. Nat. 128, 342-352. --, 1989: Courtship feeding in katydids benefits the mating male’s offspring. Behav. Ecol. Sociobiol., in press. JANETOS, A. C., 1980: Strategies of female mate choice: a theoretical analysis. Behav. Ecol. Sociobiol. 7, 107-112. KESSEL,E. L., 1955: The mating activities of balloon flies. Syst. Zool. 4, 997-1004. KNOWLTON,N., & S. R. GREENWELL, 1984: Male sperm competition avoidance mechanisms: the influence of female interests. In: Sperm Competition and the Evolution of Animal Mating Systems. (SMITH,R. L., ed.) Acad. Press, New York, pp. 62-85, KREBS,J. R., 1978: Optimal foraging: decision rules for predators. In: Behavioural Ecology: an Evolutionary Approach. (KREBS,J. R., & N . B. DAVIES,eds.) Blackwells, Oxford, pp. 23-63. LISKE,E., & W. J. DAVIS,1984: Sexual behaviour of the Chinese praying mantis. Anim. Behav. 32, 916-944. LOHER,W., & B. RENCE,1978: The mating behaviour of Teleogryllus comrnodus (Walker) and its central and peripheral control. Z. Tierpsychol. 46, 225-259. LOW, B. S., 1978: Environmental uncertainty and the parental strategies of marsupials and placentals. Am. Nat. 112, 197-213. MANNING, A,, 1966: Sexual behaviour. In: Insect Behaviour. Symp. Roy. Ent. SOC.Lond. 3,59-68. zyxwv zyxw zyxw zyxwvuts Nuptial Feeding in Insects: Mating Effort versus Paternal Investment 343 MARKOW, T. A., 1988: Drosophilu males provide a material contribution to the offspring sired by other males. Funct. Ecol. 2, 77-79. _ _ , & P. F. ANKNEY,1984: Drosophilu males contribute to oogenesis in a multiple mating species. Science ?24, 302-303. PARKER,G. A,, 1970a: Sperm competition and its evolutionary consequences in the insects. Biol. Rev. 45, 525-567. _ _ , 1970b: The reproductive behaviour and the nature of sexual selection in Scutophugu stercoruriu L. (Diptera: Scatophagidae). V. The female’s behaviour at the oviposition site. Behaviour 37, 140-168. _ _ , 1979: Sexual selection and sexual conflict. In: Sexual Selection and Reproductive Competition in Insects. (BLUM,M. S., & N . A. BLUM,eds.) Acad. Press, N e w York, pp. 123-166. _ _ , 1982: Why are there so many tiny sperm? Sperm competition and the maintenance of two sexes. J. theor. Biol. 96, 281-294; 98, 707. _ _ , 1983: Mate quality and mating decisions. In: Mate Choice. (BATESON,P., ed.) Cambridge Univ. Press, Cambridge, pp. 141-166. _ _ , 1984: Sperm competition and the evolution of animal mating strategies. In: Sperm Competition and the Evolution of Animal Mating Systems. (SMITH,R. L., ed.) Acad. Press, New York, pp. 1-60. _ _ , & L. W. SIMMONS,1989: Nuptial feeding in insects: theoretical models of male and female interests. Ethology, in press. QUINN, J. S., & S. K. SAKALUK, 1986: Prezygotic male reproductive effort in insects: why do males provide more than sperm? Fla. Ent. 69, 84-94. RUTOWSKI, R. L., G. W. GILCHRIST, & B. TERKANIAN, 1987: Female butterflies mated with recently mated males show reduced reproductive output. Behav. Ecol. Sociobiol. 20, 319-322. SAKALUK, S. K., 1986a: Is courtship feeding by male insects parental investment? Ethology 73, 161-166. _ _ , 1986b: Sperm competition and the evolution of nuptial feeding behaviour in the cricket, Gvyllodes supplicuns (Walker). Evolution 40, 584--593. - _ , 1987: Reproductive behaviour of the decorated cricket, Gyllodes supplzcuns. (Orthoptera: Gryllidae): calling schedules, spatial distribution and mating. Behaviour 100, 202-225. SIMMONS, L. W., 1986: Female choice in the field cricket, Gvyllus birnuculutus (de Geer). Anim. Behav. 34, 1463-1470. _ _ , 1987: Sperm competition as a mechanism of female choice in the field cricket, Gyllus birnuculutus. Behav. Ecol. Sociobiol. 21, 197-202. _ _ , 1988: The contribution of multiple mating and spermatophore consumption to the lifetime reproductive success of female field crickets (Gvyllus bimuculutus). Ecol. Entomol. 1 3 , 5 7 4 9 . STEELE,R. H., 1986: Courtship feeding in Drosophilu subobscuru. The nutritional significance of courtship feeding. Anim. Behav. 34, 1087-1098. THORNHILL, R., 1986a: Sexual selection and paternal investment in insects. Am. Nat. 110, 153-163. _ _ , 1986b: Sexual selection and nuptial feeding behaviour in Bittucus upiculis (Insecta: Mecoptera). Am. Nat. 110, 529-548. _ _ , & J. ALCOCK,1983: The Evolution of Insect Mating Systems. Harvard Univ. Press, London. TRIVERS, R. L., 1972: Parental investment and sexual selection. In: Sexual Selection and the Descent of Man, 1871-1971. (CAMPBELL, B., ed.) Aldine, Chicago. WALKER, W. F., 1980: Sperm utilization strategies in non-social insects. Am. Nat. 115, 780-799. WICKLER,W., 1985: Stepfathers and their pseudo-parental investment. Z. Tierpsychol. 69, 72-78. _ _ , 1986: Mating costs versus parental investment: a reply to Gwynne. Ethology 71, 78-79. _ _ , & U. SEIBT, 1985: Reproductive behaviour in Zonocerus eleguns (Orthoptera: Pyrogomorphidae) with special reference to nuptial gift guarding. Z. Tierpsychol. 69, 203-223. WRIGHT,S., 1932: The roles of mutation, inbreeding, crossbreeding and selection in evolution. Proc. 6th Int. Congr. Genet. 1, 356-368. zyxwvut and G . A. PARKER,Evolutionary Biology Research Group, Authors’ address: L. W. SIMMONS Department of Environmental and Evolutionary Biology, University of Liverpool, Liverpool L69 3BX, U.K.