HOST CASTRATION AS A PARASITIC STRATEGY

MARIO BAUDOIN Department 0/ Zoology, The University 0/ Michigan, Ann Arbor, Mich. 48104

Received April 29, 1974

The "destruction or alteration of (the Such interpretations have been rejected host's) gonad tissue by parasites" (Noble by Rothschild (Rothschild and Clay, 1952) and Noble, 1971) is widely referred to as who remarks that: parasitic castration. It is often accompa­ "In the case of larval flukes it is worth remem­ nied by a modification in the expression of bering that there can be no gradual adaptation the host's secondary sexual characteristics between host and parasite. Selection is entirely and may also have a variety of other phys­ one-sided. The parasite castrates the host or, iological and behavioral effects. That its in the case of young snail hosts, inhibits the effects may have considerable influence growth of the gonads, and therefore the more susceptible snails, and even those which survive upon the life-history of the parasite has infection the most successfully, do not repro­ been less appreciated. duce themselves and are eliminated from the Parasitic castration has attracted scien­ population. Consequently adaptation can only tific attention ever since it was first re­ be on the side of the parasite. In the cases of ported by Maln (1881) and studied by larval flukes this situation is very obvious, but it probably exists in many other cases of para­ Giard (1886, 1887, 1889; Giard and Bon­ sitism, when it is wrongly assumed that adap­ nier, 1887). The phenomenon has been re­ tation is mutual." (op. cit" p. 35). viewed for various groups of host organ­ isms, e.g., Crustacea (Reinhard, 1956; Rothschild (1941b,c) proposed increased Charniaux-Cotton, 1960), Insecta (Wiilker, host growth as the selective advantage of 1964), and Mollusca (Koie, 1969; Cheng, parasitic castration for trematodes castrat­ 1971; Wright, 1971), as well as for groups ing their first intermediate host, Peringia of parasites, e.g., Nematoda (Welch, 1965) ulvae. She demonstrated a positive correla­ and Turbellaria (Jennings, 1971). Its im­ tion between frequency of parasitism and portance has recently gained recognition in host size, which led her to postulate that a number of parasitology textbooks (e.g., parasitized hosts grew faster and to a larger Caullery, 1952; Cheng, 1972; Noble and size than non-parasitized ones. This, she Noble, 1971). said, should result in increased fitness of Several explanations of parasitic castra­ the parasite, since the number of cercariae tion have been proposed. McClelland and produced by a snail is proportionate to its Bourns (1969), for example, consider it to size. However, an alternative hypothesis be primarily an adaptive response of the to that of increased host growth has been offered by Hartnoll (Hartnoll, 1967), to host, and they use a group-selection argu­ explain crustacean castration. He postu­ ment for their explanation: lated that the typical sacculinization effects "We feel that the inhibition of molluscan re­ are produced by the parasite, which bene­ production by trematode parasites must be re­ fits from the resulting physiological and be­ garded as a major adaptation which serves on havioral changes. These changes apparently the one hand to spare snails from the double burden of producing both eggs and cercariae result in all parasitized individuals, regard­ and on the other hand as a resources-manage­ less of sex, behaving like ovigerous females; ment device whereby one portion of the snail such individuals do not reproduce while population serves its own kind and the para­ they are parasitized, but they provide pro­ sites by reproducing, while the remainder serves the parasites by incubating cercariae over an tection and ventilation to the parasites extended period of time." (op. cit., p. 145). which occupy the external location usually

EVOLUTION 29:335-352. June 1975 335 336 M. BAUDOIN taken by the host's eggs. But this explana­ fitness gained by that investment as com­ tion applies only to cases in which the para­ pared to the cost. Both gain and cost may site occupies the position of the host's off­ be dependent upon the demographic en­ spring and might thereby benefit from the vironment of the organism; at any time, host's maternal behavior; most cases of an organism faces the decision whether or parasitic castration do not fall under this not to spend effort in reproduction and, if category, and it certainly does not explain so, how much effort to spend. (This de­ the "masculinization" of stylopized (para­ cision may be either a facultative or a fixed sitized by Strepsiptera) Andrena (Perez, response, depending on the predictability 1886; Salt, 1927, 1931). and variability of the demographic en­ Although these and other explanations vironment.) have been offered for particular instances Given these assumptions, one might pre­ of parasitic castration, discussions of the dict that, in a host which has attained such subject to date fail to provide any general an adaptively balanced schedule of repro­ hypothesis or unifying interpretation of the duction, any characteristic which reduces phenomenon's adaptive significance. This reproductive effort will likely increase the paper attempts to supply such an interpre­ energy available for non-reproductive pur­ tation in terms of fitness of parasites. poses. These purposes, which include fat In developing this interpretation, I sug­ storage and growth, could result in an in­ gest that host castration can be viewed as crease in the host's survivorship. For ex­ a strategy of the parasite. By strategy I ample, in his experiments with Agonum ju­ mean a sequence of physiological or be­ liginosum and A. thoreyi Murdoch (1966) havioral states of a genotype which result found that, "other mortality factors being in a modification of a factor or factors de­ of equal intensity, the survival of adult fe­ termining the fitness of that genotype. The male Carabidae, from near the end of one persistence of the strategy in the organism's breeding season to the start of the next, is population being determined by its effects inversely proportional to the amount of on individual fitness (i.e., its optimization). reproduction in that first breeding season." Basically this strategy involves a curtail­ Similar results have been obtained for bar­ ment of the "reproductive effort" of the nacles (Barnes, 1962), domestic cats host. Reproductive effort can be defined (Hamilton et al., 1969) and man (Hamil­ as that portion of the total resources, uti­ ton, 1948). In this connection, it should lized at a given time, which is devoted to be noted that sterile mutant Drosophila reproduction (Gadgil and Bossert, 1970; show an increase in lipid content of the Williams, 1966a,b). This effort may have body fluids and in fat deposition (Doanne, a "cost" in terms of future reproduction, if 1960a,b). Furthermore, an increase in the current expenditure of time and/or en­ weight of castrated domesticated animals ergy resources reduces survivorship and/or is well known. future fecundity (Williams, 1966b; Fisher, Castrating parasites may therefore use 1958) . Organisms will tend to develop to their own advantage the results of a re­ schedules of reproduction such that future duction in the host's reproductive effort. loss will on the average be balanced by By interfering with host reproduction, they gains in fitness due to present investment may gain the advantages derived from in­ (Williams, 1966b). Characteristics which creased host survivorship (Goertz, 1966; result in increased investment in reproduc­ Wecker, 1961), increased host growth tion at no future cost will always be selec­ (Cheng, 1971) , and/or increased energy tively advantageous and are therefore un­ availability (Hughes, 1940; Kornhauser, interesting in the present context. However, 1919). My postulate is, therefore, that present reproductive effort involving some parasitic castration can be seen as an adap­ future cost will be selected in proportion to tation resulting in increased fitness of the HOST CASTRATION 337

TABLE 1. The effects which have been observed in a broad spectrum of parasite-host interactions.

Effects on the host Sec. Kin- Survivor- sex. Means ship Gonads Growth ship char. or of castra- castra- Parasite Host 'i' a 'i' a 'i' a 'i' a tion tors Source Protozoa Phytomastigophora (Crustacea) Blastodinium Pseudocalanus -+- -+- 0 - IS Cattley, 1948. contortum sp.

Telosporea (Crustacea) Agregata lnachus - - DC Smith, 1905. inachi dorsetensis

Microsporidea (Crustacea) Octosporea Gammarus R R HC Bulnheim and effeminans duebeni Vavra, 1968. (Insecta) Nosema Tribolium -+- -+- - - H C? Fisher and sp. confusum Sanbo, 1962, 1964. (Pisces) Pleistophora Notemigonus -+- +? D,H? C Summerfelt and ovariae chrisoleucas Warner, 1970.

Ciliatea (Asteroidea) Orchytophrya Asterias D M? Vevers, 1951. stellarum rubens

Ellobiopsidae (Crustacea) Thalassomyces Thysanoesa -+-?-+-? -+-?-+-? - - H?,I S Hoffman and fagei raschi Yanc, 1966; Mauchline, 1966.

Mesozoa

Orthonectidea (Ophiuroidea) Rhopalura Amphiura D,H? C Stunkard, 1954; ophiocomae squamata Fontaine, 1968. Ophiotrix [ragilis

Platyhelminthes Turbellaria (Crustacea) Kronborgia Ampellisca S Kanneworff, amphipodicola macrocephala 1965; Christen- sen and Kanne- worff, 1965. 338 M. BAUDOIN

TABLE 1. (continued)

Effects on the host Sec. Kin- Survivor- sex. Means ship Gonads Growth ship char. of of castra- castra- Parasite Host '¥ cJ' '¥ cJ' '¥ cJ' '¥ cJ' tion tors Source Trematoda (Gastropoda) Trichobilharzia Limnaea + + H C McClelland and ocellata stagnalis Bourns, 1969. (Gastropoda) "Metacercaria Littorina H(-) (-) (-) (-) H D M Rothschild, A" neritoides 1941aj Lysaught, 1941. "Cercaria Littorina + + +?+? I C Rothschild, B" neritoides 1941a. (Gastropoda) Cercaria Peringia + + +?+? I C Rothschild, oocysta ulvae 1941b,c. (Gastropoda) Zoogonoides Buccinum +?+? +?+? - D? C K~ie, 1969. viviparus undatum Cestoidea (Crustacea) Diplocotyle Gammarus + I S Stark, 1965 sp. zaddachi

Aschelminthes Nematoda (Insecta) Mermis Simulium + + +?+? - - H? S Strickland, 1911. sp. sp. (Insecta) Sphaerularia Bombus* H F Palm, 1948. bombii sp. Nematornorpha Gordius Stauroderus + - IS Ebner, 1940. sp. rammei

Arthropoda Crustacea (Crustacea, Peltogaster Pagurus +?+? +?+? H? S Samuelsen, 1970. curvatus cuanensis (Crustacea) Sacculina Microphrys +? +? - H S Hartnoll, 1967. bicuspidato bicornutus (Crustacea) Septosaccus Diogenes +? +? H? S Bourdon, 1963; rodriguezi pugilator Perez, 1928. (Crustacea) Hamiarthrus Spirontocaris +?+? +?+? I S Pike, 1960. abdominalis lilljeborgii * Sphaerularia bombii produces no external modifications in its host, an insect. This is probably due to the fact that infection takes place once the host is an imago in which all the external characters are already determined. Other cases of castration may be overlooked because of the lack of obvious effect. HOST CASTRATION 339

TABLE 1. (continued)

Effects on the host Sec. Kin- Survivor- sex. ~[eans ship Gonads Growth ship char. of of ---- castra- castra- Parasite Host ~ .s ~ <1 ~ a ~ a tion tors Source Insecta (Insecta) Stylops Andrena +? I S Smith and mellitae nigroaena Hamm, 1914. (Insecta) Elenchus Javasella -- H S Raatikainen, tenuicornis pellucida 1966; Baumert- Behrisch, 1960. (Mammalia) Cuterebra Peromyscus + n,H? S Wecker, 1962. angustijrons leu copus Cuterebra Peromyscus + + n,H? S Goertz, 1966. sp. leucopus Cuterebra Peromyscus + + n,H? S Hunter et al., grisea maniculatus 1972. Symbols used in the table: +, indicates an enlargement of, or increase in, characteristic; -, reduction of characteristic; ?, nature of effect, means of castration, or kinship of castrator inferred from what is stated in source; 0, no observed effect; (), effect apparent only in heavy infections; I, indirect castra­ tion, either hormonal or nutritional; D, direct castration; H, hormonal castration; S, single parasites can and usually are responsible for castration; C, the offspring resulting from asexual reproduction of a parasite are responsible for castration; F, the sexually produced offspring of a parasite are responsible for castration; M, multiple genotypes are probably responsible for castration; R, functional sex reversal. parasite; this increased fitness is a con­ sites as possible while still restricting the sequence of the improved environment due examples to those where interference with to host castration. host reproduction is likely to be a para­ site's adaptation. All the examples in the DIRECT EFFECTS OF PARASITIC table with the exception of Octosporea ei­ CASTRATION ON THE HOST feminans have at some time been reported Before proceeding further, it seems use­ as "parasitic castration." I have included ful to provide a brief but broad survey of O. effeminans in the table since I believe the direct effects of parasitic castration that it can shed some light on the condi­ upon the host organism. Table 1 summa­ tions under which parasitic castration is not rizes the effects which have been observed likely to evolve (see p. 345). Both "Meta­ in a broad spectrum of parasite-host inter­ cercaria A" and Orchytophrya stellarum actions. I have tried to ascertain what are examples of parasites which castrate means of castration are utilized by each their hosts but in these cases parasitic cas­ parasite and what genetic relationship is tration may not be the result of a para­ most likely between parasites in the same site's adaptation (see p. 346). These two individual host. The relevance of these last examples were included in order to con­ considerations to this paper will be dealt trast their characteristics with those of the with below (p. 346 and p. 347). The ex­ other, more typical, castrators. amples used in Table 1 are only a small The principal direct effects of parasitic though, I believe, representative sample castrators on their hosts can be classified of those found in the literature. I have as effects on a) secondary sexual charac­ tried to include as many groups of para- teristics; b) internal organs; c) internal 340 M. BAUDOIN secretions; d) viability and growth; e) gonads. The degree to which gonads are behavior; f) reproduction; and g) sex re­ modified varies from a small reduction in versal. The magnitude of these effects de­ size (Baumert-Behrisch, 1960; Welch, pends on a variety of factors, including the 1959; Christie, 1936) to total atrophy host's sex. (Perez, c., 1933; Salt, 1927, 1931). Para­ sites responsible for parasitic castration Effects on the Host's Secondary may also have profound effects on organs Sexual Characters other than the gonads (Matsumoto, 1952, Perhaps the most obvious result of para­ 1953; Fischer, 1928). Castrators can be sitic castration is a modification of the divided into three groups according to their secondary sexual characters of the host. effects on the internal organs of the host The subject has been intensively studied in (Noble and Noble, 1971). Insects and in Crustacea. The reviews by 1) Castrators which have a direct effect Reinhard (1966), Wiilker (1964), Charni­ on a number of organs including the aux-Cotton (1960), Wheeler (1910), Salt gonads (e.g., metacercariae). (1927,1931) cover the subject adequately. 2) Castrators which have a direct effect The following generalizations can be ar­ on the gonads but have little effect rived at on the basis of available literature. on other organs (e.g., Rhopalura). Host secondary sexual characters are usu­ 3) Parasites which have little direct ef­ ally modified in such a way that parasitized fect on the gonads but where there organisms of different sexes are more sim­ is a strong positive correlation be­ ilar than are those of the corresponding tween presence of the parasite and nonparasitized organisms (Reinhard, 1966; a reduction of the gonads (e.g., Wiilker, 1964). The effect on secondary Eoxenos, H emiarthrus) . sexual characters usually can be interpreted as juvenilization of the host (Reinhard, This last case seems to be the most com­ 1956; Wiilker, 1964), the exceptions being mon, at least in Insect and Crustacean hosts cases where the parasite is external and oc­ (Caullery, 1952; Reinhard, 1956; Wiilker, cupies the location of the eggs in eggbear­ 1964) (see Table 1). ing females, e.g., sacculinization (parasit­ ization by sacculinid crustacea) of some Effects on the Host's Internal Secretions crabs (Hartnoll, 1967). Secondary sexual Parasitic castration is often accompanied characters directly involved in mating or by changes in the internal fluids of the finding a mate are in some cases reduced host. The literature covering this aspect in size (Rempel, 1940; Nielsen, 1970; of parasitic castration has been reviewed Hartnoll, 1962, 1967). Secondary sexual by von Brand (1952); Fisher (1963); characters involved in competition for Cheng and Snyder (1962); Cheng (1971). mates are also usually reduced in size and No generalizations can be made for the their condition is more similar to those effects of castrators on the internal en­ found in nonaggressive conspecifics (Rever­ vironment of their hosts, but I consider beri, 1943; Baffoni, 1948). that the following observations are impor­ Secondary sex characters which serve as tant in the present context. In some cases, trophic structures involved in provisioning parasitic castration results in an increase for the offspring, such as the scopa of An­ in lipid content of the internal fluids of drena, may also be reduced (Perez, 1886; the organism (Hughes, 1940; Kornhauser, Salt, 1927, 1931; Smith and Hamm, 1914). 1919; von Brand, 1952; Reinhard, 1956). An increase in fat depositions has been re­ Effects on the Internal Organs of the Host ported in other cases (Welch, 1959; Rob­ Parasitic castration, by the definition son, 1911; Rudloff and Veillet, 1954; given, involves a modification of the host Smith, 1913). Some castrators have been HOST CASTRATIO:\, 341 shown to secrete hormones into the host's sibilities. However, laboratory studies have internal environment (Denner, 1968; definitely shown that at least in some cases Fisher, 1963). Castrators have been shown parasitic castration results in increased host to attack directly or to interfere with the growth (Rothschild and Rothschild, 1939; host's endocrinological organs (Matsumoto, McClelland and Bourns, 1969) and via­ 1952; Palm, 1948; Brandenburg, 1953, bility (McClelland and Bourns, 1969). 1956; Hattingen, 1956). In some cases, Also available are field data indicating an the presence of the castrator does not re­ increase in host viability in castrated hosts sult in any detectable modification of the (Goertz, 1966; Wecker, 1962). A much host's fluids (Reinhard, von Brand and more common report is the absence of ob­ l\lcDuffie, 1947). vious pathological conditions which would Charniaux-Cotton in an extensive review imply reduced viability. (1960) of the literature on parasitic cas­ tration in Crustacea concluded that para­ Effects on Host Behavior sitic castration in this group is accom­ The effects of parasitic castration on host plished by the parasites' interference with behavior have seldom been investigated and the secretions of the androgenic gland of most of the information is inferential. The the host (op. cit., p. 431). following are some relevent observations. Stylopization may result in drastic changes Effects on Host Viability and Growth in host behavior. Salt (1927, 1931); Perez Field data on the effects of castrators (1886) and Smith and Hamm (1914) re­ on host viability are almost totally lacking. ported that stylopized female Andrena bees Growth data of individual castrated hosts had reduced scopae and that they seldom are also lacking. There are, however, a if ever carried pollen. Nonparasitized bees number of studies that show a positive were on the other hand almost always correlation between frequency of infection found to be carrying pollen. and host size (Allen, 1966; Hartnoll, 1967; Stylopized digging wasps (Chlorion, Sphex) show smooth and unworn jaws Pike, 1960; Rothschild, 1936, 1938, 1941a, while normal females show worn and b,c; Koie, 1969; Stark, 1965; Bourdon, scratched mandibles (Salt, 1931). 1963; Mauchline, 1966; Addel-Malek, The control of dominance hierarchies in 1952). The following alternatives could shore crabs (Cyclograpsus punctatus) has produce such correlation: been shown (Caiger and Alexander, 1973) 1) The castrator is more likely to infect to be determined by the action of the an­ large hosts. drogenic gland. This gland is interfered 2) The castrator produces an increase in with by castrators and therefore one can host growth. expect that parasitic castration would re­ 3) The castrator reduces mortality with­ sult in a reduction in the aggressive be­ out affecting growth. havior (op. cit., p. 139) characteristic of 4) All host sizes are equally likely to be dominant males. infected per unit time and host size Sacculinized male and female shore crabs increases with age. (Carclnus maenas) stay in deep water 5) The castrator has a size-dependent while nonparasitized males migrate to and effect on behavior and it increases remain in shallow water. Normal females the catchability of the larger hosts. go to shallow water in order to mate (Ras­ 6) The castrator causes a high mortality mussen, 1959). After mating, ovigerous fe­ of the smaller (younger) hosts. males return to deep water. Unfortunately, the data available usually Simuliid larvae parasitized by Mermis do not allow us to choose among these pos- sp. show a reduction in mobility and an 342 M. BAUDOIN

increase in the ability to collect food Sex Reversal (Strickland, 1911). Even though the literature mentions Parasitized male chironomids show a re­ many cases of presumed sex reversals due duction in the complexity of their antennae to parasitic castration (Cattley, 1948; and tarsi (Rempel, 1940). This reduction Rempel, 1940; Potts, 1906; Okada and suggests either a reduction in their mate­ Miyashita, 1935), most of them could searching behavior or a reduction in their equally well be interpreted as retention ability to find mates. Normal chironomids of juvenile characteristics (Brandenburg, use their antennae in sound detection. 1953; Reinhard, 1956; Wlilker, 1964). Mermithized ants remain in the nest and This statement cannot be applied to cases do not forage like uninfected ants; neither do they tend the offspring (Wheeler, 1910). of parasitism of holometabolous insects, Stylopized colonial Hymenoptera show a since the immature is a larva not at all reduction in brood care (Siebold, 1843). like the adult. That secondary sexual char­ Similarly worker honey-bees, infected with acters in insects are reduced in their ex­ Nosema apis, do not tend the brood (Wang pression is, however, analogous to the re­ and Moeller, 1970). tention of juvenile structures in other host These observations suggest that para­ groups. Exceptions to this generalization sitic castration often results in changes in are the speeding up of development of some host behavior which involve a reduction in hosts (e.g., Andrena: Perkins, 1918) and sexual or parental activities. the acquisition of typical female structures and physiological characteristics by some Effects on Host Reproduction Crustacea (Hartnoll, 1967). Parasitic cas­ The effects of castrators on secondary tration does not result in true functional sexual characteristics, internal organs, in­ sex reversal (Reinhard, 1956; Wiilker, ternal secretions, and behavior of the host 1964). The effects of Octosporea effemi­ indicate that reproduction is reduced dras­ nans on Gammarus duebeni (Bulnheim and tically. In most cases it is safe to assume Vavra, 1968) constitute the only case of that there is no host reproduction for most functional sex reversal that I have come of the time of infection. Not only is gamete across. The reasons for this exception will production curtailed but all aspects of re­ be dealt with below. productive behavior and reproduction re­ Parasitic castration effects are highly de­ lated growth are reduced. The reduction in pendent on host sex. Which of the sexes is the reproductive expenditure of the host is more obviously modified seems to depend usually much larger than is apparent from more on the host than on the parasite. For consideration of only one of the variables instance, in Hymenoptera, the females are involved. usually more modified than the males, In hosts which show the effects of para­ whether they are parasitized by Strepsip­ sitic castration, the amounts of time and tera (Salt, 1931) or by Mermithidae energy devoted to reproduction are greatly (Kloft, 1949). In chironomids, males seem reduced; again, the only exceptions involve to be more obviously affected (Wiilker, cases where the parasite derives a direct 1962). In Crustacea, males seem to be benefit from the particular reproductive more modified than females but the degree behavior or structure (Hartnoll, 1967; to which they are is parasite dependent Bulnheim and Vavra, 1968; Reinhard, (Reinhard, 1956; Nilsson-Cantell, 1926). 1956; Wlilker, 1964). Permanent sterility is often a result of parasitic castration. The modification of host gonads is also Even in those cases were parasitic castra­ sex dependent. In most cases, the male's tion does not result in permanent sterility gonads are less modified than the female's the reduction in host fecundity is consid­ (Wlilker, 1961; Reinhard, 1942; Branden­ erable (Pike, 1960). burg, 1953). HOST CASTRATION 343

CHARACTERISTICS OF CASTRATORS tration cannot, however, explain gigantism Size and Energy Relationships and increased lipid concentration in the haemolymph of castrated animals, unless The literature does not often afford data it involves a very selective withdrawal of on the relative size of castrators and their key nutrients. hosts. The available references indicate c) Indirect hormonal castration has been that castrators are unusually large parasites asserted in a number of instances (Palm, (Kuris, 1974). In general castrators con­ 1948; Veillet and Graf, 1958; Baumert­ stitute 1-10% of the weight of their hosts Behrisch, 1960) and, as was mentioned be­ (Kuris, 1974). This relationship applies fore, the secretion of "castrating" hor­ to individual parasites when individuals mones has been demonstrated in the lab­ are usually responsible for castration of oratory. their hosts. When castration is usually ac­ complished by a clone or by descendants of Kinship of Castrators in the Same a propagule, then the combined weight of Individual Host the group is of that magnitude (Woodhead, 1930). If one assumes that the energy re­ Parasitic castration is almost invariably quirements of parasites are proportional to produced either by single parasites or by their mass, then one can say that castrators their immediate offspring. Some studies (Altes, 1962; Veillet, 1945) have shown have unusually large energy requirements. that in some cases castrators are associated Means of Castration with conspecifics less frequently than ex­ pected by chance alone. These observations Host castration can be the result of dif­ indicate that parasitic castration is usually ferent actions by the parasite. Depending produced either by single genotypes or by on the immediate cause, parasitic castration very closely related genotypes. It is un­ can be either a) direct, b) indirect nutri­ usual to find castration (Table 1) produced tional, or c) indirect hormonal. by a number of unrelated genotypes. Meta­ a) Direct ingestion and/or direct me­ cercarial infections in snails are exceptions chanical interference with the gonads oc­ (Rothschild, 1941a; Lysaught, 1941; Ho­ curs in a few cases (Perez, 1933; Summer­ shina and Ogino, 1951). In these cases, felt and Warner, 1970; Einarsson, 1945). however, parasitic castration may be in­ In some, the parasite damages the gonads cidental (d. p. 348). in conjunction with other organs (Atkins, 1933; Koie, 1969; Cheng and James, Parasite Fecundity and Host Size 1960), while in others the damage is re­ If one assumes that the fecundity of a stricted to the gonads (Lichtenstein, 1921). parasite is proportional to its size, then The sequence in which the organs are dam­ parasite size can be taken as an index of aged may vary: in some cases, the gonads parasite fecundity. Given this assumption, are the first organs attacked, and other or­ gans are attacked only much later (Pres­ a positive correlation of host size and para­ cott, 1960; York and Prescott, 1952); in site size would imply a positive correlation other cases, the gonads are damaged after between host size and parasite fecundity. other organs are affected (Lysaught, 1941). Some studies on Crustacea have in fact b) Indirect castration through with­ shown a positive correlation between host drawal of nutrients has often been claimed and parasite sizes. This same relationship as the immediate cause of parasitic castra­ has been observed in trematode infections tion (Salt, 1927, 1931; Reinhard, 1956). of mollusca (Rothschild, 1936, 1938, For some of the cases involved, alternative 1941a; Abdel-Malek, 1952). Rothschild explanations have been proposed (Wiilker, also observed that the number of cercaria 1964; Reinhard, 1956). Nutritional cas- produced by a snail was proportional to 344 M. BAUDOIN its size (Rothschild and Rothschild, 1939). age and/or an increased number of repro­ Similar results have been obtained for Crus­ ductions. It might also mean an increased tacea (Allen, 1966; Pike, 1960). probability of completing the parasite's own life cycle. DISCUSSION Parasitic castration has been shown to Benefits to the Parasite result in changes in the host's sexual be­ havior and in a reduction of secondary The benefits that a parasite may derive sexual characteristics associated with sexual from host castration include a) increased behavior. These activities probably involve energy availability, b) increased host via­ large amounts of risk and energy expendi­ bility and c) increased host growth. ture (Williams 1966a,b; Trivers, 1972) a) Increased energy availability. Para­ and their suppression might result in in­ sitic castration often results in an increase creased host viability. Whether these mod­ in lipid concentration or deposition in the ifications result in an absolute gain in host host. This has obvious potential advantages survivorship will depend on the demands for the parasite if the latter can utilize made by the parasite. Castrated hosts may lipids in the production of offspring. That have lesser viability than non-parasitized some parasites may indeed be able to uti­ hosts but as long as they have greater via­ lize these nutrients in this way is suggested bility than parasitized hosts which are not by the fact that some castrators responsible castrated, parasitic castration will be adap­ for increased lipid concentration in their tive for the parasite. The often reported hosts have themselves a high lipid concen­ positive correlation of host size (age?) and tration (e.g., Gyge branchialis, Hughes, frequency of parasitism could be due to 1940) . The detectable concentration of increased host survivorship (Table 1). lipids in the host will, however, depend on c) Increased host growth. Even though the intensity of the lipid utilization by the the field data are not conclusive, laboratory parasite (e.g., Peltogaster paguri, Reinhard, studies show that parasitic castration may 1942). For this reason, low lipid concen­ lead to increased host growth. This in­ tration in the host (e.g., Pagurus, Reinhard creased host growth could be advantageous and von Brand, 1944) does not indicate to the parasite for the following reasons: that castration has not resulted in increased a) there may be a positive correlation be­ energy available to the parasite. A similar tween host size and parasite fecundity; b) argument could be pursued to explain in­ increased host size may mean greater host creases in glycogen (Hughes, 1940) in cas­ viability through predator avoidance or de­ trated host body fluids. fense', and c) greater host size may result. b) Increased host viability. Reproduc­ in greater energy available to the parasite tion may involve a number of risks and provided that the larger host is also a better energy expenditure for the host. Sexual competitor. maturation, territory maintenance and ac­ quisition, nest building, competition for Costs of Castration mates, mating, nourishment and protection The parasite's secretion of hormones of the young are all activities that may in­ and/or development of structures involved volve great energy expenditure and risks. in castration requires energy. That energy Parasitic castration, if it results in reduced could be used by the parasite for repro­ host reproductive effort, could bring about duction. In this sense, parasitic castration reduced host mortality and thus increased has a cost for the parasite. In order for host viability. An increase in host viability parasitic castration to be adaptive, this would often be advantageous to the para­ cost has to be outweighed by the benefits site. It might mean an increased proba­ resulting from parasitic castration. The bility of the parasite reaching reproductive literature does not allow any quantitative HOST CASTRATION 345 estimation of this cost of castration. But stage, then they are effectively unavailable the constant secretion of antagonistic hor­ to that parasite. Host offspring produced mones that is necessary to upset the normal before the time of parasitism are not rele­ physiological feedback mechanisms of a vant since they cannot be considered a host which is an order of magnitude larger cost. Castration would less likely be adap­ than the parasite is probably quite expen­ tive if parasite transmission could occur sive if the hormones are manufactured between adult and neonate host. Only in solely for that purpose. The cost may be the cases where the parasite is restricted much reduced if these hormones are the to the host's offspring will parasitic castra­ metabolic byproduct of other activities. tion of necessity be nonadaptive. Then Even if the costs are large, these could selection would favor any characteristic still be offset by adequate increases in host that induced the host to spend more energy viability, growth and/or energy availability in reproduction, once the parasite is in­ for the parasite. fective. This selection would occur even Besides this energetic cost, parasitic cas­ if it might result in reduced fitness of the tration also has a cost in terms of the host's host. Such is the case in Gammarus due­ offspring. These offspring are after all beni parasitized by Octosporea effeminans prospective hosts for the parasite's own (Bulnheim and Vavra, 1968). O. effemi­ offspring. In a sense, then, the parasite nans can be transmitted almost exclusively reduces its own fitness by castrating its transovarially. Infected G. duebenni fe­ host. However, as far as selection is con­ males produce only female offspring while cerned, individual fitness is meaningful only noninfected produce an equal sex ratio in when compared to that of other individuals their offspring. This increases the number with alternative genotypes. If parasitic of hosts available for the parasite's off­ castration results in lowered host avail­ spring but is unlikely to increase the host's ability to other individuals in the parasite fitness. population, then parasitic castration may still be selected for. If the host's offspring Influence of Host Life History are equally available to all the genotypes If parasitic castration is indeed an adap­ in the parasite's population, then parasitic tive action on the part of the parasite, hav­ castration would not result in any reduc­ ing the function of reducing the host's re­ tion in the castrator's relative fitness, since productive effort, then one would expect this reduction would be equal for all in­ that the effects of castration would depend dividuals in the parasite's population. Only on variables affecting this reproductive ef­ if the host's offspring were more available fort. One would also expect parasitic cas­ to the castrator than to alternative geno­ tration to be more prevalent in hosts which types would parasitic castration be disad­ are spending large amounts of reproductive vantageous. Even in this case, the effect effort while they harbor the parasite. Since of parasitic castration on the parasite's fit­ the advantages of parasitic castration would ness would be the difference between this be directly proportional to host reproduc­ reduction and the gains due to increased tive effort, castration of nonreproductive host growth, survivorship and energy avail­ hosts would result in no benefit to the para­ able to the parasite. site unless parasitic castration resulted in One of the factors affecting the avail­ growth or maturation changes beneficial to ability of the host's offspring to the para­ the parasite. Most cases of parasitic cas­ site is the developmental time of the host. tration result in definite changes while the If the offspring of the host, at the time of parasite is in the host (Reinhard, 1956; prospective castration, develop so slowly Wiilker, 1964). If parasitic castration has that they are susceptible to parasitism well the function of reducing host reproductive after the parasite has reached the infective effort for the parasite's benefit, then one 346 M. BAUDOIN

would expect the effects of parasitic cas­ modification occurring are proportional to tration to depend on the host's sex, since the time that a parasite spends in a host. this effort will take different forms in each The advantages herein postulated for sex (Trivers, 1972). Females in general parasitic castration may become irrelevant spend more calories per individual gamete to the parasite once the parasite has com­ than do males (op. cit.). Thus we see that pleted development in that host. At that testes are often not reduced while the ova­ time, it may even become advantageous to ries are (Wiilker, 1964). On the other utilize as much of the host tissues as pos­ hand, males may spend large amounts of sible, as shown by castrators which could be time and energy in the defense of territories classified as parasitoids (Christensen and and in aggression with other males (op. Kanneworff, 1965; Kurochkin, 1960). The cit.) . Here we see that secondary sexual parasites may also increase the probability characteristics involved in territoriality of the host's dying if the death of the host (chelae of Crustacea) tend to be reduced results in increased parasite fitness. Thus more in the castrated males than in the the parasite may increase the likelihood of females (Reinhard, 1956). Females, how­ predation by the appropriate predators, i.e., ever, may be involved in long-term parental those that are hosts for the next stage in the care while males may have a much shorter parasite's life cycle (Holmes and Bethel, reproductive season. In such cases, we find 1972) . a reduction in those female secondary sex­ ual characteristics and behavior involved Means of Castration in parental care (Salt, 1927, 1931). For all Most of the literature on parasitic cas­ these considerations, the host phenotype tration has not been concerned with the that results in the optimum parasite fitness ultimate causes of parasitic castration. The would often be more similar to one of the concentration has been mainly on the im­ normal host sexes than to the other and mediate causes or means of castration. As would combine characters that result in far as the parasite is concerned, parasitic high host survivorship and/or low energy castration is advantageous irrespective of expenditure. Such a phenotype cannot be the means used. It is only the energetic that of a normal host, since it would be expenditure in the different means that is that of a nonreproductive individual with relevant. Those methods that are energeti­ a fitness of near zero. cally cheaper would be more advantageous to their possessors. Which method is ac­ Influence of Parasite's Life History tually used will, however, depend in large The effect that host survivorship and manner on the evolutionary history of the growth and the availability of energy have parasite. It may be energetically more on a parasite will depend on the length of economical to ingest a sex determining time that the parasite is in the host. How gland than to secrete large amounts of an long a parasite remains in a host will be antagonistic hormone. However, the latter determined by factors that are not neces­ may be the situation found in nature be­ sarily modifiable by the parasite. Some cause parasitic castration may have evolved such factors may be seasonal host avail­ as a result of the effects of some parasite ability or seasonality of the external en­ metabolites on the host. All parasites re­ vironment. The longer the time spent in quire energy from their hosts and if we the host, the greater the chance that the assume that any substantial withdrawal of parasite will be affected by the mishaps of energy would result in reduced fecundity of the host and the greater the opportunity a host, then a large number of parasites for the parasite to influence the host. could be considered castrators. Reduced Therefore, the intensity of selection for host reproductive effort would then be an host modification and the chance of that almost unavoidable result of parasitism. HOST CASTRATION 347

All reductions of host reproductive effort to grow more than a smaller one). There­ cannot then be considered the result of a fore, it is more influenced by what happens parasite's adaptation. Nutritional castra­ to the host and has a greater chance of in­ fluencing the host. tion would be a type of castration that may 2) A greater size would result in greater in­ not necessarily be an adaptation since one fluence on the host through nutrient re­ can hardly envision a parasite not requir­ quirements, metabolite effects or physical ing nutrients. Similarly, direct castration damage. This would increase the likelihood through organ ingestion could be considered that effects on host reproductive effort would develop and that they would be large incidental to the benefits derived from the enough for selection to act upon. castration act itself. In either case, para­ 3) A larger parasite size would put the host sitic castration may involve adaptation in under greater stress and therefore make re­ the form of nutrient choice and selective lief from stress the more important. Selec­ attack of host organs. The sequence in tion for relief from stress would be propor­ tional to stress. One such relief is suppression which tissues are infected may itself be of reproduction. adaptive. In any event, one would also ex­ 4) In cases where unrelated individual para­ pect that in some cases, parasitic castration sites exist in the same host, small parasites may be the incidental result of parasitism. are more likely to occur as multiple infec­ However, on the whole, the literature sup­ tions than are large parasites-a large num­ ber of large parasites would presumably be ports my contention that parasitic castra­ too heavy a load for the host. Under such tion is often a parasite's adaptation for the conditions it seems that selection would following reasons: favor those large parasites that can exclude conspecifics, perhaps by eliciting the host's 1) Most cases of parasitic castration are of the immune responses. It would also favor those indirect type and many involve hormones. parasites that can escape intraspecific com­ 2) Many cases of direct castration are the re­ petition by increasing rates of development sult of a very select attack on sex determin­ or by reducing age and/or size at maturity. ing organs. The outcome of selection in this context ,,) The effects of parasitic castration are bene­ would be a reduced likelihood of the small ficial to the parasite but not to the host (for parasites evolving as castrators. It would exceptions see p. 348). also result in an increased likelihood of the 4) The effects of parasitic castration are ex­ large parasites evolving as castrators. tremely predictable within each group of hosts even if the means of castration and Genetic Relationship of Castrators the castrators may be quite different. to N oncastrators The Evolution of Parasitic Castration In order for parasitic castration to be One of the correlates which become ap­ selected for, castrator genotypes have to parent when considering parasitic castra­ derive advantages not possessed by noncas­ tion is the large size of castrators in rela­ trators. This would not occur when a cas­ tion to their hosts when compared to other trator is in the same host as a noncastrator. parasites. Castrators are often between one Under such conditions, increased host sur­ and ten per cent of the mass of their hosts. vivorship, growth and increased energy This is much larger than the common intes­ availability would be beneficial to both parasites. Furthermore, the noncastrator tinal parasite. The aggregate weight of would be able to use the energy not spent clones responsible for parasitic castration in castration to increase its own fitness. is around the same magnitude as that of Parasitic castration would be less likely to single individuals responsible for castration. be selected for if castrators were always to These observations may be significant in be found with competing genotypes. It terms of the evolution of castration for the could be selected for only if the frequency following reasons; of co-existence of the original castrator ge­ 1) Other things being equal, a large parasite notype with noncastrators was low enough probably spends more time in its host (has to allow a positive fitness for the castrator. 348 M. BAUDOIN

This may be the reason why today the vast Redefining Parasitic Castration majority of parasites involved in host cas­ Parasitic castration cannot be considered tration are usually either single individuals simply as a "destruction or alteration of or the progeny of a single individual. Fur­ gonad tissues by parasites" because many thermore, in parasites with complex life cases of typical parasitic castration show cycles only stages undergoing schizogony very little or no damage to the gonads are involved in parasitic castration while (Wiilker, 1964). The reduction of host re­ other stages are usually not. In some cases productive effort is however, apparent in (e.g., metacercariae), parasitic castration these cases. On the other hand, parasitic results from the infection of a large num­ castration should not be considered to in­ ber of individuals which are not likely to clude all reduction in reproductive effort be related. In these cases, parasitic castra­ stemming from parasitism, since this would tion seems to be incidental to the massive make parasitic castration almost equivalent destruction of the internal organs of the with parasitism and, at the least, would host (Rothschild, 1941a) . include many cases of parasitism that have very little in common with what is normally Parasitic Castration as a Host Adaptation regarded as parasitic castration. Even if, by and large, parasitic castration I suggest that parasitic castration be re­ seems to be a parasite's adaptation, it could defined as "a destruction or alteration of conceivably develop as a host's response to gonad tissue, reproductive behavior, hor­ parasitism under the following conditions: monal balance, or other modification that 1) The host is interoparous, results in a reduction of host reproductive 2) The host has an appreciable chance of re­ effort above and beyond that which results production after the parasite abandons the from a nonselective use of host energy re­ host; and serves by the parasite." 3) The reduction in host fitness due to suppres­ sion of reproduction is more than compen­ sated by a proportional increase in the host's SUMMARY viability and/or future fecundity. This in­ This paper surveys the literature on par­ crease, resulting from reduced reproduction, asitic castration and attempts to provide an is greater the longer the parasite is in the host. evolutionary interpretation of the phenom­ enon. Previous explanations are considered The fitness of a host genotype that sup­ and a unifying hypothesis is proposed. The presses reproduction, in the presence of main theses of this paper are (1) that par­ parasitism, would be proportional to the asitic castration can be viewed as a para­ probability of being parasitized. Hence, site's adaptation and (2) that advantages parasitic castration is most likely to evolve, derived from this adaptation are a result as a host's response to parasitism, in hosts of a reduction in host reproductive effort; which are subjected to high rates of para­ which in turn gives rise to increased host sitism, at least periodically. survivorship, increased host growth and/or In most cases, parasitic castration does increased energy available to the parasite, not satisfy the previous conditions. Host thereby increasing the parasite's Darwinian reproduction after castration is usually ab­ fitness. A survey of the literature indicates: sent; when present, it is usually a very small proportion of that which was lost to a) That castration effects on the hosts are castration. As far as I can see, only Pero­ beneficial to the parasite but not to the myscus parasitized by Cuterebra (Table 1) host; b) That the genetic relationship of could satisfy these requirements. That par­ castrators within individual hosts is such asitic castration may in some cases be a that natural selection at the level of indi­ host adaptation does not make it any the vidual genotypes can account for the ob­ less advantageous for the parasite. served effects; and c) That the widespread HOST CASTRATION 349 occurrence of hormonal castration favors organe einer Homoptere. Zoologische Beitrage the interpretation outlined above. (N.F.) 6:85-126, 291-332. Incidental castration and the possibility BOURDON, R. 1963. Epicarides et Rhizocephales de Roscoff. Cah. BioI. Mar. 4:415-434. of parasitic castration as a host's adaptation VON BRAND, T. 1952. Chemical physiology of are considered, and a definition of parasitic endoparasitic animals. Academic Press, New castration is proposed. York, 339 p. BRAXDENBURG, J. 1953. Der Parasitismus der ACKNOWLEDGMENTS Gattung Stylops an der Sandbiene Andrena vaga Pz. Zeitschrift fiir Parasitenkunde 15: I wish to specially thank Armand Kuris 457--475. for sharing with me some of his ideas on --. 1956. Das endokrine System des Kopfes parasitic castration. My deepest apprecia­ von Andrena t'aga Pz. (Ins. Hymenopt.) und Wirkung der Stylopisation (Stylops, Ins. Strep­ tion to Francis C. Evans and Donald W. sipt.). Zeitschrift fiir Morphologie und Okol­ Tinkle who were kind enough to plough ogie der Tiere 45: 343-364. the early drafts and contributed their com­ BULNHEIM, H. P., AND J. VAVRA. 1968. Infection ments. Most of the manuscript was read by the microsporidian Ociosporea effeminans by Stephen P. Hubbell who contributed sp. n. and its sex determining influence in the amphipod Gammarus duebeni. J. Parasit. 54: helpful comments. I also benefited from 241-248. suggestions made by Douglas Futuyma, CAIGER, K. M., AND A. J. ALEXANDER. 1973. The Nelson G. Hairston, Jasper Loftus-Hills, control of dominance in the brachyuran crus­ Ann Pace and Henry Wilbur. Some of tacean, Cyclograpsus punctatus Mil. Edw. these ideas were discussed in the Graduate Zoologica Africana 8: 138-140. CA TTLEY, J. G. 1948. Sex reversal in Copepods. and Demography Seminars at the Depart­ Nature 161:937. ment of Zoology at The University of Mich­ CAULLERY, M. 1952. Parasitism and Symbiosis. igan. Margaret Geers contributed time and Sidgwick and Jackson Ltd. London. 340 p. comments which rendered my English more CHARNIAUX-COTTON, H. 1960. Sex determina­ intelligible. Some of the research was tion, p. 411-447. In T. H. Waterman (ed.), done while holding an Edwin S. George The physiology of crustacea 1. Academic Press, New York. Scholarship. CHENG, T. C. 1971. Enhanced growth as a manifestation of parasitism and shell deposi­ LITERATURE CITED tion in parasitized mollusks. In T. C. Cheng ABDEL-MALEK, E. T. 1952. Morphology, bio­ (cd.), Aspects of the biology of symbiosis. nomics and host parasite relations of the Pla­ University Park Press, London. 327 p. norbidae (Mollusca-Pulmonata). Ph.D. Thesis --. 1972. General parasitology. W. B. Saun­ Univ. Mich. 226 p. ders, Philadelphia. ALLEN, J. A. 1966. Notes on the relationship of CHENG, T. C., AND H. A. JAMES. 1960. The His­ the bopyrid parasite Hemiarthus abdominalis topathology of Crepidostomum sp. infection in (Kryer) with its hosts. Crustaceana 10: 1-6. the second intermediate host, Spaherium stra­ ALTES, J. 1962. Sur quelques parasites et hyper­ tum. Proc. Helminthol. Soc. Wash. 27:67-68. parasytes de Clibanarius erythropus (Latreille) CHENG, T. C., AND R. W. SNYDER, JR. 1962. en Corse. Bull. Soc. Zool. France 87:88-97. Studies on host-parasite relationships between ATKINS, D. 1933. Pinotherion uermijorme Giard larval trematodes and their hosts. 1. A review. and Bonnier, an entoniscid infecting Pinnothe­ II. The utilization of hosts glycogen by the res pisum, Proc. Zool. Soc. London 1933: 319­ intramolluscan larvae of Glypthelmins pennsyl­ 363. vaniensis Cheng and associated phenomena. BAFFONI, G. M. 1948a. Annotazione compara­ Trans. Amer. Microsc. Soc. 81: 209-228. tive sull' azione prodotta dai Rizocefali paras­ CHRISTENSEN, A. M., AND B. KANNEWORFF. 1965. siti di Eupagurus prideauxii. Pubbl. staz. zool. Life history and biology of Kronborgia am­ Napoli 21:237-255. phipodicola Christensen and Kannerworff BARNES, H. 1962. So-called anecdysis in Balanus (Turbellaria, Neorhabdocoela). Ophelia 2:237­ balanoides and the effect of breeding upon the 251. growth of calacareous shell of some common CHRISTIE, J. R. 1936. Life history of Agamermis barnacles. Limnol. Oceanog. 7: 462--473. decaudata, a nematode parasite of grasshoppers BAUMERT-BEHRISCH, A. 1960. Der Einfluss des and other insects. J. Agricult. Res. 52: 161-198. Strepsipteren-Parasitismus auf die Geschlechts- DENNER, M. W. 1968. Biology of the nematode 350 M. BAUDOIN

Mermis subnigrescens Cobb. Ph.D. Thesis, chez les Crustaces Decapodes. Bull. sci. du University Microfilms, Ann Arbor, Michigan. Dep. du Nord 18: (2) 1-28. DOANNE, W. W. 1960a. Developmental physiol­ 1889. Sur la castration parasitaire des ogy of the mutant female sterile (2) adipose Typhlocyba par une larve d'Hymenoptere of Drosophila melanogaster. I. Adult Morphol­ (Aphelopus melaleucus Dalm.) et par un larve ogy, longevity, egg production and egg lethal­ de Diptere (Ateleneura spuria Meig). Compt. ity. ]. Expt!. Zoo!' 145: 1-21. rend. Acad. Sci. 109: 708-710. 1960b. Developmental physiology of the GIARD, A., AND ]. BONNIER. 1887. Contributions mutant female sterile (2) adipose of Drosoph­ a l'etude des Bopyriens. Trav. de L'Inst. ZOOI. ila melanogaster. II. Effects of altered envi­ Lille et Lb. Marit. Wimereux. 5:3-272. ronment and residual genome on its expression. GOERTZ, ]. W. 1966. Incidence of warbles in ]. Expt!. Zoo!. 145:23-42. some Oklahoma rodents. Am. Mdlnd. Nat. 75: DOGlEL, V. A. 1966. General Parasitology. Aca­ 242-245. demic Press, New York, 516 p. HAMILTON, ]. B. 1948. The role of testicular EBNER, R. 1940. Veranderungen an Orthopteren secretions as indicated by the effects of cas­ durch parasitische Wiirmer. Transactions of tration in man and by studies of pathological the International Congress of Entomolgy 6: conditions and the short lifespan associated 341-471. with maleness. Rec. Prog. Horm. Res. 3:257­ EINARSSON, H. 1945. Euphasiacea, I. Northern 322. Atlantic species. The Carlberg Foundation HAMILTON, ]. B., R S. HAMILTON, AND G. E. Oceanographical Expedition Round the World, MESTLER. 1969. Duration of life and causes 1928-1930 and previous "Dana" expeditions. of death in domestic cats: influence of sex, Dana Rept. 5: 1-185. gonadectomy and inbreeding. ]. Gerontology ETGES, F., AND W. GRESSO. 1965. Effects of 24: 395-411. Schistosoma mansoni infection upon fecundity HARTNOLL, R G. 1962. Parasitic castration of in Australorbis glabratus. ]. Parasit. 51:757­ Macropodia longirostris (Fabricius) by a Sac­ 760. culinid. Crustaceana 4:295-300. FISCHER, E. 1928. Sur les modifications d'un --. 1967. The effects of Sacculinid parasites organisme (Crabe) enhavi par un parasite on two ] amaican crabs. ]. Linn. Soc. London (Sacculine). Compt. rend. soc. bioI. 98:837­ (Zoo!.) 46:275-296. 839. HATTINGEN, R 1956. Zentralblatt Iiir Bakteri­ FISHER, F. M. 1938. Production of host endo­ ologie, Parasitenkunde und Infection Krank­ crine substances by parasites. In Cheng, T. C. heiten. Abt. II. 109: 236-249. (cd.). Ann. N.Y. Acad. Sci. 113:63-73. HOFFMAN, E. G., AND R M. YANCEY. 1966. El­ 1964. Nosema as a source of juvenile hor­ lobiopsidae of Alaskan coastal waters. Pacif. mone in parasitized insects. BioI. Bull. 126: Sci. 20: 70-78. 235-252. HOLMES, ]. C., AND W. M. BETHEL. 1972. FISHER, F. M., AND R C. SANBORN. 1962. Pro­ Modification of intermediate host behavior by duction of insect juvenile hormone by the parasites. In E. U. Canning and C. A. Wright microsporidian parasite Nosema. Nature 194: (eds.), Behavioural aspects of parasite trans­ 1193. mission. Zoo!.]. Linn. Soc. London. 51: (supl. FISHER, R. A. 1958. The genetical theory of 1) 123-149. natural selection. Second ed. Dover, New HUNTER, D. M., R M. F. S. SADLEIR, AND ]. M. York. 291 p. WEBSTER. 1972. Studies on the ecology of FONTAINE, A. R. 1968. A new ophiuroid host cuterebrid parasitism in deermice. Can. ]. for Rhopalura ophiocomae Giard (Orthonec­ ZOOI. 50:25-29. tida: Mesozoa). ]. Parasit. 54: 1251-1252. HUGHES, T. E. 1940. The effects on the fat and FRENTZ, R, AND A. VEILLET. 1953. Teneur en starch metabolism of Gebia by the parasite lipide et determinisme des caracteres sexuels Gyge branchialis. ]. Exptl. BioI. 17:335-336. externes chez Ie crabe Carcinus maenas Pennt, ]ENNINGS, ]. B. 1971. Parasitism and commen­ parasite par Ie Rhizocephale Sacculina carcini salism in the turbellaria. Adv, Parasit. 9: 1-32. Thompson. Compt. rend. 236:2168-2170. ] EPPS, M. W. 1937. On the protozoan parasites GADGIL, M., AND W. H. BOSSERT. 1970. Life of Calonus finmarchius in the Clyde Sea area. Historical consequences of natural selection. Quart. ]. Microscop. Sci., New Ser. 316 79 Amer. Natur. 104: 1-24. (4) :589-658. GIARD, A. 1886. De l'influence de certains para­ KANNEWORFF, B. 1965. Life cycle, food and sites Rhizocephales sur les caracters sexuels ex­ growth of the auphipod Anyselisca macro­ terieurs de leur hote. Compt. rend. 103:84-86. cephalo lilgeborg from the Presund. Ophelia 1887. La castration parasitarie et son in­ 2:305-318. fluence sur les caracters exterieurs du sexe male KLOFT, W. 1949. Ober den Einfluss von Mer- HOST CASTRATIOr\ 351

misparasitismus auf den Stoffwechsel und die sex-reversal in parasitized male crabs. Mem. Organbildung bei Ameisen. Zeitschrift fuer Coli. Sci. Kyoto Imp. Univ. B10:169-208. Parasitenkunde 14:390-422. PALM, X B. 1948. Normal and pathologica KOIE, M. 1969. On the endoparasites of Buc­ histology of the ovaries of Bombus Latr. cinim undatum L. with special reference to (Hymcnopt.), Opusc. Etomol. (Sup. 7): 1-101. the trematodes. Ophelia 6:251-279. PAN, C. T. 1965. Studies on the host-parasite KORNHAUSER, S. J. 1919. The sexual character­ relationships between Schistosoma mansoni istics of the Membracid Thelia bimaculata and the snail Australorbis glabratus. Amer. J. Fabr. 1. External changes by Aphelopus theliae Trop. Med. Hyg. 14:931-976. Grahan. J. Morphol. 32: 531-636. PEREZ, C. 1933. Processes de resorption pahgo­ KURIS, A. M. 1974. Trophic interactions: Sim­ cytaire des oocytes dans l'ovaire chez les M ac­ ilarity of Parasitic Castrators to Parasitoids. ropodia sacculinees, Compt. rend. soc. bioI. Quart. Rev. BioI. 49: 129-148. 112: 1049-1051. KUROCHKIN, Yu V. 1960. (The nematode Het­ PEREZ, J. 1886. Des effts due parasitisme des erot ylenchus pawlowskyi sp. n. castrating fleas Stylops sur les Apiaires du genre Andrena. that transmit plague). Dokl. Akad. Nauk. Actes Soc. Linn. Bord. 10:21-60. S.S.R. (AIBS Translation). 135: 1281-1284. PERKINS, R. C. L. 1918. Further notes on Sty­ LVSAUGHT, A. M. 1941. The biology and trem­ lops and stylopized bees. Entomol. Month. atode parasites of the gastropod Littorina neri­ Mag. 54: 115-129. toides (L.) on the Plymouth breakwater. J. Mar. BioI. Ass. U.K. 25:41-67. PIKE, R. B. 1960. The biology and post larval development of the bopyrid parasites Pseudione LICHTENSTEIN, J. L. 1921. Ophryoglena collini, ajiinis G. O. Sars and Hamiarthus abdominolis n. sp., parasite coelomique des larves d'Ephe­ (Kroyer) (= Phryxus abdomina/is (Kroyer) ). meres. Compt. reno Soc. BioI. et Fil. 85:794­ J. Linn. Soc. London, (Zool) 44:239-251. 796. MALM, A. W. 1881. Om Cirripeder funna vid POTTS, F. A. 1906. The modification of the sex­ Bohuslans Kust. Goteborgs Naturhist. Mus. ual characters of the hermit crab caused by zool.-zoot. Afdel. 3: 26-32. the parasite Pelto gaster (castration parasitaire of Giard). Q. Microscop. Sci. 50: 599-621. MATSUMOTO, K. 1952. On the sacculinization of J. Charibdis japonica (A. Milne-Edwards). BioI. RAATIKAINEN, M. 1966. The effect of different J. Okayama Univ. 1:84-89. sexes of the parasite Elenchus tenuicornis 1953. On the epicaridization of the fresh­ (Kirby) on the morphology of the adult]ava­ water crab Eriocheir japonicus. Zool. Mag. sella pellucida (F.) (Hom., Delpahacidae). Ann. Tokyo 62:354-361. Entomol. Fenn. 32: 138-146. MAUCHLINE, J. 1966. Thalasomyces [agei an RASMUSSEN, E. 1959. Behaviour of sacculinized ellobiopsid parasite of the euphassid crustacean shore crabs (C arcinus maenas Pennant). Na­ Thysanoesa raschi. J. Mar. bioI. Ass. U.K. 46: ture 183:479-480. 531-539. REINHARD, E. G. 1956. Parasitic castration of MURDOCH, W. W. 1966. Population stability Crustacea. Expl. Parasit. 5: 79-107. and life history phenomena. Amer. Natur. 100: REINHARD, E. G., T. VON BRAND, AND S. F. Mc­ 5-11. DUFFIE. 1947. Observations on the fat con­ MCCLELLAND, G., AND T. K. R. BOURNS. 1969. tent of hermit crabs parasitized by a bopyrid. Effects of Trychobilharzia ocellata on growth, Proc. Helmithol. Soc. Wash., D.C. 14:69-73. reproduction and survival of Lymanaea stag­ REMPEL, J. G. 1940. Intersexuality in Chiron­ nalis. Exptl. Parasit. 24: 137-146. omidae induced by Nematode parasitism. J. NIELSEN, S. 1970. The effects of the Rhizo­ Expl. Zool. 84:261-289. cephalan parasites Pelto gaster paguri Rathke REVERBERI, G. 1943. Sui significato della "cas­ and Gemnosaccus sulcatus (Lilljeborg) on five trazione parassitaria." La trasformazione del species of paguridan hosts (Crustacea: Decap­ sesso nei Crostacei parassitati da Bopiridi e oda). Sarsia 42: 17-32. da Rizocefali. Pubbl, staz. zool. Napoli 19: NILSSON-CANTELL, C. A. 1926. Uber Verander­ 225-316. ungen der sekundaren Geschlechtsmerkmale bei Rt:VERBERI, G. 1944/45. La determinazione del Paguriden durch die Einwirkung von Rhizo­ sesso ne i Crostacei e i fenomeni della castra­ cephalen. Ark. Zool.18A (13):1-21. zione parassitaria. Rend. ist. lombardo sci. mat ~OBLE, E. R., AND G. A. NOBLE. 1971. Parasi­ nat. 78:217-246. tology, the Biology of Animal Parasites. Lea Roasox, G. C. 1911. The effect of Sacculina and Febiger, Philadelphia 617 p. upon the fat metabolism of its host. Quart. J. OKADA, Y. K., AND Y. MIVASHITA. 1935. Sac­ Microscop. Sci. 57:267-278. culinization in Eriocheir japonicus de Haan, ROTHSCHILD, A., AND M. ROTHSCHILD. 1939. with remarks on the occurrence of complete Some observations on the growth of Peringia 352 M. BAUDOIN

uluae (Pennant, 1777) in the laboratory. Novi­ systematic relations of the Mesozoa. Quart. tates Zool. 41:240--247. Rev. BioI. 29:230-244. ROTHSCHIT.D, M. 1936. Gigantism and variation SUMMERFELT, R. C., AND M. C. WARNER. 1970b. in Peringia ulvae Pennant 1777, caused by in­ Geographical distribution and Host-parasite fection with larval trematodes. J. Mar. BioI. relationships of Plistophora ovariae (Micro­ Ass. U.K. 30: 537-546. sporidia, Mosematidae) in Motemigonus cry­ 1938. Further observations on the effect soleucas. J. Wild. Dis. 6:457-465. of trematode parasites of Peringia uluae (Pen­ TRIVERS, R. L. 1972. Parental investment and nant 1777). Nov. Zool. 41:84-102. sexual selection, p. 136-179. In B. Campbell 1941a. The effect of trematode parasites (ed.) Sexual selection and the descent of man on the growth of Littorina neritoides (L.). 1871-1971. Aldine Pbl. Co., Chicago. J. Mar. BioI. Ass. U.K. 25:84-102. VEILLET, A. 1945. Recherches sur Ie parasitisme des Crabes et des Galathees par les Rhizoceph­ 1941b. Observations on the growth and ales et les Epicarides. Ann. inst. oceanog. trematode infections of Peringia ulvae (Pen­ Paris 22(4) :193-341. nant 1777) in a pool in the Tamar Saltings, VEILLET, A., AND F. GRAF. 1958. Degenerescence Plymouth. Parasit. 33:406-415. de la glande androgene des Crustaces decapodes 1941c. Gigantism and variation in Perin­ parasites par les Rhizocephales. Bull. Soc. gia ulvae Pennant 1777, caused by infection scient. Nancy, N.S. 18:123-127. with larval trematodes. J. Mar. BioI. Ass. VEVERS, H. G. 1951. The biology of Asterias U.K. 20:537-546. rubens L. 11. Parasitization of the gonads by ROTHSCHIT.D, M., AND T. CLAY. 1952. Fleas, the ciliate Orchitophya stellarum Cepede. J. flukes and cuckoos. Macmillan, London. Mar. BioI. Assoc. U.K. 29:619-624. RUDLOFF, 0., AND A. VEIT.LET. 1954. Influence WANG, DER I., AND F. E. MOELLER. 1970. The du Rhizocephale Septosaccus cuenoti sur Ie division of labor and queen attendance behav­ metabolisme lipidique du Pagure Diogenes pu­ ior of nosema-infected worker honey bees. J. gilator. Compt. rend. soc. bioI. 148: 1464-1467. Econ. Entomol. 63: 1539-1541. SALT, G. 1927. The effects of stylopization on WECKER, S. C. 1962. The effects of bot fly aculeate Hymenoptera. J. Exp. Zool. 48:223­ parasitisms in a local population of the white­ 231. footed mouse. Ecology 43: 561-565. 1931. A further study of the effects of WELCH, H. E. 1959. Taxonomy, life cycle, de­ stylopization on wasps. J. Expl. ZooI. 59: 133­ velopment and habits of two new species of 163. Allantonematidae (Nematoda) parasitic in dro­ SAMUELSON, T. J. 1970. Peltogaster curuatus sophilid flies. Parasitology 49:83-103. Kossman a Rhizocephalan parasite new to the 1965. Eutomophilic nematodes. Amer. Norvegian fauna with notes on the synonymy Rev. Ent. 10:275-302. of one of the hosts, Pagurus prideauxi Leach. WHEELER, W. M. 1910. The effects of parasitic Sarsia 43:81-86. and other kinds of castration in Insects. J. VON SIEBOLD, G. T. 1843. tiber Strepsiptera. Exp, Zool. 8:377-438. Arch. f. Naturg., Jahrg. 9, H.l, S.:137-162. WIT.LIAMS, G. C. 1966a. Adaptation and nat­ SMITH, G. W. 1905. Note on a gregarine (Ag­ ural selection. Princeton Univ. Press, Prince­ regata inachi n. sp.) which may cause the ton, New Jersey. 307 p. parasitic castration of its host (lnachus dorset­ 1966b. Natural selection, the costs of re­ tensis). Mitth. ZooI. Stat. Neapel 17:406-410. production and a refinement of Lack's prin­ 1913. Studies in the experimental analysis ciple. Amer. Natur. 100:687-690. of sex. Part 10. The effect of Sacculina upon WOODHEAD, A. E. 1930. Life history studies on the storage of fat and glycogen, and on the the trematode family Bucephalidae. No. II. formation of pigment by its host. Quart. J. Trans. Am. Micro. Soc. 49:1-17. Microscop. Sci. 59:267-295. WRIGHT, C. A. 1971. Flukes and snails. Science SMITH, G. W., AND A. H. HAMM. 1914. Studies of biology series No.4. George Allen and Un­ on the experimental analysis of sex. Quart. J. win Ltd., London, p. 168. Microscop. Sci. 60:435-461. WULKER, W. 1961. Untersuchungen iiber die STARK, G. T. C. 1965. Dyplocotyle (Eucestoda) Intersexualitat der Chironomiden (Dipt.) nach a parasite of Gammarus Zaddachi in the estu­ Paramermis-Infektion. Archiv. Hydrobiol. Sup. ary of the Yorkshire Esk. Britain. Parasit. 25: 127-181. 1962. Parasitare und nicht parasitare 55:415-420. geschlechtliche Aberrationem bei Chironomi­ STRICKLAND, E. H. 1911. Some parasites of den. Zool. Anz. Sup. 25:132-139. Similium larvae and their effects on the devel­ 1964. Parasite-induced changes of internal opment of the host. BioI. Bull. 21:302-338. and external sex characteristics in insects. Exp. STUNKARD, H. W. 1954. The life-history and Parasitol. 15:561-597.