6 pages

Oviposition and incubation environmental effects on embryonic diapause in a ground cricket* 1

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Oviposition and incubation environmental effects on embryonic diapause in a ground cricket* 1
   Anim. Behav. , 1998,  55,  331–336 Oviposition and incubation environmental e ff  ects on embryonic diapause in aground cricket ALEXANDER E. OLVIDO*, STEPHEN BUSBY† & TIMOTHY A. MOUSSEAU* *Department of Biological Sciences, University of South Carolina at Columbia † South Carolina Governor’s School for Science and Mathematics (Received 20 February 1996; initial acceptance 2 May 1996; final acceptance 20 January 1997; MS. number:   7529) Abstract.  Maternal e ff  ects on o ff  spring phenotype are well known in organisms inhabiting seasonalenvironments. Mothers that perceive climatological changes indicative of winter’s onset will oftenproduce increasing numbers of o ff  spring that enter a state of arrested morphogenesis known as‘diapause’. In this study of bivoltine  Allonemobius socius  ground crickets, we manipulated the abioticenvironment experienced by ovipositing females (i.e. maternal-oviposition environment) and thatexperienced by o ff  spring incubating as eggs (i.e. egg-incubation environment) to assess the degree towhich mothers contribute to the expression of embryonic diapause in her o ff  spring. Analysis of variancecomponents indicated that variation in maternal-oviposition environment contributed only about 4% tothe total variation in diapause incidence, compared to about 24% from embryos responding directly tovariation in their incubation environment. Moreover, the 8% contribution from between-familyvariation was significant, suggesting that parental genes and maternal biotic and abiotic environmentscontribute to diapause expression in o ff  spring. Although these findings suggest that maternal physi-ology itself contributes little to embryonic diapause variation in  A. socius , they do not preclude othermaternal behaviours (e.g. placement of eggs at di ff  erent soil depths) that may a ff  ect o ff  spring diapause.   1998 The Association for the Study of Animal Behaviour A maternal e ff  ect is a developmental or behav-ioural influence in which ‘genetic or environ-mental di ff  erences in the maternal generation areexpressed as phenotypic di ff  erences in the o ff  -spring’(Mousseau 1991, page 1053). For example, precociously weaned  Mus domesticus  pups showmore play and exploratory behaviour comparedto siblings weaned at regular times from theirmothers (Terranova & Laviola 1995). Swissmice,  M. musculus , develop a preference for theirmother’s diet over a control food source(Valsecchi et al. 1993). In animals with externalfertilization, increased maternal investment toovum size leads to hatching larvae with largerbody size (Oriental fire-bellied toad,  Bombinaorientalis : Kaplan 1991; poeciliid fish: Reznick  1991). Thus, o ff  spring traits can be shaped by thephysiological and/or social milieu provided bymothers.For animals that inhabit seasonal environ-ments, maternal e ff  ects link o ff  spring phenotypeto climate, notably in temperate-zone insects over-wintering in the egg stage (e.g.  Aedes atropalpus :Beach 1978). Female insects that perceive changesin ambient temperature and/or photoperiod layeggs that can enter a state of dormancy knownas embryonic (or egg) diapause (Danks 1987). Unlike quiescence, embryonic diapause is anattentuation of development, or complete cess-ation of morphogenesis, which allows embryosto survive adverse climatic conditions and toresume development during more favourableconditions (Tauber et al. 1986; Danks 1987). Thus, embryonic diapause has major implicationsfor behaviour because it reflects how an animaldeals with environmental challenges.The ground cricket,  Allonemobius socius , pro-vides a useful system for investigating whethermothers provide embryos with climatologicalinformation that initiates embryonic diapause, orwhether o ff  spring detect it directly. Bivoltinepopulations occur in the southeast U.S.A. Correspondence: T. A. Mousseau, Department of Biological Sciences, University of South Carolina,Columbia, SC 29208, U.S.A. (email:–3472/98/020331+06 $25.00/0/ar970624   1998 The Association for the Study of Animal Behaviour 331  (Howard & Furth 1986), with first-generation adults emerging in mid-summer to produceprimarily non-diapausing progeny. As the grow-ing season progresses and winter approaches,however, the incidence of embryonic diapauseincreases, a pattern attributed to adaptivematernal influence by female crickets of increasingage (Mousseau 1991; Bradford & Ro ff   1993).In this study, we used the incidence of embryonic diapause as a measure of a femalecricket’s ability to transmit climatological infor-mation to her developing embryos. By comparingthe relative contribution of maternal-ovipositionenvironment and egg-incubation environment tovariation in diapause incidence, we can assess thedegree to which a mother’s abiotic environmentinfluences development of her o ff  spring. METHODSCricket Stocks All subjects were one generation removed fromthe field, having been produced from laboratory-reared crickets caught as early-instar nymphsfrom a wet, grassy field adjacent to a greenhouseat the University of South Carolina at Columbiaduring June–September, 1994. Prior to the startof this experiment, all crickets were reared andmaintained at 31  C on a 15:9 h light:dark cycle(Olvido & Mousseau 1995). Experimental Design We paired each of 21 females once with males.All mating pairs were housed individually inplastic cages, 9  9  8 cm. Female crickets ovi-posit within 24 h after copulation (Loher &Dambach 1989; A. Olvido, personal observation),so that all matings occurred in an assigned ovi-position environment. We initially assigned 11females an ‘autumn’ oviposition environment(24  C, 11:13 h L:D) at the same time that theremaining 10 were assigned a ‘summer’ ovipos-ition environment (31  C, 15:9 h L:D). We allowedfemales to oviposit at least 20 eggs (per egg batch)in damp cheesecloth strips loosely rolled in plasticvials that we placed against the inside wall of theirmating cage. We collected eggs 7 days from thedate of initial male–female pairing, then collectedbatches of eggs every 7–10 days. Hence, femalecrickets oviposited in their initially assigned en-vironments for at least 14 days before we allowedthose same crickets to oviposit in the alternativeenvironment for a comparable duration.We randomly divided each collection of eggs inhalf. The first half of each egg batch was incu-bated exclusively in either an ‘autumn’ (24  C,11:13 h L:D) or ‘summer’ (31  C, 15:9 h L:D)environment, while the other half was placed inthe alternative incubation environment. This split-brood design allowed (1) half of all developingembryos to experience a di ff  erent environmentfrom the maternal-oviposition environment and(2) properly set-up paired incubation controls,in the form of eggs exposed to an environ-ment identical to their mother’s ovipositionenvironment. Scoring Embryonic Diapause We allowed the o ff  spring 15–18 days to showsigns of non-diapause development, such as eye-spots on eggs (Mousseau & Ro ff   1989) or thepresence of newly hatched nymphs. Fewer than10% of eggs in any treatment rotted (usuallywithin 10–14 days from collection), and these eggswere discarded to avoid confounding diapausewith non-viability of eggs. Diapause incidence wascalculated as the number of healthy, unhatchingeggs (  N  Diapause =2509) divided by the number of healthy eggs (  N  Total =4812) 15 days after wecollected and assigned them to an incubationenvironment. Statistical Analysis To establish conditions of normality inmeasures of diapause incidence, we performed anangular transformation on all percentage dataprior to analysis (Sokal & Rohlf 1981, page 427).We carried out most of the statistical tests usingSAS for Windows, Version 6.10 (SAS 1989). A normal probability plot indicated su ffi cientnormality to justify running a mixed-modelanalysis of variance (ANOVA), which considersboth fixed and random e ff  ects. The full ANOVAmodel appears as follows: Y  ijklm =   . . .+Hatch i +Ovip  j +Family k  +(Hatch  Ovip) ij +(Hatch  Family) ik  +(Ovip  Family)  jk  +(Hatch  Ovip  Family) ijk  +  1(ijk) ,  Animal Behaviour, 55, 2 332  where  Y  ijklm  is variance in the proportion of viableeggs that diapause;    . . . and   1(ijk)  represent thetrue mean and error term, respectively; Hatch i is diapause incidence variation due to egg-incubation environment (fixed), that is, variancein o ff  spring development reflecting the e ff  ects of the abiotic environment experienced by the eggduring the 15–18-day incubation period; Ovip  j  isvariation due to maternal-oviposition environ-ment (fixed), that is, variance in o ff  spring devel-opment reflecting the e ff  ects of the abioticenvironment experienced by the mother duringoviposition; Family k   is variation between families(random), that is, variance in o ff  spring develop-ment reflecting the confounded e ff  ects of parentalgenes and both abiotic and biotic environments of the mother during oviposition. We consideredall interaction terms, except (Hatch  Ovip), asrandom e ff  ects (Neter et al. 1991, pp. 1016–1021). To assess the relative contribution of maternal-oviposition and egg-incubation environments tototal variation in diapause incidence, we used arestricted-maximum likelihood variance com-ponents estimated procedure in SAS, that is, ‘procvarcomp’ (SAS 1989; see also Shaw 1987). Total variance in diapause incidence was partitionedaccording to the terms given in the full ANOVAmodel.We calculated statistical power of each  F  -testusing power analysis software (Hintze 1992). Statistical power is measured as (1   ), where   equals the probability of committing a Type IIerror. Hence, power indicates the probability of correctly detecting a significant treatment e ff  ectgiven the size of the e ff  ect, sampling variance,sample size and a fixed Type I error rate (  =0.05)(Sokal & Rohlf 1981, pp. 157–169, 262–264;Fairweather 1991). RESULTS Both maternal-oviposition and egg-incubationenvironments (‘Ovip’ and ‘Hatch’, respectively)showed significant e ff  ects on diapause incidence(Table I). Mothers tended to lay more diapausing o ff  spring in the autumn than in the summeroviposition environment (Fig. 1). The o ff  -spring themselves responded similarly to di ff  erentseasonal climates: diapause incidence was higherin the autumn versus the summer incubationenvironment (Fig. 1). Significant between-family di ff  erences (‘Family’ in Table I) suggest that maternal-oviposition environment, confoundedwith genetic factors (Falconer 1989, pp. 158–160), contributed to the observed diapause incidencepatterns.Two of the four interaction terms were signifi-cant, despite low statistical power (1   <0.80).There was a significant interaction betweenmaternal-oviposition and egg-incubation environ-ments (‘Ovip  Hatch’ in Table I), probablybecause the e ff  ects of maternal ovipositionenvironment were much greater when eggs wereincubated in the autumn environment (Fig. 1).The highly significant interaction between Table I.  Results of a mixed-model analysis of variance on percentage of diapausing eggsSource  df   Type III MS  F   1-  Hatch 1 5.564 61.20*** 0.903Ovip 1 1.456 16.02*** 0.137Family 20 0.514 5.65*** 0.994Hatch  Ovip 1 0.479 5.26* 0.146Hatch  Family 20 0.146 1.61   0.209Ovip  Family 20 0.286 3.15*** 0.562Hatch  Ovip  Family 20 0.073 0.80   0.209Error 84 0.091Total 167An angular transformation was performed on all percentage data prior to any analysis.For the power analysis, Type I and Type II error rates were set at 0.5 and 0.20,respectively. Main e ff  ects are: Ovip=the abiotic environment experienced by ovipositingfemales (‘autumn’ versus ‘summer’); Hatch= the abiotic environment experienced byo ff  spring incubating as eggs (‘autumn’ versus ‘summer’).* P <0.05; ** P <0.01; *** P <0.001;   ,  P >0.05. Olvido et al.: Adaptive maternal e  ff  ects?  333  maternal-oviposition environment and family(‘Ovip  Family’ in Table I) reflects environment-specific genotypic expression, i.e. G  E inter-action.Egg-incubation environment ( V  Hatch ) con-tributed approximately six-fold more to diapauseincidence than did maternal-oviposition environ-ment ( V  Ovip ) (Table II). Variance in the inter- action between family and maternal-ovipositionenvironment (‘Ovip  Family’), as well as betweenfamilies (‘Family’), significantly contributed tototal variation in diapause incidence (Tables Iand II).The variance contributions from the inter-actions between egg-incubation environment andfamily (‘Hatch  Family’) and between egg-incubation and maternal-oviposition environ-ments and parental genes (‘Hatch  Ovip  Family’) were less than 6% combined (Table II). The variance contribution from the inter-action between egg-incubation and maternal-oviposition environments and parental genes( V  Hatch  Ovip  Family  in Table II) was zero, suggesting that the three-way interaction isnegligible or absent. DISCUSSION In bivoltine populations of   A. socius , maternalenvironment determines whether embryos enterdiapause. Our results showed that diapause inci-dence in ovipositing female crickets tends toincrease from a summer to an autumn environ-ment (Fig. 1). One explanation is that maternal influences are incidental by-products of physio-logical processes detected by o ff  spring. Alterna-tively, embryonic diapause may be initiated by‘diapause hormone’, which in many lepidopteransoriginates from the suboesophageal ganglionof mothers responding to shortened photo-period (Fukuda 1951; Hasegawa 1957). A similar physiological mechanism may operatein maternally induced embryonic diapause of   A. socius , and probably involves other maternallyderived hormones (Denlinger 1985; Mousseau 1991).A post hoc ANOVA failed to show an e ff  ect of maternal-oviposition order on diapause incidence( F  3,152 =1.82;  P =0.1457). This finding suggeststhat a female cricket laid diapausing eggs inresponse to her immediate oviposition en-vironment and not as a delayed response to herprevious environment; that is, there were no‘carry-over e ff  ects’ from a 7–10 day exposure to aparticular oviposition environment.Embryos themselves respond to their environ-ment. When we controlled for maternal-oviposition environment, diapause incidencewas higher in an autumn versus a summeregg-incubation environment (Fig. 1). Kidokoro & Masaki (1978) mentioned a similar phenomenon 1.00.0    P  r  o  p  o  r   t   i  o  n   d   i  a  p  a  u  s   i  n  g 0.40.2 Autumn Summer0.80.6Egg-incubation environment Figure 1.  A bar graph showing the e ff  ect of bothmaternal-oviposition and egg-incubation environments.An angular transformation was performed on all per-centage data prior to analysis. Height of each histogramcolumn gives the back-transformed arithmetic meanfor that treatment. Vertical bars indicate width of theback-transformed 95% confidence intervals. Maternal-oviposition environment: , autumn; , summer. Table II.  Restricted-maximum likelihood estimates of variance components of diapause incidenceSourcePercentageof totalHatch 23.8Ovip 3.7Family 8.4Hatch  Ovip 3.7Hatch  Family 5.8Ovip  Family 19.8Hatch  Ovip  Family 0.0Error 34.8Total 100.0Sources of variation are coded as in Table I.  Animal Behaviour, 55, 2 334  in a Japanese ground cricket,  Dianemobius fascipes  (formerly  Pteronemobius fascipes ;Shimizu & Masaki 1993). Apparently,  A. socius embryos can modify their own developmentalprocesses in response to seasonal cues (e.g.decreasing ambient temperature).The e ff  ects of both maternal and embryonicenvironments reinforce each other, as indicated bythe general drop in diapause incidence whenenvironmental conditions shifted from autumn tosummer (Fig. 1). Several lines of evidence suggestthat embryos are responsive enough to climato-logical shifts to alter maternally induced diapause:(1) the significant interaction between maternal-oviposition and egg-incubation environments(Table I); (2) the non-parallel drop in diapause incidence across egg-incubation environments andwithin maternal-oviposition environment (Fig. 1);and (3) the substantial contribution of egg-incubation environment to total variation indiapause incidence (Table II). Riska (1991, page 720) proposed that maternale ff  ects are ‘strongest in the early stages of development and growth’. This notion is logicalsince embryos can gauge changes in the environ-ment as they progressively develop a functionalnervous system. We therefore speculate that  A. socius  embryos are less subject to maternalinfluence (e.g. maternally induced diapause) atlater stages of development.The pattern of decreasing intensity of maternale ff  ects with time is not restricted to natural insectpopulations (Mousseau & Dingle 1991a, b). In laboratory mice, dietary supplements of manganese inhibit expression of the mutation  pallid   only when administered early in the pup’slife, that is, either directly or via milk fromthe pup’s mother fed a high-manganese diet(Roubertoux et al. 1990). Similarly, the magnitude of uterine genotypic e ff  ects in mice 3–12 days afterbirth was 7% of the average gain in progeny taillength (an index of body size), but was non-significant thereafter (Cowley et al. 1989). In general, maternal e ff  ects subside as o ff  springdevelop.We conclude that  A. socius  mothers do not‘programme’embryonic diapause (i.e. through thephysiology of oogenesis) in response to seasonalchanges. Other maternal behaviours, such aschoice of oviposition substrate (Loher &Dambach 1989; Walker & Masaki 1989) or soil depth (T. A. Mousseau & A. Anderson,unpublished data), and embryonic responseto environment may be more important toembryonic diapause expression in wild  A. socius populations. ACKNOWLEDGMENTS Special thanks to the Busby family for helpin collecting data, to Holmes Finch in theDepartment of Statistics for statistical advice, andto Jerry Hilbish, Chuck Fox and Sally Woodin forfruitful discussion. Chuck Fox, Fran Groeters,Hugh Dingle, David Wise, Fred Dyer and threeanonymous referees gave valuable critiques onearlier versions of the paper. Extra special thanksto Patti Loesche for substantially clarifying thetext. This study was supported by a Ford pre-doctoral fellowship to A. E. Olvido, by a SouthCarolina University Research and Educationfellowship to S. Busby and by NSF grantDEB-9409004 to T. A. Mousseau. REFERENCES Beach, R. 1978. The required day number and timelyinduction of diapause in geographic strains of themosquito,  Aedes atropalpus .  J. Insect Physiol. ,  24, 449–455.Bradford, M. J. & Ro ff  , D. A. 1993. Bet hedging andthe diapause strategies of the cricket  Allonemobius fasciatus .  Ecology ,  74,  1129–1135.Cowley, D. E., Pomp, D., Atchley, W. R., Eisen, E. J. &Hawkins-Brown, D. 1989. The impact of maternaluterine genotype on postnatal growth and adult bodysize in mice.  Genetics ,  122,  193–203.Danks, H. V. 1987.  Insect Dormancy: An EcologicalPerspective . Ottawa, Ontario: Biological Survey of Canada.Denlinger, D. L. 1985. Hormonal control of diapause.In:  Comprehensive Insect Physiology, Biochemistry,and Pharmacology, Vol. 8   (Ed. by G. A. Kerkut &L. I. Gilbert), pp. 353–411. Oxford: Pergamon Press.Fairweather, P. G. 1991. Statistical power and designrequirements for environmental monitoring.  Austral. J. mar. Freshwat. Res. ,  41,  555–567.Falconer, D. S. 1989.  Introduction to QuantitativeGenetics . 3rd edn. Essex: Longman Scientific &Technical.Fukuda, S. 1951. Factors determining the production of non-diapause eggs in the silkworm.  Proc. Japan. Acad. ,  27,  582–586.Hasegawa, K. 1957. The diapause hormone of thesilkworm,  Bombyx mori .  Nature, Lond. ,  179,  300–301.Hintze, J. L. 1992.  SOLO Statistical System Power  Analysis . Los Angeles, California: BMDP StatisticalSoftware. Olvido et al.: Adaptive maternal e  ff  ects?  335
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!