Seasonality, developing time and protandry in three populations of the

neotropical grasshopper Sphenarium histrio in an altitudinal gradient

Víctor Hugo Ramírez-Delgado & Raúl Cueva del Castillo*

UBIPRO, Laboratorio de Ecología, Universidad Nacional Autónoma de México, FES

Iztacala, A.P. 314, Tlalnepantla, 54090. México, México

*Corresponding author: [email protected]

Running Head: Seasonality and protandry in Sphenarium histrio

Key Words: Sexual selection, natural selection, protandry, local adaptation, sexual

dimorphism
 ABSTRACT

Large body size is correlated with high fitness. A larger body size could be consequence

of longer development time. In seasonal environments organisms must mature and

produce offspring on a short period of time. In that condition, a fast development could

be favored by selection. Because the differences in pressures selection between sexes,

females and males could adopt different strategies, like sexual size dimorphism and

development time differences. We evaluated the traits of history life and sexual

dimorphism in three populations of the grasshopper Sphenarium histrio from different

altitudinal sites and environments in a common garden experiment. Due to the lower

seasonality, grasshoppers from low lands are bigger, have longer developing times, have

more nynphal instars, and females are more fecund but growth rates were not different

between populations. Nonetheless, males from the midland population are protandrus,

which suggest that protandry is adaptive and not a byproduct of selection on male’s body

size. Seasonality should be consider as important factor in change climate models due it

has an important impact in the life cycle, body size, and inter sexual variation of the

neotropical insects.

Key words: Body Size, Tropical Grasshopper, Life Cycle, Protandry, Sexual Size

Dimorphism, Development Time, Clinal Variation.

The emergence or sexual maturation of males prior to females is known as protandry and

is a common phenomenon in insects and other animal with not overlapping generations

(Thornhill and Alcock 1983, ). Protandry can be adaptive if early mature males

mate with the first sexually mature females that emerge in the season and/or if they

obtain a higher number of matings than males that mature late (Wiklund and Fagerstrom

1977; Thornhill and Alcock 1983, Cueva del Castillo and Núñez Farfán 1999). Protandry

maximize the males mating opportunities if there is temporal variation in female

availability, thus in stronger seasonal environments it is expected a greater level of

protandry (Singer,1982; Wiklund, 1995, Morbey 2013). Nonetheless, protandry cannot be

adaptive, because in many insect species females are bigger than males and have longer

development periods (Wiklund and Fagerstrom 1977; Thornhill and Alcock 1983), it has

been suggested that males can reach the adult stage in less time than females (Wiklund &

Fagerström, 1977). Under this argument, interpopulation differences in the degree of

protandry must be associated to size and sexual dimorphism levels. Moreover,

protandrous males should be smaller than those males that mature synchronically with

females, although if the growth rates in males are higher than in females, the former can

reach similar sizes to females in less time, cushioning differences in body size between

females and males (Ayres & Scriber, 1994).

The natural selection pressures can differ in direction and / or magnitude,

generating local adaptation (Blanckenhorn, 2007, Olvido & Mousseau, 2012). In places

with higher altitude or far from the equator, the time available to complet the

development is less than sites with lower seasonality, so at high elevations natural
selection favors a faster maturation and reproduction than in lowlands [7–10). However,

the inter population phenotype difference also can be a result of phenotypic plasticity

(Roff, 1980). If natural selection has molded the genetic variance of populations in

different ways, by exposing organisms from contrasting environments to a common

environment during their development, we would expect that at some level the

phenotypic differences between populations will be maintained.

In this study, in a common garden experiment we evaluated the possibility of

protandry (developing time), growth rate and adult body size in three populations of

Sphenarium histrio in an altitudinal gradient in Chiapas, a southern state of Mexico. As

result of local adaptation, we expected that the developing time and body size were

smaller at high altitudes. Moreover, that we would find protandry or higher levels of

protandry at a high altitude, and associated to higher levels of sexual size dimorphism

(SSD) bias to females. Nonetheless, if the phenotypes response were mainly plastic, we

would expect that the phenotypic differences would decrease as they developed in the

same environment (Olvido & Mousseau, 2012).

S. histrio is an univoltine neotropical Pyrgomorphidae grasshopper (Uvarov

1977). The species is distributed in elevations ranging approximately from 15 to 2225 m

a.s.l. in Chiapas, Guerrero, Oaxaca, Tabasco, Veracuz, in southern Mexico; and in the

states of Huehuetenango and Santa Rosa in Guatemala (Sanabria-Urbán et al., 2015). Due

to its wide altitudinal distribution range the species is found in a variety of climates and

seasonality (Kevan, 1977; Sanabria-Urbán et al., 2015). Nymphal S. histrio emerge at the

beginning of the rainy season (July-August Cueva del Castillo pers obs.). In places with a

lower seasonality adults can live until the end of February (Sanabria-Urbán et al., 2017).
Considerable variation in body size has been recorded in both sexes in the field. This

species is sexually dimorphic: females have a larger thorax than males, and femur I and II

are wider in males (Sanabria-Urbán et al., 2015, 2017).

MATERIAL AND METHODS

SAMPLING

Adult females an males of S. histrio were collected in late October 2014 from three

localities from Chiapas, Mexico: Chiapa de Corzo and Arriaga, which are dry deciduous

tropical forests, and Teopisca; a coniferous forest. These localities were chosen because

despite that the geographical distances among them are relatively short, their altitudinal

(and climatic) differences are large (see Table 1). Moreover, a previously study in neutral

genes (ribosomal DNA S12, transcription rRNA S5.8 and rRNA 28S) showed that they

are not genetically differentiated (Sanabria-Urbán et al., 2015, 2017). Thus, the

phenotypic differences among these grasshopper populations may be due to local

adaptation.

The adult grasshoppers were collected by hand or or sweep nets, placed in plastic

recipients and transported to the laboratory. The third Femur 3 Length (3FL), the first and

second Femur Width (1FW, 2FW), and the Thorax Width TW of each adult were

measured. A digital caliper was used to take the body size measures to the nearest 0.1

mm (Mitutoyo Corp., Tokyo, Japan). The 3FL is a general indicator of body size because

this trait is usually high correlated with body size in Sphenarium grasshoppers (Cueva del

Castillo et al., 1999, Sanabria-Urbán et al. 2015), whereas TW, IFW and 2FW are
sexually dimorphic traits. In order to test phenotypic differences, the measurements of the

four morphological traits for females and males of the three populations were analyzed.

LABORATORY HANDLING

Grasshoppers from the three populations were maintained in the laboratory in the same

conditions. The collected stock from each population was placed in 7- liter plastic

containers. In each container 15 females and 15 females were placed. Before to place the

grasshoppers in the container, each one of them was provided by a seven cm commercial

flower potting soil layer as oviposition substrate. The organisms were maintained at 29 ±

3 °C on a photoperiod of 12 : 12 h (L:D). Every third day the grasshopper were feed up

ad libitum with Apium graveolens. In order to avoid dehydration the containers were

daily sprayed out ad libitum with water. When an individual dye, it was removed from

the container and preserved in ethanol at 70%.

Ones all the individuals in the containers die, we inspected the soil looking for egg pods.

Each egg pod was transferred to 1-liter plastic jar and covered with 3 cm flower potting

soil layer maintaining the population and mother (egg pod) identity. In order to simulate

the rainy season, starting on February 25th, 2015, every plastic jar was daily sprayed out

with 40 ml of water. The firs day of watering was considered as the onset of the rainy

season. In order to ensure that water was not a hatching limitation, the water supply was

maintained for all the jars until 45 days after the last day that a nymph of any population

was found in the plastic jars. Then we recorded the number of no hatching eggs

remaining in each jar. The first hatching day was considered as the onset of hatching

period emergence. Consecutive hatching day were considered until the last day that a
hatching from any population happens. The hatched nymphs were placed individually in

100 ml plastic recipients, and daily feed up with celery leaves ad libitum until they reach

the adult instar. For each grasshopper the hatching day, number of molts, and instars were

recorded. Moreover, in order compare inter and intra population onset of adult emergence

we considered the beginning of the reproductive season for the three populations and the

beginning of the season for each population separately. The former was estimated as the

first day that any individual(s) of any population reached the adult intar, whereas for each

population it was estimated as the first day that any individual(s) of the population

reached the adult intar. Such day was considered as the onset of adult emergence (OAE).

Consecutive OAE were considered until the last day that an individual of any population

was found in the jars. The OAE of each collected organism was considered for further

analyses as a continuous variable.

In order to estimate the growing trajectories of each grasshopper, we took digital pictures

of the dorsal view of each instar of the individuals using a digital camera ( Marca??

Modelo??), then we measured the third femur length (3FL) and thorax width (TW) of

each instar using the software imageJ (Abràmoff, Magalhães, & Ram, 2004). The adults

3FL, the first and second femur width (1FW, 2FW), and the TW of each adult were

measured. A digital caliper was used to take the body size measures to the nearest 0.1

mm (Mitutoyo Corp., Tokyo, Japan). In addition, individual Growth rate was estimated

as follows: (adult 3FL – nymph 3LF) / developing time.

STATISCAL ANALYSIS
Inter population differences in adult body size collected in the field

In order to evaluate inter population differences in adult body size collected in the field

we analyzed differences between females and males for the four morphological traits

performing a MANOVA, considering Population, Sex and the interaction Population *

Sex as independent variables. Because the MANOVA was highly significant (see below),

univariate tests were performed for each variable.

Inter population differences in adult body reared in the laboratory

In order to evaluate inter population differences in adult body size reared in the

laboratory that reached the adult instar, we analyzed differences between females and

males for the four morphological traits performing a MANOVA, considering Population,

Sex and the interaction Population * Sex as independent variables. Because the

MANOVA was highly significant (see below), univariate tests were performed for each

variable.

Inter population differences in hatching day, number of instars, OAE and growth

rates

In order to evaluate inter population differences in hatching day, number of instars, OAE

and growth rates we performed a MANOVA, considering Population, Sex and the

interaction Population * Sex as independent variables. The 3FL (an indicator of the

grasshoppers general body size) was used to estimated the growth rates as follow: (adult

3FL – nymph 3LF) / developing time. Previous to the analysis ORS values were square

root transformed and adult body size measurements were log transformed in order to fit
the parametric assumption of the residuals distribution. Because the MANOVA was

highly significant (see below), univariate tests were performed for each variable.

Intra population OAE differences

In order to evaluate if males mature significantly earlier than females (protandry) females

and males OAE (square roots) were compared by t test in each population.

RESULTS

In field 533 grasshoppers from the three populations were collected. In the laboratory 151

egg pods were obtained, 1425 nymphs hatched, but just 337 became adults. All the

analyzes from the laboratory results were performed on the data of the 337 female and

male grasshoppers that reached the adult instar.

Inter population differences in adult body size collected in the field

In field 531 grasshoppers were collected, 282 from Arriaga (Females: 135; Males: 147),

135 from Chiapa de Corso (Females: 71; Males: 64), and 114 from Teopisca (Females:

52; Males: 62). The MANOVA on the four female and male morphological traits for the

three populations analyzed was highly significant (Wilk’s  = 0.020, F(25, 1936.9) = 143.96,

P < 0.0001). The analyzes detected inter population differences in body size traits:

(Wilk’s  = 0.233, F(10, 1042) = 111.370, P < 0.0001); sexual size dimorphism: (F(5, 521) =

8.586, P < 0.0001), and differences in the magnitude o sexual size dimorphism among

populations, interaction population*sex: (Wilk’s  = 0.940, F(10, 1042) = 3.251, P<

0.0004). Univariate test for the four body size traits were significant. 3FL: Population: F(2,
525) = 599.90; P< 0.0001, Sex: F(1, 525) = 9.51; P< 0.0022, Population*Sex: F(2, 525) = 3.06;

P= 0.047. 1FW: Population: F(2, 525) = 101.097; P < 0.0001, Sex: F(1, 525) = 263.482; P <

0.0001, Population*Sex: F(2, 525) = 4.32; P = 0.013. 2FW: Population: F(2, 525) = 132.530; P

< 0.0001, Sex: F(1, 525) = 1242.226; P < 0.0001, Population*Sex: F(2, 525) = 8.500; P =

0.0002. TW: Population: F(2, 525) = 240.689; P < 0.0001, Sex: F(1, 525) = 701.401; P <

0.0001, Population*Sex: F(2, 525) = 11.183; P < 0.0001. In Arriaga and Teopisca were

collected respectively the largest and the smaller grasshoppers (Fig. a-d).

Inter population differences in adult body size reared in the laboratory

In laboratory 337 grasshoppers reached the adult instar, 289 from Arriaga (Females: 149;

Males: 142), 24 from Chiapa de Corzo (Females: 10; Males: 14), and 22 from Teopisca

(Females: 17; Males: 7). The grasshoppers reared in the laboratory were smaller than

their fathers collected in the field. Nonetheless, the relative differences among them

remain, grasshoppers which their fathers came from Arriaga and Teopisca were

respectively the larger and the smaller ones (Fig. a-d). The MANOVA on the four

female and male morphological traits for the three populations analyzed was highly

significant (Wilk’s  = 0.035, F(20, 1088.8) = 90.585, P < 0.0001). The analyzes detected

inter population differences in body size traits: (Wilk’s  = 0.487, F(8, 656) = 35.555, P<

0.0001); sexual size dimorphism: (F(4, 328) = 257.129, P < 0.0001). However, the

differences in the magnitude o sexual size dimorphism among populations was not

significant, interaction population*sex: (Wilk’s  = 0.970, F(8, 656) = 1.222, P = 0.283).

Univariate test for the four body size traits were significant. 3FL: Population: F(2, 331) =
147.392; P< 0.0001, Sex: F(1, 331) = 7.824; P< 0.0022, Population*Sex: F(2, 331) = 3.844; P=

0.022.

1FW: Population: Population: F(2, 331) = 45.316; P< 0.0001, Sex: F(1, 331) = 46.859; P<

0.0022, Population*Sex: F(2, 331) = 0.620; P= 0.586.

2FW: Population: F(2, 331) = 28.392; P< 0.0001, Sex: F(1, 331) = 129.173; P< 0.0001,

Population*Sex: F(2, 331) = 1.033; P= 0.357.

TW: Population: F(2, 331) = 55.884; P< 0.0001, Sex: F(1, 331) = 176.516; P< 0.0022,

Population*Sex: F(2, 331) = 1.269; P= 0.283.

Inter population differences in hatching day, number of instars, growth rates and

OAE

The MANOVA was highly significant (Wilk’s  = 0.336, F(2o, 1088.8) = 21.164, P<

0.0001). The analyzes detected inter population differences: (Wilk’s  = 0.271, F(8, 656) =

45.616, P = 0.654); no differences between females and males: (F(4, 328) = 0.6.54, P=

0.0001), and a significant interaction population*sex: (Wilk’s  = 0.947, F(8, 656) = 2.276,

P = 0.021).

Inter population differences in hatching day

The univariate test detected inter population differences in hatching day; Population: F (2,

331) = 38.502; P< 0.0001. However there were not differences between females and males;

Sex: F(1, 331) = 0.394; P< 0.0022, and the hatching pattern was similar for females and

males in the three populations Population*Sex: F(2, 331) = 1.269; P= 0.283.

The univariate test detected inter population differences in the number of instars;

Population: F(2, 331) = 70.539; P< 0.0001. However there were not differences between

females and males in the number of instars; Sex: F(1, 331) = 0.844; P = 0.358, and the

instars was similar for females and males in the three populations Population*Sex: F(2, 331)

= 1.505; P= 0.224.

Inter population differences in growth rates

The univariate test detected inter population differences in the number of instars;

Population: F(2, 331) = 4.448; P = 0.012. However there were not differences between

females and males in growth rates; Sex: F(1, 331) = 0.044; P = 0.834. Nonetheless, there

were inter population differences in the growth rates from females and males;

Population*Sex: F(2, 331) = 3.731; P= 0.025.

Inter population differences in OAE

The univariate test detected inter population differences in the OAE; Population: F (2, 331) =

38.502; P = 0.012. However there were not differences between females and males in

OAE; Sex: F(1, 331) = 0.374; P = 0.542, and the females and males OAE pattern from the

three populations; Population*Sex: F(2, 331) = 0.389; P= 0.677.

Intra population female and male OAE differences

( X females = 32.16 days, X males = 32.55 days), t = 1.06, DF = 287, P = 0.14, Tepisca: ( X females

= 32.16 days, X males = 32.55 days), t = 1.54, DF = 20, P = 0.10. However, in Chiapa de

Corzo the population was protandrous ( X females = 22.36 days, X males = 16.15 days), t = 2.34,

DF = 22, P = 0.013.

Discussion

Our results suggest the seasonality and the time available for reach the adulthood may be

the force shaping many of the history life traits and protandry of Sphenarium histrio

populations. We found in both, field and laboratory, interpopulation differences in body

size and sexual size dimorphisms. In addition, there are interpopulation differences in

eclossion, hatching, growth rate and maturation time. Teopizca, and Arriaga, respectively

the higher and lower altitude populations showed the earlier and later response. Even

though Teopizca had the earlier maturation time, Chiapa de Corso was the only

protandrous population. S. histrio body size can be affected by environmental conditions

like diet and/or temperature ([4]. The grasshoppers collected in the field were larger than

the organisms reared in the laboratory. However, despite this environmental component,

the relative interpopulation differences in body size, eclossion, hatching, growth rate and

maturation time suggest some level of local adaptation. The genetic differentiation may

be due stochastic factors like genetic drive or selective pressures (Olvido & Mousseau,

2012). Nonetheless, because these populations are not differentiated in neutral genes, the

differences among them can be the result of natural and sexual selection.
 Generally, seasonality increases with elevation, constraining the available time for

development and reproduction [7,8]. This can favor the evolution of shorter development

times, decreased time to reach maturity, and smaller adult body size at high elevations

than in lowlands [7–10]. By shortening the development time, the possibility of a pre-

reproductive death is reduced (Wiklund & Fagerström, 1977; Thornhill & Alcock, 1983).

However, this can have a negative impact in fecundity and longevity (Berner, Körner and

Blanckenhorn, 2004, Blanckenhorn & Demont, 2004, Chown and Gaston, 2010).

The mating possibilities also decrease, since protandrous males with frequencies are

smaller than males that mature synchronously with females (Wiklund & Fagerström,

1977; Thornhill & Alcock, 1983).

Growth rates are positively affected by temperature, food quantity and quality [4];

whereas development time is constrained by seasonality, which determines the duration

of optimal developmental conditions, such as temperature and food availability [5,6].

In this study in three populations of the neotropical grasshopper S. histrio in a common

garden experiment we found a trade-off between development time and body size across

the populations but not inside the populations. Our results suggest the seasonality and the

time available for reach the adulthood may be the force shaping many of the history life

traits and protandry of the grasshopper’s populations. Also, we found evidence

suggesting there is genetic differences in body size between the grasshopper’s

populations possibly due natural selection and local adaptation.

The trade-off between development time and body size across populations is consistent

with the evidence found in other grasshoppers (Blanckenhorn, 1998; Plaistow et al.,

2005; Berner & Blanckenhorn, 2006) and many other arthropods (reviewed by

Blanckenhorn et al., 2011). This trade-off usually occurs due the seasonality (Roff, 1980,

2000). Therefore the grasshoppers populations from higher elevation and pronounced

seasonality shown faster development time, smaller body size, and fewer number of

molts but with growth rates similar like the other populations.

Nevertheless, inside the populations the trade-off of the development time and body size

could disappear perhaps because the local adaptation of body size to specific condition in

every locality, with restricted genetic flow between populations (Kawecki & Ebert,

2004). The specific selective pressures in the localities could homogenize the body size

inside the populations via natural selection and maintain the differences between

populations at the same time (Nuismer & Gandon, 2008). Grasshoppers of the genus
Sphenarium get competitive advantage if they reach maturity early and great body size

(Cueva del Castillo & Núñez-Farfán, 2002), however the optimization of the trade-off in

different sites should vary accord with the time available for reach maturity, the food

available, the abundance of coespecifics and the danger to death before the reach the

maturity in every locality (Roff, 1980).

The body size similarities patterns between the grasshoppers from the field and the

laboratory suggests already the populations are genetically differentiated (Kawecki &

Ebert, 2004) the genetic differentiation may be due stochastic factors like genetic drive or

selective pressures (Olvido & Mousseau, 2012).

This could also support the hypothesis of local adaptation of the populations, above

mentioned. The overall smaller adults obtained in the laboratory respect to the collected

adults in field could indicate that the diet in the laboratory not be optimal and is

indication of certain degree of plasticity in the body traits. Plasticity and genetic

differentiation are not excluding alternatives (Mousseau & Roff, 1989; Laiolo, Illera, &

Obeso, 2013). It is possible both effects are present in the grasshopper from the 3

localities.

In respect to fecundity. Arriaga due its low elevation and less seasonality favor the

biggest and more fecund females. Usually the fecundity in invertebrates is correlated with

the body size of the females (Honek, 1993; Cueva del Castillo, Núñez-Farfán, & Cano-
Santana, 1999). This was also found in other species of Sphenarium (Cueva del Castillo

et al., 1999) and in other grasshoppers (Akman & Whitman, 2008).

The protandry found in Corzo cannot be explain for the Wiklund & Fagerström, (1977)

hypothesis, explained before. In Corzo population the sexes don’t differ between

development time, therefore protandry is not consequence of SSD. However, the males

arrive before to adulthood than females. This implies the males hatched before the

females and the causes behind the protandry may be a faster embryonic development of

the males (Berner & Blanckenhorn, 2007) or earlier diapause break in males (Fielding,

2006). This hypothesis possibly will be addressed in future works.

Also is interesting that only in Corzo is founded protandry. Corzo is the locality with less

precipitation due less intense or shorter raining season. Precipitation could be a good

indicator of available food for herbivore insects (Hodkinson, 2005). So it’s possible

Corzo is the locality with less food or shorter period with available food. Seasonality and

narrow periods for mating could favor protadry via maximizing the opportunities for

mating (Wiklund & Forsberg, 1991) or reducing the chances for die before mating

(Cueva del Castillo & Núñez-Farfán, 2002; Morbey, 2013).

We provide evidence suggesting there are genetic differences due local adaptation

between populations but some limitations deserving be noted. Is necessary complement

our results with other experiments like genetic analysis in gens associated with selective

traits, population genetics studies for evaluation of the heritability, or doing other

transplant experiments with more reliable fitness assessments are possible alternatives.

As well, we consider a best evaluation of seasonality like periodical long term

help to improve the understanding of dynamics in traits related to seasonality.

In summary, the variation in body size and development time in S. histrio supports the

trade-off between these traits. Also we found evidence that suggests S. histrio body size

is locally adapted. Thus an important factor causing local adaptation must be the

differences in seasonality of the different populations studied and probably the

environmental variation of each site has molded genetic differences in history life traits in

the populations. The protandry found in Corzo could be explained for faster embryonic

development or early diapause break in males. This hypothesis should be tested in the

future. Seasonality is a variable with profound implication in the life cycle, body size,

and many other related traits in neotropical insects. It is related with a clinal variation of

the populations and probably with many other phenomena. Seasonality should be

consider as important factor influencing tropical insects in models of climate change.

More future studies addressing the implication of the seasonality in the life story of

tropical insects should be necessary.

References

Akman O & Whitman D. 2008. Analysis of body size and fecundity in a grasshopper.

Journal of Orthoptera Research 17: 249–257.

Berner D & Blanckenhorn WU. 2006. Grasshopper ontogeny in relation to time

constraints: Adaptive divergence and stasis. Journal of Animal Ecology 75: 130–139.

Berner D & Blanckenhorn WU. 2007. An ontogenetic perspective on the relationship

between age and size at maturity. Functional Ecology 21: 505–512.

Body Size in the Yellow Dung Fly. Evolution 52: 1394–1407.

Blanckenhorn WU, Stillwell RC, Young KA, et al. 2011. When Rensch Meets

Bergmann : Does Sexual Size Dimorphism Change Systematically With Latitude?

Evolution 60: 2004–2011.

Cueva del Castillo R & Núñez-Farfán J. 2002. Female mating success and risk of pre-

reproductive death in a protandrous grasshopper. Oikos 96: 217–224.

Cueva del Castillo R, Núñez-Farfán J & Cano-Santana Z. 1999. The role of body size in

mating success of Sphenarium purpurascens in Central Mexico. Ecological Entomology

24: 146–155.

Fielding D. 2006. Optimal diapause strategies of a grasshopper, Melanoplus sanguinipes.

Journal of insect science 6: 1–16.

Hodkinson ID. 2005. Terrestrial insects along elevation gradients: species and

community responses to altitude. Biological Reviews 80: 489–513.

Honek A. 1993. Intraspecific variation in body size and fecundity in insects : a general

relationship. Oikos 66: 483–492.

Kawecki TJ & Ebert D. 2004. Conceptual issues in local adaptation. Ecology Letters 7:

1225–1241.

Laiolo P, Illera JC & Obeso JR. 2013. Local climate determines intra- and interspecific

variation in sexual size dimorphism in mountain grasshopper communities. Journal of

Evolutionary Biology 26: 2171–2183.

Theoretical Biology 339: 93–99.

Mousseau TA & Roff DA. 1989. Adaptation to Seasonality in a Cricket: Patterns of

Phenotypic and Genotypic Variation in Body Size and Diapause Expression Along a

Cline in Season Length. Evolution 43: 1483–1496.

Nuismer SL & Gandon S. 2008. Moving beyond Common‐Garden and Transplant

Designs: Insight into the Causes of Local Adaptation in Species Interactions. The

American Naturalist 171: 658–668.

Olvido AE & Mousseau TA. 2012. Geographical Variation. eLS.

Plaistow SJ, Tsuchida K, Tsubaki Y, et al. 2005. The effect of a seasonal time constraint

on development time, body size, condition, and morph determination in the horned beetle

Allomyrina dichotoma L. (Coleoptera: Scarabaeidae). Ecological Entomology 30: 692–

699.

Roff D. 1980. Optimizing Development Time in a Seasonal Environment : The ’ Ups and

Downs ’ of Clinal Variation. International Association for Ecology 45: 202–208.

Roff DA. 2000. Trade-offs between growth and reproduction: an analysis of the

quantitative genetic evidence. Journal of Evolutionary Biology 13: 434–445.

Wiklund C & Fagerström T. 1977. Why do males emerge before females? Oecologia 31:

153–158.
Wiklund C & Forsberg J. 1991. Sexual Size Dimorphism in Relation to Female

Polygamy and Protandry in Butterflies: A Comparative Study of Swedish Pieridae and

Satyridae. Oikos 60: 373–381.