<%BANNER%>

Effects of Weather and Behavior on Body Temperature and the Consequences of Temperature Fluctuations on Development and ...


PAGE 1

1 EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE AND THE CONSEQUENCES OF TEMPERATURE FLUCTUATIONS ON DEVELOPMENT AND REPRODUCTION IN Schistocerca americana : IMPLICATIONS FOR PHENOLOGY AND POPULATION MODELING BY JASON G. FROEBA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Jason Froeba

PAGE 3

3 To my Aunt Linda, who always nurtured my scie ntific interests, and to my wonderful wife Emily, without whom this accomplishment would not have been possible

PAGE 4

4 ACKNOWLEDGMENTS I thank my supervisory committee chair, Dr John L. Capinera, for his advice and support, his firm motivational styl e, and for always keeping my best interest in mind. I thank my other committee members, Dr. Daniel Hahn and Dr Heather McAuslane, for their mentoring and constructive criticism, which greatly improved th is work. I would like to thank Seth Bybee and Jennifer Zaspel for their help and anyone else who provided input, support, or materials. Finally, I would like to thank my wife Emily for her support and encouragement.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ..............9 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................11 2 EFFECTS OF WEATHER AND BEHA VIOR ON BODY TEMPERATURE....................30 Methods and Materials.......................................................................................................... .30 Study Site..................................................................................................................... ....30 Measurement of Body Temperature................................................................................30 Behavioral Observations.................................................................................................33 Environmental Conditions..............................................................................................34 Statistical Analysis...................................................................................................... ....34 Results........................................................................................................................ .............35 Measurement of Body Temperature...............................................................................35 Environmental a nd Behavioral Observations.................................................................36 Temperature Fluctuations and Adverse Weather Conditions........................................38 Discussion..................................................................................................................... ..........40 Measurement of Body Temperature...............................................................................40 Environmental a nd Behavioral Observations.................................................................43 Temperature Fluctuations and Adverse Weather Conditions........................................47 3 EFFECTS OF TEMPERATURE FLUC TUATIONS ON DEVELOPMENT AND REPRODUCTION.................................................................................................................69 Methods and Materials.......................................................................................................... .69 Temperature Treatments................................................................................................69 Cages.................................................................................................................... ..........70 Nymphal Development Time.........................................................................................70 Body Size and Reproduction..........................................................................................70 Statistical Analysis..................................................................................................... ....71 Results........................................................................................................................ .............72 Nymphal Development Time.........................................................................................72 Body Size............................................................................................................. ..........74 Days to Oviposition...................................................................................................... .74 Egg Pods / Female........................................................................................................ ..75 Eggs / Pod............................................................................................................... .......76

PAGE 6

6 Discussion..................................................................................................................... ..........77 Conformity of Laboratory Treatments to Field Study Results............................................77 Nymphal development time................................................................................................78 Body size..................................................................................................................... ........83 Days to Ovioposition.......................................................................................................... 84 Egg Pods / Female............................................................................................................. ..86 Eggs / Pod................................................................................................................. ..........87 4 CONCLUSIONS..................................................................................................................10 2 LIST OF REFERENCES............................................................................................................. 112 BIOGRAPHICAL SKETCH.......................................................................................................116

PAGE 7

7 LIST OF TABLES Table Page 2-1 Summary of mean body temperatures, behavior, and environmental data .......................49 2-2 Regression statistics for grass hopper body temperature and environmental parameters..................................................................................................................... .....52 2-3 Rates of temperature decrease du e to adverse weather conditions....................................53 3-1 Nymphal development time: ANOVA results...................................................................90 3-2 Mean comparisons of nymphal development time............................................................91 3-3 Mean comparisons of nymphal devel opment time for low and high frequency treatments combined..........................................................................................................92 3-4 Body size: ANOVA results................................................................................................93 3-5 Reproductive data: ANOVA results..................................................................................94 3-6 Mean comparisons of reproductive data............................................................................95

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Wiring method used for record ing internal body temperature...........................................54 2-2 May 19 2005................................................................................................................ ......55 2-3 June 14 2005............................................................................................................... .......56 2-4 June 17 2005............................................................................................................... .......57 2-5 June 22 2005............................................................................................................... .......58 2-6 June 29 2005............................................................................................................... .......59 2-7 July 5 2005................................................................................................................ .........60 2-8 July 8 2005................................................................................................................ .........61 2-9 November 17 2005........................................................................................................... ..62 2-10 Variation in the percent temperat ure difference between sun constrained grasshoppers and ambient temperature in relation to sunlight intensity............................63 2-11 Regression analysis of s un constrained grasshoppers........................................................64 2-12 Regression analysis of sh ade constrained grasshoppers....................................................65 2-13 Regression analysis of behavioral data against percent body temperature differences.....66 2-14 Regression of behavior al orientation rating agai nst ambient temperature ........................67 2-15 Regression analysis of free roaming grasshoppers............................................................68 3-1 Fluctuating laboratory temperature treatments..................................................................96 3-2 Nymphal development time of high frequency treatments................................................97 3-3 Nymphal development time of low frequency treatments.................................................98 3-4 Nymphal development time at fixed levels of frequency .................................................99 3-5 Mean femur and overall length for high and low frequency treatments..........................100 3-6 Mean number of days to first oviposition, egg pods / female, and eggs / pod for high and low frequency treatments..........................................................................................101

PAGE 9

9 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE AND THE CONSEQUENCES OF TEMPERATURE FLUCTUATIONS ON DEVELOPMENT AND REPRODUCTION IN Schistocerca americana : IMPLICATIONS FOR PHENOLOGY AND POPULATION MODELING By Jason Froeba May 2007 Chair: John Capinera Major: Entomology and Nematology Phenology modeling is central to many pest ma nagement strategies and often consists of modeling temperatures by using am bient temperatures and insect developmental data derived from constant temperature studies. Not surprisi ngly, discrepancies betw een predictions and the actual timing and size of insect populations can ofte n be attributed to differences between actual body temperature and the temperature used in the model. In addition, temperature fluctuations are a potentially important, yet largely ignored area in insect thermal biology and modeling, as most models are based on data collected at constant temperature. Grasshoppers such as Schistocerca americana are well known for maintaining body temperatures different from ambient temperature. This, plus th e fact that it is one of the fe w grasshoppers in the eastern USA that reach damaging levels, make S. americana a relevant subject on which to assess the effects of thermoregulation on development and reproduction. Body temperatures of grasshoppers, along with several environmental parameters, were monitored continuously to ascertai n the effects of weather and be havior, and to determine what body temperatures grasshoppers could attain and w ould prefer in the field. Body temperatures averaged about 30 and 38C during the daylig ht hours for grasshoppers on overcast and sunny

PAGE 10

10 summer days, respectively, compared to averag e ambient temperatures of 28 and 32 for overcast and sunny conditions. Daytime temperatur es fluctuated frequently, especially on overcast days. There were significant positive relationships between body temperature in S. americana and sunlight intensity, ambient temperatur e and behavior. The rate of increase in body temperature due to direct sola r radiation was as high as 2.5 C /min, which greatly elevated body temperatures, up to 19.6C above ambient. Adverse weather conditions (cloud cover and rainfall) reduced mean daily body temperatures by up to 8C and caused up to a 38% decrease in body temperature at a mean rate of 0.978C/ minute and 0.552C/minute for cloud cover and rainfall, respectively. Laboratory studies were conducte d at constant temperatures and fluctuating temperatures of two amplitudes and frequencies for two mean temperatures of 29.5C and 31.5C for a total of ten treatments. A mean daily temperature in crease of 2C caused a decrease in mean development time (up to about 6 d) and increase in mean reproductive out put (up to 175 eggs / female increase). Amplitude of temperature fluc tuation was also shown to affect development rates and fecundity, but more so in the 29.5C treatments than in the 31.5C treatments. The effects of frequency of temperat ure change were significant for all parameters tested except days to oviposition, but could not be separated from effects which may have been caused by the separation of treatments into two time periods. Ov erall, temperature fluctuations significantly affect development and reproduction in S. americana The results suggest that grasshopper body temperatures are often very different from am bient temperatures, that phenology models based on ambient temperatures can be inaccurate, and that use of more relevant mean body temperatures should be considered for species that thermoregulate or attain body temperatures different than ambient.

PAGE 11

11 CHAPTER 1 INTRODUCTION Temperature plays an important role in the life processes of many organisms. This is especially true in poikilothermic ectotherms, whic h rely on outside sources of heat to maintain internal body temperature. Temperature affects the rate at which biochemical processes take place, and can affect metabolism, developm ental rate, reproduction, and body size. The developmental stage of Atlantic cod, Gadus morhua at hatch is negatively affected by suboptimal temperatures, with development being incomplete at both low and high temperatures (Jordaan et al. 2006). Incuba tion time in northern shrimp, Pandalus borealis, was reduced from 214 days at 2C to 123 days at 8C (Brillon et al. 2005). Insects are particularly sensitive to changes in temperature. Listronotus texanus Stockton (Coleoptera: Cu rculionidae) exhibited a decrease in total development time of 35 days from 65 to 30 days when the temperature was raised from 20C to 27.5C (Woodson and Edelson 1988). Euzopherodes vapidella Mann (Lepidoptera: Pyralidae) showed an increase in mean fecundity from 51.8 eggs per female at 20C to 124.4 eggs per female at 33C. However, wing span and body length was significantly less at 33C than at lower temperatures tested (Ashamo and Odeyemi 2000). As advantageous as temperature increases seem to be, there is a li mit to the extent temperature increases can be beneficial. Extremes in temperature can have seve re negative effects on an organism (e.g. desiccation and deactivation of enzymes) causing thermal death. The temperatures at which severe negative effects occur define the tolera nce range of an organism, which estimates the range of temperatures over which survival is possible (Huey and Stevenson 1979). While temperatures outside of the tole rance range lead to death, the e ffects of less extreme suboptimal temperatures are usually more subtle (e.q., pr olonged developmental time and diapause, and

PAGE 12

12 decreased fecundity and body size). In Mallada desjardinsi (Navas) and Chrysoperla nipponensis (Okamoto) (Neuroptera: Chrysopidae) body size was reduced at both low and high suboptimal temperatures (Nakahira et al. 2005). Castillo et al. ( 2006) showed that developmental time in Quadrastichus haitiensis (Gahan) (Hymenoptera: Eulophid ae), decreased with increasing temperature up to a certain point after which increases in te mperature caused increases in developmental time. This increase in developm ental time at suboptimal temperatures is very common in insects, and the temperatures at whic h the decrease occurs ar e often called critical thermal temperatures. There are both critical thermal maximum and critical thermal minimum temperatures, and they are often used to define ecological and be havioral thermal limits (Hu and Appel 2004). The critical thermal minimum and ma ximum temperatures help define what is known as the thermal performance breadth of an organism, which esti mates the range of temperatures over which an organism performs well (Huey and Stevenson 19 79). When compared to larger organisms, insects exhibit a broad thermal performance breadt h. This may be because insects will encounter the same thermal environments as many other organisms but, due to their small size, body temperature will be more sensitive to small ch anges in the thermal environment (Heinrich 1993). In addition to having a broad thermal performa nce breadth, some insects (e.g., termites) have been shown to have critical temperatures that fluctuate seasonally (Hu and Appel 2004). Of even more importance than the thermal performan ce breadth is the thermal optimum or optimal temperature at which an insect will experien ce the highest fitness level possible (Huey and Stevenson 1979), usually by a combination of in creased developmental and reproductive rates. However, environmental temperatures are often no t those that would be considered optimal for insects, and so body temperature must be co ntrolled in order to optimize fitness.

PAGE 13

13 All insects are poikilothermic and can be ect othermic (relying on ex ternal sources of heat), endothermic (able to use he at generated by muscle activity), or both. In either case, insects must thermoregulate to maintain functional body temperatures when in a suboptimal thermal environment. Thermoregulation is the maintena nce of an optimal internal body temperature different from environmental temperatures. Thermoregulation not onl y includes raising body temperature above ambient, but also systemati cally decreasing body temperature in cases where ambient temperatures become too high. While almo st all insects possess some ability to avoid undesired temperatures (Casey 1981), many insect s can maintain body temperatures within an optimal range by actively alte ring specific behaviors. Insect orders in which behavioral th ermoregulation has been observed include Lepidoptera, Odonata, Orthoptera, Coleoptera, He miptera, Hymenoptera, and Diptera (Heinrich 1993). Methods of thermoregulation in insects ar e variable and differ significantly between orders. Hymenoptera make use of heat produced by the shivering of flight muscles. Honeybees and bumblebees have been known to use this me thod to keep nests warm (Heinrich 1996). Among Lepidoptera, Hemileuca oliviae Cockerell (Lepidoptera: Saturniidae) larvae ascend vegetation and rest during the heat of the day to stay cool. They shift position so that the ventral surface of the body is always facing the sun a nd covered by vegetation (Capinera et al. 1980). Codling moth larvae, Cydia pomonella (L.) (Lepidoptera: Torticidae), exhibit thermoregulatory behavior by selecting favorable microhabitats wi thin apples (Khrt et al. 2005). Khrt showed that 74% of larvae built larger cavities in the warmer hemispheres of apples closest to the heat source. Khrt also found that, unlike the previous larval stages, when larvae sought to cocoon there was no difference in distribution of the ma ture larvae between the temperature gradient and the control, suggesting littl e if any preference for temperature at this stage. He concluded that at

PAGE 14

14 least in the codling moth, behavioral thermoregul ation is life stage dependent, and that larvae change their thermoregulation behavior duri ng development based on benefits, needs, and constraints (Khrt et al.2005). One group in which behavioral thermoregul ation has been extensively studied is Orthoptera, specifically Acridida e, due to their great economic importance. Some Orthoptera, such as katydids, have been observed shivering f light muscles to raise muscle temperature before singing, similar to muscle twitching in Hyme noptera (Heinrich 1993). However, Acrididae do not use their wing muscles to sing, and so far no such warming mechanism has been observed in the group. Consequently, most Acrididae rely so lely on external sour ces of heat obtained behaviorally (Casey 1981), prim arily radiant heat from the s un, also known as insolation. The movement into or out of sunlight is the most common form of behavioral thermoregulation in Acrididae (Lactin & Johnson 1998b). Moving into and remaining in a radiant heat source, usually direct sunlight, as a way to increase body temperature is known as basking. While there are other sources of heat, such as radiant heat from the ground and small amounts of metabolic heat, radiant energy acquired by basking is the most important heat source for any acridid (Uvarov 1966a). In one study conducted on the genus Locusta body temperatures rose from 27.7C to 36C at a rate of 0.83C per minute for ten minutes, thereafter rising at a rate of 0.104C per minute until reaching 42.7 C, when exposed to radiant heat (Uvarov 1966a). While metabolic heat may have played a role in caus ing basal body temperature to be different from ambient temperature, it is very unlikely to raise body temperature by 0.83C per minute, especially in a resting grasshopper. Lac tin and Johnson (1996) f ound that nymphs of Melanoplus sanguinipes (Fabricius) oriented themselves around arti ficial heat sources and reached average body temperatures (near 40C) that were different than those expect ed if the grasshoppers were

PAGE 15

15 randomly dispersed. They considered this conclusive evidence that M. sanguinipes was actively thermoregulating. In a later study, Lactin and Johnson (1998b), showed grasshoppers thermoregulate not only in the laboratory, but also in the field. They estimated the body temperature of free ranging grasshop pers using a series of comple x equations that accounted for different environmental factor s affecting body temperature. Th ey also placed objects with thermodynamic properties similar to those of gra sshoppers at random in the field and recorded the temperature of these objects. The estim ated body temperatures of the free ranging grasshoppers differed significantly fr om those of the objects that were placed at random. This was one of the first field studi es that provided strong eviden ce that grasshoppers actively thermoregulate in the field by choosing warm er locations and altering body posture. When trying to understand the thermal biology of an organism it is not only essential to know if the organism possesses the ability to ther moregulate, but also wh at other factors might affect that ability and body temperature. Fort unately, an extensive amount of work has been conducted for the purpose of understanding the individua l mechanisms of heat transfer that affect a grasshoppers body temperature a nd ability to thermoregulate. However, pathways of heat transfer are very complex and largely independe nt of each other, making them very hard to quantify. Some factors affecting th ese pathways include insect or ientation, body texture, color, and size, radiant intensity, substrate conductiv ity, wind speed, and turbulence (Chappell and Whitman 1990). While basking, acridids will adopt different orie ntations with relation to the suns rays (Uvarov1966b). There are two basic orientations that a grasshopper will adopt, perpendicular or parallel. In perpendicular basking, the long axis of the body is oriented perpendicular to the suns rays (Uvarov 1966b). While in this position, gras shoppers will often assume a flanking position,

PAGE 16

16 in which the hind leg shaded by the abdomen is raised, and the other is lowered to prevent shading of the abdomen (Heinrich 1993). This exposes the largest am ount of surface area possible for heating. Parallel posture is one in which the long axis of the body is parallel to the suns rays. In most acridid spec ies the parallel postur e only exposes about one-fifth the surface area that perpendicular fla nking basking does (Uvarov 1966b). Body size and color are two other factor s that have the pot ential to affect body temperature. Rates of temperature increase and fi nal equilibrium temperature (the point at which temperature no longer increases as a result of a fixed amount of radiant energy) were examined in 1st and 5th instar stages of Schistocerca gregaria (Forskal) (Uvarov 1966a). First instars had a higher rate of temperature increase but lower equilibrium temperatures than 5th instars. In contrast, Willot (1997), while investigating ther moregulation in four species of grasshoppers, found there to be no significant temperature di fferences between males and females within species. Together, the two studies suggest th at while large size differences between 1st and 5th instar nymphs and possibly species affect temper ature, the smaller size differences in adults between males and females has little if any effect on body temperature. Indeed, Lactin and Johnson (1997) found only a 2C difference in body temperature between a body mass of 0.03g and 0.30g, a ten-fold difference in weight. Forsma n (1997) conducted a st udy to determine if differences in color and size w ould affect body temperature in Tetrix subulata (Linnaeus) (Orthoptera: Tetrigidae). He found that body size had no effect on body temperature. However, black individuals achieved higher body temperatures than brown or white individuals. He then conducted a study to see if these differences in body temperature affected reproductive performance in females of the T. subulata Forsman (2000) again found significant variation in body temperature due to color morphs in females, but no difference in overall performance. He

PAGE 17

17 concluded that even though darker color mor phs achieved higher temperatures, there were possible physiological differences between morphs that caused them to required higher optimum temperatures. Similarly, in their study on H. oliviae Capinera et al. (1980) found that artificially blackened larvae attained highe r body temperature than normal larvae, however, performance implications were not evaluated. There are several factors that can potentia lly lower a grasshoppe rs body temperature, including convective and evapora tive cooling, and long wave radiation, which refers to the fact that all objects lose heat by em ission of energy in the form of infrared radiation (Uvarov 1966a). S. gregaria will climb onto vegetation, increasing c onvective cooling when ambient temperature rises to 40C (Heinrich 1993). So me grasshoppers can extend thei r legs to raise the body above the ground, a behavioral mechanis m called stilting, which helps the grasshopper to escape high surface temperatures while increasing convectiv e cooling (Heinrich 1993). There have been several studies concerned with the effects of convection and eva porative cooling on body temperature in grasshoppers. Lactin and Johnson (1998a) showed that orientation to the wind had no effect on body temperature in M. sanguinipes ; however, wind speed did, and as wind speed increased, body temperature decrease d. In a study on the caterpillar H. oliviae, body temperature was shown to be inversely related to air spee d, but body temperature was always above ambient temperature, even at air speeds of 4m/s (Capin era et al. 1980). This suggests that while wind cannot prevent temperature incr ease, it can reduce the maximum attainable temperature and possibly the rate of increase. Evaporative co oling can become a factor in low humidity environments, where evaporation rates are high. Th is can be seen in an experiment in which nymphs of S. gregaria were held at different humidity le vels, and the rate at which thermal equilibrium occurred was measured (Uvarov 1966a). Nymphs held in high humidity

PAGE 18

18 environments reached equilibrium more quickly th an nymphs held in low humidity environments (Uvarov 1966a). The higher humidity reduced he at loss through evaporation allowing the nymph to retain more heat and attain the equilibrium temperature faster. Grasshoppers have also been known to pant, increasing brea thing rate and increasing eva porative cooling, and therefore lowering body temperature (Casey 1981, Heinri ch 1993). Prange ( 1990) conducted a study on the effect of respiratory rate on evaporativ e cooling in three sp ecies of grasshoppers Schistocerca nitens (Thnberg), Locusta migratoria (L.), and Tmethis pulchripennis (Bolivar). He showed that rate of evaporation and ventilati on frequency remained relatively constant up to about 45C, after which both rates increased significan tly. Grasshoppers were able to maintain temperatures below the lethal limit of 48C in air temperature as high as 52-53C. The maintenance of such temperatures caused water loss in the grasshoppe rs to reach 8% of their body mass per hour. Grasshoppers could tolerate a water loss of 33% of their mass, meaning they could maintain sublethal temperatures for about 4 hours. With the exception of deserts, there are few places where ambient temperature can rise to such lethal leve ls requiring such an extr eme form of behavioral thermoregulation. By far the most determining factor of body temperature in grasshoppe rs is radiant energy from the sun and is the main reason body color an d orientation can affect body temperature. It is common knowledge that objects heat up when e xposed to direct sunlight, but more specific parameters such as maximum attainable temper ature and rate of temperature increase differ depending on the nature of the object. There is al so diffuse solar radiation from the sun that reflects off surrounding objects which could possibl y affect temperature. Lactin and Johnson (1997) conducted a study on M. sanguinipes to determine the effects of direct and diffuse solar radiation on body temperature. They found that di rect solar radiation ha d a highly significant

PAGE 19

19 effect on body temperature while diffuse solar radiat ion did not. It is possible that diffuse solar radiation does not have enough energy to heat ob jects, or perhaps body temperature did not differ from ambient because diffuse solar radiation was also warming the surrounding environment. Either way, diffuse solar radiati on likely plays litt le if any part in maintaining body temperatures different from ambient. In simulations based on this work, Lactin and Johnson found that direct solar radiation can cause a temp erature increase of 0.008C / W/m2. In another study, Fielding (2004) placed freshly killed M. sanguinipes grasshoppers perpendicular to the sun at ground level. A thermocouple was inserted into the sternum to obtain temp eratures under direct insolation. Fielding found that body temperatur e could be raised 15-20C above ambient temperature near the ground when exposed to di rect insolation. He also compared predicted developmental times using ambient temperature a nd solar adjusted temperatures and showed the time spent as nymph could be cut in half. While sunlight is the most significant source of temperature elevation in grasshoppers, it is not always available due to adverse weather conditions. Because the sun is the most important factor for raising body temperature, adverse weather conditions or the absence of the sun, is th e most important factor hindering the ability to raise body temperature. In environments such as mountain ranges where ambient temperatures are low and the warmer seasons are short, cloud y days can prevent grasshoppers from obtaining the necessary amount of heat to comple te development (Berner et al. 2004). Aside from behavioral responses and abioti c environmental factors, there are ecological factors which can also play a role in a grasshopp ers ability to thermoregulate. When considering an insects ability to thermoregulate, one must al so consider the biotic environment in which the insect lives. The vegetation surrounding the gra sshopper can dramatically change how each of the abiotic factors mentioned ear lier will affect the grasshopper. Vegetation can provide shelter

PAGE 20

20 from sunlight and wind, while at the same time providing a perch that a grasshopper may use to increase exposure to either element. Vegetation can also produce different thermal environments with varying temperature and humidity levels. W illot (1997) studied the temperature differences between swards (large expanses of grass covered so il) of different height. His results showed that shorter swards reach higher temperatures during the day than taller swards. Differences in food quality of vegetation in the environment will al so play a role in where a grasshopper will spend its time (Pitt 1999). Along with the surrounding ve getation, other animals can also influence a grasshoppers ability to thermoregulate. Predat ors can drive grasshoppers away from optimal conditions and into suboptimal temperatures and areas of reduced food quality. In a study by Pitt (1999) on M. femurrubrum (DeGeer), avian predators were shown to drive grasshoppers down into vegetation away from high quality food and sunlight into c ooler temperatures. It may seem like grasshoppers go through a la rge amount of trouble to maintain optimal temperatures, but there is merit in their e ffort. Temperature can have profound effects on physiological processes in acridids and in insects as a whole. Most chemic al reactions occur at faster rates when subjected to higher temperatures. This generali zation can be applied to most processes in the biological world, and is especi ally important to poikilothermic organisms and specifically in grasshoppers. Some processes that benefit from higher temperatures include speed of muscle contraction, food consumption, metabolism, immune response, development, and reproduction. Grasshoppers usually need to maintain body temperatures above ambient in order to optimize these processes. The speed at which muscle contractions occu r is highly dependent on temperature. This has profound effects on minimum temperatures at which any insect moves, and more importantly, flies. This dependence on a minimum temperature is so important that some insects

PAGE 21

21 can even shiver their flight muscles to warm up to minimum flight temp eratures (Heinrich 1996). In S. gregaria the minimum temperature required for flight is around 20C, while optimum is around 35C (Uvarov 1966b). Feeding rates are also positively correlated with temperature; both chemical digestion and the muscle contractions of the digestive tract ar e temperature dependent. Whitman (1988) conducted a study on thermoregulation in Taeniopoda eques (Burmeister), part of which was measuring feeding rates at different te mperatures. The study showed that as temperature increased, rates of feeding and defecati on increased. Harrison and Fewell (1995), in a similar study on M. bivittatus (Say), showed that there were minimum thresholds for feeding, 10C for lab studies and 25 C for field studies. They also showed there to be a maximum rate of consumption in the laboratory be yond which feeding does not increas e even with an increase in temperature. This suggests an upper feeding th reshold at which rate of consumption would exceed the maximum rate of digestion. In additi on, Harrison and Fewell (1995) also showed that net energy intake increased dramatically from 0.008W at 15C to 0.38W at 35C. This was also accompanied by an increase in metabolic rate. Gndz and Glel (2002) conducted a study on S. gregaria, investigating food consumption and body weight in relation to two different temperatures, 25C and 30C. They found that food consumption increase s with each stage and then begi ns to decline after the end of the first week of adult life, and that during this time consumption increased with an increase in temperature. After the first week of adult life, te mperature increases result ed in decreases in food consumption. Gndz and Glel reported high weight gain from first instar to the first week of adult life in high temperature treatments rela tive to low temperature treatments. However, average weight at the end of the experiment wa s not significantly differe nt between treatments. They concluded that the high we ight gain was due to increased food consumption rates, that

PAGE 22

22 there was some critical weight which was needed for reproduction to begin, and that temperature does not affect the critical weight ne eded but how fast it is attained. In M. sanguinipes, body weight was lower at the low (21C and 24C) and high (39C and 42C) extremes of the temperatures tested relative to intermediate temperatures (Fie lding 2004). It is possible that body weight was affected in this study and not th e other because this study used sub-optimal temperatures. Temperature also plays an important role in a grasshoppers ability to fight infection. Several studies have investigated levels of mycosis at different temperatures in grasshoppers. In M. sanguinipes continuous exposure to high temperatur es was detrimental to the mycosis of Beauveria bassiana Nymphs that were inoculated with B. bassiana and then placed on a heat gradient remained in warmer areas of the gr adient (Inglis et al. 1996). Carruthers (1992) performed a study on the mycosis of Entomophaga grylli in the clear wing ed grasshopper, Camnula pellucida (Scudder). The study showed that exposu re to temperatures of 38C-40C for more than four hours each day was detrimental to survival of E. grylli The ability of high temperatures to fight infection gives gr asshoppers one more reason to maintain body temperatures well above ambient. The effects of temperature on developmen t and reproduction in grasshoppers can be substantial and have been well documented. A study by Begon (1983) showed that when 4th instars of Chorthippus brunneus (Thunberg) were exposed to a radiant heat source, developmental rate could be 5.6 times greater th an when the grasshoppers were not exposed. The same study also showed that adult females held in cages with longer periods of radiant heat laid more egg pods than females in cages with s horter periods of radiant heat. Putnam (1963) conducted a study that shows development time in three species of grassh opper ranging from 53

PAGE 23

23 days at 24C, to 17 days at 38C. In another st udy by Willot (1992) on four species of Acrididae, development and reproduction were either zero or very low at temperatures below 25C, while the optimum temperature for growth and de velopment was between 35C and 40C. In S. gregaria, nymphal development was 10 days faster and sexual maturation 19 days earlier at 30C when compared to 25C (Gndz and Glel 2002). Taeniopoda eques required 60 days from nymph to adult at 25C and only 35 days at 30C (Whitman 1986). Parker (1930) showed an increase from 27C to 32C reduced lengt h of larval development by 27 days in M. mexicanus Saussure. While these studies show that temperature in creases as small as 4C can cut development time in half, it is likely that the more important parameter is how close the temperature approximates the optimal temperature. An Arizona population of Hesperotettix viridis (Thomas) did not develop at 15C and 20C and developed slowly at 23C At higher temperatures of 30C, 35C, and 40C, development was the same, approximately 40 days (Gardner and Thompson 2001). Here, small increases of 5C did not cause an increase in developmental time. Almost all studies show that increases in temp erature can increase development, but the amount of increase and actual be nefits associated with these incr eases depend highly on the thermal biology of the organism in question. Temperature increases development by increa sing biological processes in general, but this can also have a negative side effect. Increa sed metabolic rates tend to shorten an organisms life span. In one study conducted on C. pellucida adults survived for 16 days at 37C and 32.6 days at 27C (Uvarov 1966a). At first glance one might conclude that those grasshoppers living at 27C have a higher fitness. However, at 27C females laid an average of only one egg pod, as opposed to the average of four egg pods laid at 37C (Uvarov 1966a). Even though the longevity

PAGE 24

24 of the grasshopper had been reduced, its fecundity had quadrupled. These results were similar to those of Begon (1983). Overall, basking can help a grasshopper maintain an optimal body temperature and increase development rate and egg production. Incr eases in development rate are beneficial to most grasshoppers, as adult grasshoppers have fe wer natural enemies than nymphs, and the faster one reaches adulthood, the sooner one can leav e behind certain risk factors (Kemp 1986). Increases in developmental rate due to temperatur e are also essential to the survival of some species of grasshoppers that live in cooler climat es which could not complete their life cycle in a single season if it were not for the ability to ba sk and increase their developmental rate (Heinrich 1996). In a study on T. eques Whitman (1988) found that this species required 850 degree-days to complete its life cycle, while the air temperature of the environment only provided 692 degree-days. The deficit in th e supply of heat was made up by the grasshoppers ability to thermoregulate. Most of the studies mentioned above were c onducted at constant temperatures. While this simplifies the execution and analys is of experiments, it fails to mimic natural conditions. In nature, terrestrial organisms are submitted to dail y fluctuations in temperature, often exceeding a 10C difference (Petavy et al. 2001). Daily temp erature fluctuation, wh ich is most commonly thought of as a day / night cycle, is referred to as thermoperiod. Beck (1983) conducted a review on thermoperiod in insects and found that ther moperiod is known to affect development, fecundity, circadian rhythms, diapause, and biol ogical clocks in insects. The effects of thermoperiod can differ between different insect s; under cyclic temperatures some species will develop more rapidly, others will show no difference, and even a few will develop more slowly (Beck 1983). For example, the European corn borer, Ostrinia nubilalis Hbner (Lepidoptera,

PAGE 25

25 Pyralidae) exhibited no difference in development time under thermoperiodic conditions when compared to constant temperatur es, but did exhibit larger late instar larvae under thermoperiodic conditions (Beck 1983). On the other hand the pitch plant mosquito, Wyeomyia smithii (Coquillett) developed more slowly under thermoperiod ic conditions. However, larvae produced larger, heavier, and more fecund mosquitoes wh en compared to larvae reared under constant conditions (Beck 1983). While the benefits of th ermoperiodic conditions in these two species may not be readily apparent in developmental da ta, their increased fitness becomes apparent in reproductive aspects of their life history. On th e other hand, the benefits of thermoperiodic conditions may be blatantly obvious. Satar et al. (2005) tested th e effects of thermoperiod on the aphid Brevicoryne brassicae (L.). At alternating temperatures of 25C and 30C, developmental time decreased, mortality decreased, longevity increased, and reproduction increased when compared to those reared at a constant 30C even though they had rece ived less overall heat. Many studies regarding thermoperiod in inse cts have yielded ambiguous results and those that have produced clear results often conflict with results from othe r studies (Beck 1983). Studies dealing with thermope riod often compare fluctuating treatments against a constant temperature at the average or midpoint temperatur e of the alternating treatment. The premise for this is that if development is the same unde r both alternating temperature and the midpoint temperature, then the relationship between deve lopment time and temperature can be assumed to be linear. However, the relationship is usually not linear and there is often a deceleration or acceleration of developmen t rate which is called the Kaufmann effect (Petavy et al. 2001). In such situations, midpoint temperatures are not suitable for predicting development times under alternating conditions. An alternat ive must be used and is called the equivalent development temperature (EDT), which is the constant temper ature that provides the same developmental time

PAGE 26

26 as that observed under a given alternating temper ature (Petavy et al. 2001 ). Petavy et al. (2001) conducted a study on relatively simple thermoperiods in Drosophila They consisted of two 12 hr phases (day and night) and total of 14 treatments with mid range temperatures from 10C to 27C and amplitudes of 6C to 22C. Overall, there was a decrease in development time with increasing temperature. Mortalit y was 100% at alternating temper atures of 4C/26C, 9C/33C, and 21C/34C, and temperatures above 28C cau sed increases in developmental time. As expected, mortality was caused by extremes in temperature, while suboptimal temperatures caused less severe deleterious effects. They determined that the relationship between development time and temperature is a positive f unction of amplitude and a negative function of midpoint. The study showed that development time under alternating temperatures can be up to 20% superior or inferior to the expected de velopment time at the midpoint. Depending on the species, a given amount of heat from alternating temperature c ould provide some developmental advantage over the same amount of heat provided at cons tant temperature. This poses a problem for the common practice of using degree-day accumulation based on average ambient temperature, or the midpoi nt, and known developmental rates obtained at constant temperatures to deve lop phenology models for pest management purposes. This method assumes there to be a linear rela tionship between life history trai ts and accumulated degree days. For many insects the relationship is not linear and helps explain why Khrt et al. (2005) report that in the codling moth and other insects there are time di screpancies between predicted population numbers and actual numbers observe d in the field. Grasshopper phenology is a foundation of grasshopper management (Gardne r and Thompson 2001) and understanding the effects of temperature on phenology is essentia l to developing reliable pest management techniques.

PAGE 27

27 It is known that grasshoppers develop faster and in general are h ealthier when reared under day / night temperature cycles (Uva rov 1966b). More recently, in his study on M. sanguinipes Fielding (2004) found that grasshoppers reared at alternating temperatures developed faster than those reared at the midpoint temperature. As with any terrestrial organism, grasshoppers are affected by daily temperature cycles. However, grasshoppers behaviorally thermoregulate by exposing themselves to sunlight, and changes in this behavior or availability of sunlight will affect the cycle of temperatures th ey experience, specifically daily fluctuations in temperature other than the day / night cycle mo st commonly investigate d. There has been little field work conducted to see what types of temp erature fluctuations gr asshoppers experience under natural conditions, and inves tigated even less are the effects these fluctuations may have on life history traits. If developmen t time is affected by the way in which heat is obtained and not just the absolute amount of heat obtained, then other factors such as fecundity and body size might also be affected. The grasshopper S. americana (Drury) is commonly the most economically important grasshopper in Florida and has b een known to cause severe damage to citrus and ornamental crops (Capinera 1993). In Florida it is known to have a spring or early summer generation followed by an autumn generation (Kuitert and C onnin 1952). This coincides with the report that the numbers of nymphs increase in early summer and late September (Squitier and Capinera 2002a). The eggs of overwintering adults lack a prolonged diapause and ha tch in the year they were deposited. Consequently, in Florida, S. americana overwinters as an adult (Squitier and Capinera 2002a). S. americana has adapted to nearly every Florida habitat (Squitier and Capinera 2002b). Like many Schistocerca species, it is a tree and shrub dweller and can usually be found in abundance in areas where both food a nd perches are available. Disturbed habitats

PAGE 28

28 near the edges of crop fields and ro ads often contain large numbers of S. americana Squitier and Capinera (2002b) report that abund ance in pine plantations is a bout 20 times greater when pine trees are young and small than in mature pine stan ds. This may be due to several factors. One, a mature pine stand usually has a less diverse plant community as compared to younger stands (Squitier and Capinera 2002b). Second, fully grow n pine trees may provide more shade than S. americana prefers. When Miles ( 1985) placed adults of S. americana in thermally heterogeneous environments, he found that on average they spen t about 90% of their ti me in locations where thoracic temperature was 32C 44C. Like ma ny of the grasshoppers mentioned earlier, S. americana prefers to maintain high internal body te mperatures and does so by basking. Hence, shaded habitats, such as mature pine stands, would likely not be preferred. In 1991 there was an outbreak of S. americana in Florida. Considerable damage occurred in Pasco, Polk, Sumter, and Hernando countie s (Capinera 1993b). According to Capinera (1993a), the outbreak was likely due to a 5year drought which provided an abundance of sunshine and increased the ambient temperature. This was coupled with a mild winter in 1990-91 which allowed for high overwintering adult surviv al rates and abundance of suitable habitat. All of these combined factors may have induced a gradual population increase over several years to produce an outbreak population. Most grasshopper research is conducted on species that occur in the western United States where outbreak populati ons frequently occur. The concentration of work on other species and lack of frequent outbreaks of S. americana has caused this species to have not been well studied (Capinera 1993a). Ho wever, its close relation to two species well known for outbreaks, S. gregaria in Africa and S. piceifrons in Mexico, along with the 1991 outbreak underscores their pote ntial to reach outbreak populati ons and warrants further study.

PAGE 29

29 In preliminary laboratory testing, S. americana adults oriented themselves around a radiant heat source and reached temperatures of over 40C for short periods of time when allowed to position themselves in front of 100w incandescent light bulbs. Achieving temperatures this high in the field is very unlik ely without the presence of direct sunlight. During hours when a heat source (in this case an incandescent light bulb) was provided, S. americana adults often did not maintain a constant temp erature, but allowed their body temperature to fluctuate between 30C and 41C. This corres ponds with the idea sugge sted by Kemp (1989), that grasshoppers will only use a portion of th e heat available to them on a given day. These preliminary tests, along with observations that S. americana adults do not bask all hours of the day, suggest that sunlight is only required for a po rtion of the day. This is relevant in Florida, where during the summer there are very few days with no cloud cover, allowing for a full day of sunlight. Also, there are times when cloud cover can remain for several days, which could be detrimental to grasshopper development and fecundity. Our first objective was to assess the eff ects of sunlight on the body temperature of S. americana in the field, to determine how adverse weather conditions such as rain and cloud cover may affect body temperature, and to ascert ain what sorts of temperature fluctuations a grasshopper might experience throug hout the course of a day. Th e second objective of this study was to determine the effects ecologically rele vant cycles of frequency and amplitude of temperature change on development and fecundi ty by developing standardized alternating temperature regimes based on what grasshoppers e xperience in the field. Such information will allow for development of more accurate phenolo gical models that are not based on ambient temperature, but more accurate body temperatures, not only for this species but for others as well.

PAGE 30

30 CHAPTER 2 EFFECTS OF WEATHER AND BEHA VIOR ON BODY TEMPERATURE Methods and Materials The purpose of this portion of the study was to assess the effects of sunlight, adverse weather, and behavior on the body temperature of S. americana in the field, and more specifically, tries to determine what kinds of temperature fluctuations a grasshopper might experience due to these factors. To accomplish th is, grasshoppers were wired to thermocouples and placed in the field where observations of body temperature, behavior and environmental conditions were recorded. Study Site Members of the genus Schistocerca are shrub and tree dwellers. Many of the shrubs and trees used as perches are also host plants, such as citrus. In an effort to promote natural behavior, citrus trees were used as the site of the fiel d study. Three potted grapef ruit trees (1.5 2m high) were placed on a light gray gravel surface wh ich was assumed to have similar thermal and reflective properties as sandy soil, the normal substrate for citrus. The trees were arranged as if they were a single tree with a dense canopy and several protruding branches. Trees were watered as needed and occasionally sprayed with a copper solution to control scal e insects (Hemiptera : Margarodidae). Trees were also fitted with ant barriers (Line Guard Inc ., Elyria, OH) to deter predation by ants. Observations were made i rregularly from mid-May to mid-November of 2005. Measurement of Body Temperature Grasshoppers used in the field study we re pre-reproductive adult females of S. americana taken from a lab colony at the University of Flor ida. Only pre-reproductiv e females were used to prevent any differences in behavi or that may be due to necessity of heat intake, which could differ between stage and sex. In order for continuous body temperatures to be taken,

PAGE 31

31 grasshoppers were fitted with a permanen t thermocouple (36 gauge, Teflon insulated, thermocouple wire, part# TT-T-36-SLE, Omega Engi neering, Inc., Stamford, CT) (Fig. 2-1). The thermocouple consists of a separate copper and c onstantan wire, with a Teflon coating. One end of the pair was connected to a male terminal, copper to positive and constantan to negative (Sub Mini T/C Connector, Part # SMP-T-M, Omega E ngineering, Inc., Stamford, CT). The copper side of the other end was threaded through a se wing needle (size 28), which was then inserted laterally into the centr al mesothorax, and pulled through so that the end of the wire runs completely through the thorax. The copper wire was then soldered to the constantan wire forming the site of temperature measurement, which was then pulled back into the center of the thorax. Hot glue (SuperPower Slow Setting Hot Melt Glue, Model No. BSS6-4, Arrow Fastener Co. Inc., Saddle Brook, NJ) was used to seal off th e wounds and to hold the wire in place. There seemed to be no behavioral effects due to the process, and grasshoppers lived for more than a week in this state. Similar results have also been noted in other studies where thermocouple wires were inserted into the th orax of grasshoppers (Carruther s et al. 1992, Lactin and Johnson 1996). Later dissection of the wired grasshoppers showed there to be no damage to internal organs. This method allows for continuous temperat ure readings to be ta ken and is much more accurate than the common grab and stab me thod used in many other studies (Beck 1983, Begon 1983, Kemp 1986, Willot 1997, Bl anford and Thomas 2000,). Four grasshoppers were wired with thermoc ouples each morning the tests were to take place. One grasshopper was hot-glued to a branch near the center of the canopy to provide continuous shade (shade constrained grasshoppe rs), allowing the measurement of minimum temperatures that could be enc ountered by a grasshopper on that pa rticular day. This could then be used as a relative ambient temperature measur ement of the grasshopper and be compared with

PAGE 32

32 readings for actual ambient air temperature. A s econd grasshopper was hot-glued to a green pipe cleaner which was attached to a branch fully expo sed to sunlight and positioned so that the long axis of the body was perpendicular to the suns rays (sun constrained grasshoppers), theoretically allowing the measurement of maximum temperat ures that may be encountered. After a few minutes, both stationary grasshoppers quit str uggling to get free, therefore minimizing any effects of metabolic heat. The fully exposed grasshopper only began struggling again when it began reaching lethal temperatures. The remaini ng two grasshoppers were marked with either red or blue lettering enamel (Sign Painters 1 Shot Paint Peinture Lettering Enamel, 153-L Process Blue and 165-L Rubine Red, Consumers Pa int Factory, Inc., Gary, IN) on the hind tip of the forewings. This allowed them to be easily distinguished when behavi oral observations were being recorded. These two grasshoppers were pl aced on the grapefruit trees to roam freely, allowing them to behaviorally thermoregulate an d to maintain desired body temperatures. Free roaming grasshoppers were provided with e nough thermocouple wire to allow unlimited movement within the trees. This provided measuremen ts of preferred temperat ures in contrast to the minimum and maximum measurements provi ded by the two stationary grasshoppers. Once all four grasshoppers had been placed on the trees, they were connected to data loggers (Easy View Dual Input Thermometer, model# EA15, Extech Instruments, Waltham, MA). Each data logger had two terminal inputs; hence, one logger was used for the shaded and exposed grasshoppers and one for the two free roaming grasshoppers. The data loggers were housed in plastic containers with a watertight lid. Temperature readings were logged every ten seconds throughout the duration of the tests to provide a more accurate view of how quickly internal body temperature re sponded to weather changes.

PAGE 33

33 Behavioral Observations After the data loggers had been started and it had been determined th at all were working properly, grasshoppers were allowed to settle fo r a short period before behavioral monitoring began. Visual monitoring took place every 15 mi n. during the time when grasshoppers were most actively thermoregulating, usually from the start of the experiment (9:00 a.m.) until about 4:00 p.m. Behavioral observations for the free-roaming grasshoppers were recorded as five different categories: completely shaded, partial sun exposure (about 25-75% of body surface) with a parallel orientation, partial sun exposure with a perpendicular orientation, complete sun exposure with a parallel orientation, and complete sun exposure with a perpendicular orientation. These behavioral orientation responses were th en numerically rated as 0, 0.375, 0.50, 0.75, and 1.00, respectively. Previous research (Uvarov 1966b) has shown that orient ation of the long axis of the body plays an important role in temperature re gulation and that a perp endicular orientation provides for higher attainable te mperatures. Therefore, a rating of 1.00 represents a grasshopper receiving the most solar radia tion possible. The 0.75 rating for the parallel orientation was derived from measurements taken that showed a 25% reduction in body surface area exposed to sunlight when moving from perpendicular to paralle l orientation. Each rati ng was then arbitrarily halved for partial exposures. After behavioral observations were made, any tangles that may have developed in the wires since the last obs ervation were removed. Also, during this time the two stationary grasshoppers were checked to make sure they re mained in position, and that the exposed grasshoppers orientation remained perpe ndicular. Adjustments needed to maintain a perpendicular orientation were accomplished by repos itioning the pipe cleaner. A total of 6 d (2 grasshoppers on each day) yielded behavioral data suitable for st atistical analysis.

PAGE 34

34 Environmental Conditions Measurements taken throughout the test pe riod included sunlight intensity, ambient temperature, and humidity. Sunli ght intensity was take n every second and the average recorded every 10 sec as W/m2 with a Silicon Pyranometer smart sensor (Part # S-LIB-M003, Onset Computer Corp., Bourne, MA) attached to a Hobo Micro Station (Par t # DOC-H21-002, Onset Computer Corp., Bourne, MA). The light sensor was placed as close to the test trees as possible without becoming shaded. Temperature and humidity readings were recorded every 10 sec with a Hobo data logger (Hobo U12 Temp/RH/Light/Ext ernal Data Logger, Part # U12-012, Onset Computer Corp., Bourne, MA) placed on top of a pl astic plate supported by cork legs, which was then placed in the shade of a bu ilding away from any vegetation. Statistical Analysis Descriptive statistics for field data were calculated using Microsoft Excel (Microsoft Corporation, Redmond, WA). A simple linear re gression was performed on sunlight intensity and percent temperature difference between sun constrained grasshoppers and ambient temperature. Simple linear regressions were also performed on mean behavior and ambient temperature, and temperature differences betw een constrained grasshoppers and free roaming grasshoppers and behavior. Body temperatures, ambient temperature, % RH, and sunlight intensity for each day were plotted. Rates of temp erature increase and decrease were calculated using the linear portions of body te mperature plots. SAS Analyst 9.0 (SAS Institute Inc., Cary, NC) was used to perform multiple linear regression analysis and plot partial regressions on the combined data of body temperatures and behavior ambient temperature, and sunlight intensity from June 29, July 5, and July 8, 2005.

PAGE 35

35 Results Measurement of Body Temperature Body temperature, ambient temperature, % RH and sunlight intens ity were plotted for each of the 7 d tested (Fig. 2-2 2-9). For each of the 7 d tested, mean values of body temperature were calculated (Table 2-1). Because of the very large sample sizes (>2500), even extremely small differences between body temper atures (such as 0.05C) were considered significant by ANOVA, causing every temperature to be considered different. Therefore, this information was omitted from the table. As expected, sun constrained grasshoppe rs attained the highest maximum body temperature of all the grasshoppe rs tested, with the exception of one day tested in November (Table 2-1). What was not expected was how high maximum body temp erature reached. On three of the 7 d tested, internal body temperat ures rose above 50C, causing thermal death. Increased respiration rates were observed in sun constrained grass hoppers at such high temperatures, possibly as a cooling mechanism (Heinrich 1993, Casey 1981). Unfortunately, this behavior could not be maintained indefinitely and ev entually some of the grasshoppers reached lethal temperatures. Body temperatures at their maximum were 12.4-19.6C above ambient temperature, often around 75% a bove ambient temperature. Sun constrained grasshoppers were meant to represent the highest attainable temperatur e for that day. For the mo st part this was true, but there were factors acting on sun constrained grasshoppers, such as wind and rain, that the free roaming grasshoppers could avoid. Cons equently, mean body temperatures of sun constrained grasshoppers were not always the highest throughout the day. Shade constrained grasshoppers tended to ha ve the lowest average body temperatures for each day, with the exception of the two hottest and sunniest days, July 5 and July 8, where the

PAGE 36

36 body temperature of the shade constrained grassh opper was very similar to those of the free roaming grasshoppers (Table 2-1). On these days free roaming grasshoppers spent most of their time shaded from direct sunlight and consequen tly had body temperatures similar to those of the shade constrained grasshopper. On five of the se ven days tested, shade constrained grasshoppers had a mean body temperature greater than ambien t temperature, with one day having a maximum body temperature that was 26.9% above ambient temperature. The highest body temperature reported in a shade constrained grasshopper wa s on the hottest day tested, July 8, where body temperature reached 41.3C. Mean body temperatures for free roaming grasshoppers were somewhat harder to generalize, but for the most part they were between those for shade and sun constrained grasshoppers (Table 2-1). With the exception of November 17, maximum body temperature of free roaming grasshoppers never surpassed the maximum for sun constrained grasshoppers. On all days but November 17, free roaming grasshop pers raised their body temperature above 40C for short periods of time, which corresponds well with what was found in preliminary laboratory tests. On several occasions, night-time body temper atures were also recorded for all four grasshoppers. For the most part, all grasshoppers maintained body temperatures very similar to that of ambient, usually within 1C above or below ambient. Environmental and Behavioral Observations For each of the days where data were avai lable, mean values of sunlight intensity, ambient temperature, relative humidity, behavior as well as differences between constrained grasshopper body temperatures and ambient te mperature were calculated (Table 2-1). Measurements of sunlight intensity were availabl e for four of the days tested. The higher the

PAGE 37

37 mean sunlight intensity the higher the mean per cent difference between body temperatures of sun constrained grasshoppers and ambient temperature. Simple linear regression analysis between percent difference in temperature an d sunlight intensity produced an R2 value of 0.99 (Fig. 2-10). A multiple linear regression analysis with sun constrained body temperature and ambient temperature and sunlight intensity revealed si gnificant positive linear relationships between body temperature and both ambient temperature and su nlight intensity (Tab le 2-2, Fig. 2-11). Not surprisingly, sunlight was the most signifi cant factor, accounting for 0.8664 of the total r 2. The data tend to be clustered in th e partial regression plots because most body temperatures of sun constrained grasshoppers were on the higher end a nd at high levels of sunlight. A multiple linear regression analysis with the body temperature of shade constrained grasshoppers and ambient temperature and sunlight intensity also revealed significant positive linear relationships between body temperature and both ambient temperature and sunlight intensity (Table 2-2, Fig. 2-12). In contrast to the regression analysis for the s un constrained grasshopper, ambient temperature was the most significant factor, acc ounting for 0.9539 of the total r 2. The models given for predicting body temperature of sun and shade constrained gr asshoppers (Figs. 2-11c, 2-12c) were found to be reliable. Unlike the constrained grasshoppers, free roaming grasshoppers were allowed to thermoregulate behaviorally. Both parallel and perpendicular or ientations to the sun were observed, but no grasshopper was ever observed in the flanking position. As a general rule, grasshoppers tended to begin bask ing immediately in the morning sun after being placed on the test trees. In all but one inst ance, the grasshopper with the high er mean behavioral orientation rating (the one that spent more time basking) of the two free roami ng grasshoppers had the higher mean daily body temperature (Table 2-1). In creases in mean behavi oral orientation rating

PAGE 38

38 for different days reflected increases in the mean difference between the body temperature of free roaming grasshoppers and those of shade cons trained grasshoppers (Fi g. 2-13a). Conversely, increases in mean orientation behavior cause d decreases in the mean difference between body temperature of free roaming grasshoppers and those of sun constrained grasshoppers (Fig. 213b). Mean daily behavioral orie ntation rating decreases with increases in mean daily ambient temperature (Fig. 2-14). A multiple linear regression analysis betw een the body temperatures of free roaming grasshoppers and ambient temperature, sunlight intensity, and behavior al orientation rating showed significant positive linear relationshi ps between body temperature and all three parameters (Table 2-2, Fig. 2-15). All three parame ters were found to be significant contributors to the variation in body temperature, with ambien t temperature being the mo st significant factor, accounting for 0.7819 of the total r 2 while behavior only account ed for 0.0063 of the total r 2. The relationship between body temperature and ambi ent temperature and sunlight intensity for free roaming grasshoppers follows somewhat si milar patterns to those of the constrained grasshoppers (Figs. 2-15a, 2-15b). Unlike the data for constrained grasshoppers, the data for free roaming grasshoppers are much more dispersed along the y-axis, indicating some effect of behavior. The partial regression plot for body te mperature and behavior (Fig. 2-15c) takes on a columnar appearance. However, a pattern still emer ges; as behavioral orientation rating increases there is a slight increase in body temperature. The model for free roaming grasshoppers (Fig. 2-15d) is not as accurate as t hose for constrained grasshoppers. Temperature Fluctuations and Adverse Weather Conditions Rates of body temperature increas e in sun constrained grassho ppers were calculated from linear portions of body temper ature plots where sunlight intensity was 800 W/m2 or greater

PAGE 39

39 (sunny conditions). Sunlight intensity fluctuated too rapidly to allow for rates of temperature increase to be measured at a constant inte nsity. The mean ( SD) rate of body temperature increase for sun constrained grasshoppers was 1.24C/min 0.009, with a maximum of 2.5C/min, and a minimum of 0.71C/min. Large fluctuations in body temperature (> 15 % change in temperature) occurred several times a day, averaging 3.67 1.50 (Mean SD) ch anges per day, and were often the direct consequence of changes in weather or sunlight intensity (Figs. 2-2 -2-9), with a few due to behavioral thermoregulation. We used a 15% change in temperature to define a large fluctuation. This eliminated from the calculations, any sma ll changes that might have been caused by very brief changes in sunlight intensity, where the fu ll effect of the change could not have been observed. Additionally, body temperatures of sun and shade constrained grasshoppers show a high frequency of lower amplitude temperature changes not caused by cloud cover or rain which were also eliminated by the use of the > 15% rule. The rates of change and amplitude (% temperature change) were calculated for large temp erature fluctuations caused by cloud cover or rainfall (Table 2-3). Cloud cove r caused a higher mean rate of temperature decrease than did rainfall, but had a lower mean % decrease. The frequency and amplitude of temperature fluctuations for the laborato ry portion of this study were based on these observations. The lowest temperature recorded as a resu lt of adverse weather conditions, 20.1C, was due to rain on June 22 and was 7.6% below ambi ent temperature. On June 22 and June 29, when frequent cloud cover and rainfall were record ed, mean body temperatures were up to 8C lower for all grasshoppers than on predominately sunny days (Table 2-1).

PAGE 40

40 Discussion Measurement of Body Temperature Sun constrained grasshoppers achieved body temperatures that were 12.4-19.6C above ambient. Fielding (2004) reported similar resu lts of 15-20C above ambient temperature. However, Fieldings grasshoppers were at gr ound level and dead, reduc ing both convective and evaporative cooling. The grasshoppers in the current study may have reached even higher temperatures had they been treated in a si milar manner. Uvarov ( 1966a) reports that in L. migratoria maximum temperatures reach 42.7C, but doe s not specify the source of radiant heat. S. americana exhibited maximum temperatures several de grees higher (Table 2-1), even in free roaming grasshoppers. Though sun constrained gra sshoppers did not always attain the highest body temperatures, they provided a reasonable re presentation of maximum body temperatures that may be attainable by a grasshopper on a summer day in Florida. The body temperatures of sun constrained grasshoppers reported in this study were often around 75% above ambient temperature. Theoretically, grass hoppers with the ability to thermo regulate could also reach such temperatures and, in fact, they do. The fact that grasshoppers possess the ability to raise their body temperature so high above ambient, puts in to question the method of using mean ambient temperature to model phenology and population si zes. Surprisingly, body temperatures of sun constrained grasshoppers attained lethal levels ( > 50C) when exposed to direct sunlight, emphasizing the need for cooling. To the auth ors knowledge there is no other study that documents thermal death in Schistocerca species from exposure to direct sunlight. Shade constrained grasshoppers, for the mo st part, had the lowest mean daily body temperature. However, there were 5 d when mean daily body temperatures were above ambient, sometimes as much as 26%. The likely explanat ion for shade constraine d grasshoppers having a

PAGE 41

41 higher body temperature than ambient is that the gr asshoppers were located within the grapefruit trees and surrounded by vegetation. Sheltering vegetation can provide some amount of insulation by reducing air flow and preventing heat escape, causing air temperature within the trees to be different from that measured out in the open ai r. The surface of the vegetation can also reflect light back into the canopy and possibly incr ease temperature even more. The thermal and reflective properties of the gravel also likely ad ded to the increased temperature. In addition, when night time temperatures we re recorded they were shown to be very similar to ambient, suggesting that temperature differences between ambient and shade constrained grasshoppers were caused by radiant energy from the sun be ing trapped within the tree canopy. These data show that even when completely shaded, grassh oppers will experience body temperatures much higher than that of ambient temperature (up to 26 %) due to differences in microclimate, and that measurement of microclimatic conditions woul d likely be more appropriate for modeling purposes. As expected, the mean daily body temperat ures of free roaming grasshoppers were usually between the body temperatures of the constrained grasshoppers, which represent the extremes in temperature that could be experien ced. Kemp (1989) suggested that grasshoppers do not utilize all of the available heat in a day and th at sunlight is not requir ed all hours of the day for optimal body temperature to be maintained. Th e data from this study correspond with this on days when ambient temperature is high and sunlight is available for most of the day. Data from July 5 and 8 shows that mean body temperatures of free roaming grasshoppers were well below those of sun constrained grasshoppers, the theo retical maximum (Table 2-1). On days when ambient temperature was cooler or sunlight was limited, grasshoppers seemed to be attempting to utilize most of the heat available to them. Data from June 14, 17, 22, and 29 shows increased

PAGE 42

42 basking behavior and body temperatures of free roaming grasshoppers very close to, and in two instances higher than, th ose of sun constrained grasshoppers (Table 2-1). However, free roaming grasshoppers did not bask every moment of the day even on cooler days. All of this suggests that there is a certain amount of heat a grasshopper ne eds obtain to achieve optimal fitness and that they will not utilize more heat than is n ecessary to do so. High body temperatures do have negative side effects and would be avoided wh en there is no longer a benefit provided by increasing body temperature. July 5 and 8 would represent days when mo re than enough heat was available, while June 14, 17, 22, and 29 were days where heat was more limited and grasshoppers took advantage of most of the heat that was available. November 17, 2005 was the only day tested during the winter months. November 17 provided some interesting data. As mentioned ear lier, sun constrained gr asshoppers attained the highest maximum body temperature of all the grass hoppers tested, with the exception of this day, when free roaming grasshoppers attained th e highest maximum and mean temperatures. November 17 was the coolest day tested, and was characterized by a stiff breeze and low sunlight intensity. Because there were no rainfall events, the most likely explanation for lower sun constrained body temperatures is that the br isk breeze prevented the exposed sun constrained grasshopper from reaching higher temperatures, while free roaming grasshoppers might have been able to position themselves to bask while avoiding exposure to the wind. What is also interesting about November 17 is that the m ean body temperature of the shade constrained grasshopper was below mean ambient temperatur e, and is probably a result of the high wind speed and low sunlight intensity. This helps supp ort the idea that, on dates other than November 17, radiant energy from sunlight trapped by the vegetation was causing shade constrained grasshopper temperatures to rise above ambient.

PAGE 43

43 Environmental and Behavioral Observations A simple linear regression revealed a very significant positive relationship (R2=0.99) between mean sunlight intensity and the mean difference between body temperatures of sun constrained individuals and ambi ent temperature. As expected, the higher the mean sunlight intensity, the higher the body temperature rose above ambient temperature. When a multiple regression analysis was conducted on the raw da ta, accounting for the effects of both sunlight and ambient temperature on body temperature, s unlight was the most significant factor, accounting for 0.8664 of the total r 2 (Table 2-2). Such a dependency of body temperature on sunlight was expected in sun c onstrained grasshoppers, and show s how influential sunlight can be on body temperature. These results are in ag reement with many other studies. Lactin and Johnson (1997) found direct sunlight to signifi cantly affect grasshopper body temperature. Fielding (2004) also reported a very strong relationship betw een grasshopper body temperature and sunlight. The majority of data points shown in the plot of predicted body temperatures vs. actual body temperatures given by the multiple regression analysis (Fig. 2-11c) follow the general linear pattern predicted by the model. A model predicting the body temperature of constrained grasshoppers is not likely to prove useful in trying to model the phenology and population of free moving grass hoppers; however, the model does provide some interesting information. It provides a method for calcu lating a rough estimate of maximum grasshopper body temperature when the ambient temperature and sunlight intensity are known. Additionally, there are some outlying data points which seem to reduce the models predictive power. These points are artifacts of the delaye d response of body temperature to changes in sunlight intensity. Changes in sunlight intensity were recorded im mediately, while the effects these changes had on body temperature took some time to manifest them selves in the body temperature recordings.

PAGE 44

44 In contrast to sun constrained grasshoppe rs, the body temperature of shade constrained grasshoppers was much less dependent on sunlig ht intensity and more dependent on ambient temperature. The multiple regression analysis showed ambient temperature having a linear relationship with body temperature a nd accounting for 0.9539 of the total r 2 (Table 2-2). This result was expected, because shade constraine d grasshoppers were deni ed access to direct sunlight. Therefore, the body temp erature of shade constrained grasshoppers should be very dependent on ambient temperature. This is partially supported by the fact that the data for shade constrained grasshoppers in the partial regression plot betwee n body temperature and sunlight intensity (Fig. 2-12b) are much mo re dispersed along the y-axis than the data for the equivalent sun constrained plot (Fig. 2-11b). In addition, the plot of predic ted vs. actual body temperatures (Fig. 2-12c) does not exhibit nearly as many ou tlying values as that of the plot for sun constrained grasshoppers (Fig. 2-11c). This furt her shows the reduced effect of sunlight on body temperature in shade constrained grasshoppers. It is suspected that the effect of sunlight on body temperature and the positive rela tionship between the two (Fig. 2-12b) are due to the partial dependency of ambient temperature on sunlight in tensity (even though tests for co-linearity were negative) and differences in microclimatic condi tions which would also be partially dependent on sunlight intensity. Blanford and Thomas (2000), among many others, report a non-linear relationship between ambient temperature and body temper ature in grasshoppers. They report that body temperatures reach equilibrium at higher ambient temperatures. There is a simple explanation for the discrepancy between their study and this st udy. They were trying to show behavioral thermoregulation and at very hi gh temperatures grasshoppers were actively cooling their bodies, while at cooler temperatures grasshoppers were basking to raise their body temperature. In the

PAGE 45

45 current portion of this study, behavior has been removed from the equation for constrained grasshoppers and body temperature is solely de pendent on environmental conditions. For this reason there is a linear relationship between body temperature of shaded grasshoppers and ambient temperature. Free roaming grasshoppers adopted both paralle l and perpendicular or ientations. The fact that grasshoppers were never observed in a flanking position (when a grasshopper raises one hind leg and lowers the other while in a perpendi cular orientation to maximize exposure) is not entirely surprising. Flanking has never been reported in S. americana Unfortunately, the experiment was not designed to investigate how body orientation and body temperature interact, and the duration of orientation was never recorde d. Such information is necessary to determine how orientation directly affects body temperature in real time a nd vice versa. However, mean behavioral ratings were used to determine the effects of behavior on mean daily body temperature. Increased mean behavioral orient ation rating reflected increases in the mean difference between the body temperature of fr ee roaming grasshoppers and those of shade constrained grasshoppers (Fig. 213a) and decreases in the mean difference with sun constrained grasshoppers (Fig. 2-13b). As grasshoppers spend more time basking, their mean body temperatures rise above those of shaded grasshoppers and toward those of sun constrained grasshoppers which proves behavior has an effect on body temperature in S. americana Additionally, of the two grasshoppe rs tested each day, the one with the higher mean behavioral orientation rating always had a higher mean body te mperature for that day. This shows that the behavioral orientation ratings used in this study were suitab le for describing grasshopper behavior with respect to sunli ght absorption. The fact that mean daily behavioral orientation rating decreases with increases in mean daily ambient temperatur e (Fig. 2-14) corresponds with

PAGE 46

46 the previous idea that grasshoppers do not use all the heat availabl e to them on very warm days. It also shows that grasshopper behavior is dependent on environmental conditions. The more ambient heat that is available to a grasshoppe r, the less heat it must obtain through basking. Behavior obviously has an effect on body temper ature, and the linear plots of body temperature (Figs. 2-2 2-9) reinforce this fact even further by showing differences in body temperature between the two different free roaming grasshop pers for each day which experienced the same exact environmental conditions. The differences alone do not provide enough evidence for behavioral thermoregulation, but help make a strong case in conjunction with the other behavioral data, especially the comparison betw een mean behavioral rating and mean ambient temperature (Fig. 2-14). The multiple regression analysis for free roam ing grasshoppers revealed results similar to those for constrained grasshoppers. Both sunlig ht and ambient temperature had significant positive linear relationships. Again, the rela tionship between body temperature and ambient temperature remained linear. Behavior was also found to be a significant factor. However, behavior only accounted for a small portion of the total r 2 (Table 2-2). The data plotted for the partial regression plot between be havioral orientation and body temp erature (Fig. 2-15c) tries to show the direct effects of behavior on body temper ature. However, this data takes on a columnar appearance because the times of data collection for a continuous variable, body temperature, do not perfectly coincide the times of data coll ection for a discontinuous variable, behavior. As mentioned before, the duration of behavioral orie ntations were not reco rded and behavior was only recorded every 15 min, then paired with th e corresponding 15 min of temperature data. This data cannot account for any changes in behavior between the 15 min intervals. Even so, a general trend can be seen, that as beha vioral orientation ra ting increases so does body temperature (Fig.

PAGE 47

47 2-15c). Unlike the data for constrained grasshop pers, the data for free roaming grasshoppers are much more dispersed along the y-axis (Figs. 215a and 2-15b). This is most likely because behavior allowed grasshoppers to maintain simi lar body temperatures over a broader range of environmental conditions and shows how behavi or can affect body temperature by controlling for environmental conditions. Why ambient temperature is the most signifi cant factor contributing to the variation in body temperature in free roaming grasshoppers is uncer tain. It could be that the behavioral data are discontinuous and do not accurately represent how behavior affects body temperature. It is more likely, though, that the effects of ambient te mperature on body temperature are not affected by grasshopper behavior (unlike th e effects of sunlight which can be avoided by seeking shade) and, therefore, ambient temperature is the most consistent factor affecting body temperature. The fact that the model for free roaming grasshoppers (Fig. 2-15d) is not as accurate as those for constrained grasshoppers is likely due to the tim e discrepancies and discontinuous nature of the behavioral observations. Temperature Fluctuations and Adverse Weather Conditions Rates of body temperature incr ease in sun constrained gra sshoppers averaged 1.24C/min 0.009 (mean SD). These results are somewhat si milar to the rate of increase of 0.828C/min in Locusta reported by Uvarov (1966a). Lactin and Johnson (1997) provide a measurement of overall temperature increase per W/m2, but give no measurement of rate. The maximum rate recorded, 2.5C/min, is very high and at such rates grasshoppers could reach optimal body temperature in a very short period of time. Rate s of decrease and % decrease caused by adverse weather conditions (Table 2-3) s how just how significant the imp act of cloud cover or rain can be on grasshopper body temperature. Unexpectedly, cloud cover caused a faster rate of decrease

PAGE 48

48 than did rainfall. However, rates were calculate d using the temperature when the event started and the lowest temperature record ed within the decrease. Rainfall caused much larger percentage decreases in temperature than did cloud cover, and at lower temperatures the rate of body temperature decrease tapered off, therefore causi ng decreases due to rainfall to have a reduced rate. There were many small fluctuations in grasshopper body temperature throughout the day. These changes are possibly due to several factors, including small changes in sunlight intensity, changes in evaporative cooling, and changes in wind speed. The timing of these small changes in body temperature is similar between sun and shad e constrained grasshoppers, but the amplitude of change is often larger in sun constrained grasshoppers than in shade constrained grasshoppers. Sun constrained grasshoppers were more expos ed to small changes in both wind speed and sunlight intensity, whereas shade constrained gr asshoppers were sheltere d from such changes by vegetation. All of these observa tions suggest that these sma ll temperature changes are due mainly to small changes in abiotic factors. Aside from limiting a grasshoppers ability to achieve optimal body temperature, adverse weather conditions caused grasshoppers to experien ce an average of 3.67 large fluctuations in body temperature (> 15%) during daylight hours. Wh en trying to model insect populations, it is common to use data collected from subjects reared under constant conditio ns. Whether or not it is proper to use such data should be inves tigated on an individual basis. Depending on the answer, it might be possible to use daily averages for environmental parameters to model insect populations, or it may be necessary to record and include fluctuations that occur in environmental parameters. The next part of th is study attempts to determine if there are developmental and reproductive cons equences associated with daily fluctuations in temperature, such as those observed in th e field portion of this study

PAGE 49

49 Table 2-1. Grasshopper body temperatures and beha vioral orientation ratin gs, and environmental data (mean SD) obtained from field st udies. Differences between constrained grasshoppers and ambient temperature are gi ven in percent difference. See text for explanation of treatments and e nvironmental parameters. June 14 2005 June 17 2005 June 22 2005 Parameter Mean Min Max Mean Min Max Mean Min Max Free Roaming Blue (C) 37.03 2.99 29.80 44.40 37.37 2.16 32.90 43.40 28.19 7.42 20.90 47.20 Blue Behavior 0.56 0.43 0.38 0.41 0.73 0.40 Free Roaming Red (C) 35.61 2.61 30.60 42.70 41.71 1.84 34.70 44.90 27.45 6.25 20.80 43.80 Red Behavior 0.36 0.35 0.78 0.23 0.46 0.45 Shade Constrained (C) 33.61 2.52 28.20 38.80 33.82 1.87 28.10 36.10 26.20 4.27 21.10 37.10 Sun Constrained (C) 39.31 3.11 33.20 45.80 40.36 3.30 32.60 45.70 28.13 7.77 20.20 50.40 Ambient Temperature (C) 30.41 1.77 27.80 33.40 28.95 1.45 26.60 31.40 27.68 1.37 25.10 30.80 Shade Const. Ambient (%) 10.44 2.91 1.40 18.30 16.86 4.67 -0.60 26.90 8.76 6.26 -7.90 22.10 Sun Const. Ambient (%) 29.64 12.44 2.69 62.60 39.35 8.58 14.94 55.90 30.26 18.22 -7.60 76.40 % RH 58.43 8.66 36.80 76.70 68.19 5.91 54.30 79.90 Sunlight Intensity (w / m2)

PAGE 50

50 Table 2-1. Continued. July 5 2005 July 8 2005 Parameter Mean Min Max Mean Min Max Free Roaming Blue (C) 38.17 1.45 33.80 41.60 37.40 1.85 32.50 41.30 Blue Behavior 0.12 0.24 0.06 0.13 Free Roaming Red (C) 38.32 1.87 32.10 42.90 37.35 2.09 31.60 42.20 Red Behavior 0.13 0.25 0.07 0.15 Shade Constrained (C) 37.68 1.68 32.80 40.10 37.76 2.27 32.10 41.30 Sun Constrained (C) 45.90 2.96 33.20 52.10 46.13 3.42 33.50 51.60 Ambient Temperature (C) 32.20 1.04 28.50 33.30 32.58 1.25 28.70 34.50 Shade Const. Ambient (%) 17.00 2.29 9.56 21.35 15.83 3.55 2.90 22.90 Sun Const. Ambient (%) 42.47 6.08 16.60 57.70 41.66 9.90 2.40 56.20 % RH 51.07 3.28 46.50 63.70 56.19 6.52 47.80 74.10 Sunlight Intensity (w / m2) 801.40 99.50 420.60 979.40 805.70 204.80 0.60 1014.40

PAGE 51

51 Table 2-1. Continued. June 29 2005 November 17 2005 Parameter Mean Min Max Mean Min Max Free Roaming Blue (C) 30.30 3.62 25.30 40.40 27.21 4.28 17.30 38.40 Blue Behavior 0.60 0.40 Free Roaming Red (C) 30.88 3.92 25.10 41.40 27.63 5.12 18.70 37.60 Red Behavior 0.60 0.37 Shade Constrained (C) 28.46 2.29 25.40 33.60 21.47 2.60 13.80 24.80 Sun Constrained (C) 31.84 4.57 25.30 47.80 26.35 3.94 15.80 33.10 Ambient Temperature (C) 27.29 1.09 25.00 29.10 21.84 1.43 20.70 27.40 Shade Const. Ambient (%) 4.13 4.79 -5.70 17.20 -0.90 15.19 -40.90 16.33 Sun Const. Ambient (%) 16.41 13.55 -6.60 66.60 21.58 21.79 -29.70 55.90 % RH 81.18 4.32 71.50 87.20 29.29 2.18 24.60 33.90 Sunlight Intensity (w / m2) 316.80 223.40 39.40 1164.40 460.20 196.80 51.90 731.90

PAGE 52

52 Table 2-2. Regression statistics for multiple re gressions between grasshopper body temperatures and environmental parameters from June 29, July 5, and July 8 2005. SS F-value P r 2 Variable Parameter Estimate t value P Type II SS Partial r 2 Sun Constrained Model 489263 50939 <0.0001 0.93 Intercept -6.6499 -17.12 <0.0001 1408 Error 37531 Ambient Temperature 1.2070 84.03 <0.0001 33910 0.0644 Sunlight Intensity 0.0165 138.46 <0.0001 92073 0.8644 Shade Constrained Model 180960 141308 <0.0001 0.97 Intercept -13.9944 -98.69 <0.0001 6236 Error 5003 Ambient Temperature 1.5149 288.82 <0.0001 53412 0.9539 Sunlight Intensity 0.0033 74.59 <0.0001 3562 0.0192 Free Roaming Model 192781 18664 <0.0001 0.82 Intercept -1.4235 -4.14 <0.0001 59 Error 42494 Ambient Temperature 1.0921 86.68 <0.0001 25867 0.7819 Sunlight Intensity 0.0049 46.30 <0.0001 7382 0.0312 Behavior 1.2106 20.68 <0.0001 1472 0.0063

PAGE 53

53 Table 2-3. Rate of temperature decrease and percent temperature decrease (mean SD) of body temperature in sun constrained gra sshoppers under cloudy or rainy weather conditions. Condition Rate of Decrease Min Max % Decrease Min Max Cloud Cover 0.0163 C/sec 0.0063 0.0074 C/sec 0.0269 C/sec 22.34 3.98 17.60 31.20 Rain Fall 0.0087 C/sec 0.0058 0.0031 C/sec 0.0146 C/sec 38.83 9.49 30.20 49.00

PAGE 54

54 Figure 2-1. Diagram of wiring method for r ecording internal body temperature of grasshoppers showing how to conn ect copper and constantan wires.

PAGE 55

55 Figure 2-2. Linear plot of continuous body temperatures of free roaming grasshoppers and cloud cover from May 19 2005. Cloud cover is represented by the discontinuous bars in the upper portion of the plot.

PAGE 56

56 Figure 2-3. Linear plot of c ontinuous body temperatures of (a) constrained grasshoppers and cloud cover and (b) free roaming gra sshoppers and cloud cover from June 14 2005. Cloud cover is represented by the di scontinuous bars in the upper portion of the plot.

PAGE 57

57 Figure 2-4. Linear plot of continuous body temperatures of (a ) constrained grasshoppers with ambient temperature and cloud cover a nd (b) free roaming grasshoppers with ambient temperature and cloud cove r from June 17 2005. Cloud cover is represented by the discontinuous bars in the upper portion of the plot.

PAGE 58

58 Figure 2-5. Linear plot of continuous body temperatures of (a ) constrained grasshoppers with ambient temperature and cloud cover a nd (b) free roaming grasshoppers with ambient temperature and cloud cove r from June 22 2005. Cloud cover is represented by the discontinuous bars in the upper portion of the plot.

PAGE 59

59 Figure 2-6. Linear plots of continuous body temperatures and environmental parameters from June 29 2005. a) Constrained grasshoppers. b) Free roaming grasshoppers. c) Ambient te mperature and relative humid ity. d) Sunlight intensity. Secondary axis of c given in %RH.

PAGE 60

60 Figure 2-7. Linear plots of continuous body temperatures and environmental parameters from July 5 2005. a) Constrained grasshoppers. b) Free roaming grasshoppers. c) Ambient te mperature and relative humid ity. d) Sunlight intensity. Secondary axis of c given in %RH.

PAGE 61

61 Figure 2-8. Linear plots of continuous body temperatures and environmental parameters from July 8 2005. a) Constrained grasshoppers. b) Free roaming grasshoppers. c) Ambient te mperature and relative humid ity. d) Sunlight intensity. Secondary axis of c given in %RH.

PAGE 62

62 Figure 2-9. Linear plots of c ontinuous body temperatures and environmental para meters from November 17 2005. a) Constrained grasshoppers. b) Free roaming grasshoppers. c) Ambient te mperature and relative humid ity. d) Sunlight intensity. Secondary axis of c given in %RH.

PAGE 63

63 Figure 2-10. Variation in the pe rcent temperature difference betw een the body temperature of sun constrained grasshoppers and ambi ent temperature in relation to mean sunlight intensity from June 29, July 5, July 8, and November, 17 2005 (F=255.13, P=0.004 df = 1, 2).

PAGE 64

64 Figure 2-11. Regression analysis of sun constrained grasshoppers from June 29, July 5, and July 8 2005. a) Partial regression pl ot of body temperature and ambient temperature. Y axis represents residu als of regression between body temperature and sunlight intensity. X axis represents residuals of regressi on between ambient temperature and sunlight intensity. b) Partial regression plot of body temperature and sunlight intensity. Y axis represen ts residuals of regression between body temperature and ambient temperature. X axis represents resi duals of regression between sunlight intensity and ambient temperature. c) Plot of actual vs. predicted body temperature (t = ambien t temperature, s = sunlight intensity).

PAGE 65

65 Figure 2-12. Regression analysis of shade constr ained grasshoppers from June 29, July 5, and July 8 2005. a) Partial regression pl ot of body temperature and ambient temperature. Y axis represents residu als of regression between body temperature and sunlight intensity. X axis represents residuals of regressi on between ambient temperature and sunlight intensity. b) Partial regression plot of body temperature and sunlight intensity. Y axis represen ts residuals of regression between body temperature and ambient temperature. X axis represents resi duals of regression between sunlight intensity and ambient temperature. c) Plot of actual vs. predicted body temperature (t = ambien t temperature, s = sunlight intensity).

PAGE 66

66 Figure 2-13. Regression analysis of behavioral data from June 14, June 17, June 22, June 29, July 5, and July 8 2005. a) Mean behavior al orientation rating plotted against the difference between free roaming body te mperature and shade constrained body temperature in (F=13.14, P=0.005, df =1, 10). b) Mean behavi oral orientation rating plotted against the difference between free roaming body temperature and sun constrained body temperat ure (F=168.85, P=<0.0001, df =1, 10).

PAGE 67

67 Figure 2-14. Regression analysis for mean behavi oral orientation rating plotted against ambient temperature (F=27.58, P=0.0004, df =1, 10).

PAGE 68

68 Figure 2-15. Regression analysis of free ro aming grasshoppers from June 29, July 5, a nd July 8 2005. a) Pa rtial regression plot of body temperature and ambient temperatur e. Y axis represents residuals of re gression between body temperature and sunlight intensity and behavioral or ientation rating. X axis represents resi duals of regression between ambient temperature and sunlight intensity an d behavioral orientation ra ting. b) Partial regression pl ot of body temperature and sunlight intensity. Y axis represen ts residuals of regression between body te mperature and ambient temperature and behavioral orientation rating. X axis represents residuals of regression betw een sunlight intensity and ambient temperature and behavioral orientati on rating. c) Partial regressi on plot of body temperature a nd behavioral orientation rating. Y axis represents residuals of regression between body temperature a nd ambient temperature and sunlight intensity. X axis represents residuals of regression between behavioral orient ation rating and sunlight intensity and ambient temperature. d) Plot of actua l vs. predicted values for body temperature (t = ambient temperature, s = sunlight intensity, b = behavior al orientation rating).

PAGE 69

69 CHAPTER 3 EFFECTS OF TEMPERATURE FLUC TUATIONS ON DEVELOPMENT AND REPRODUCTION Methods and Materials This portion of the study was conducted in the laboratory in rearing chambers to determine how development time, adult body size, and female fecundity is affected by ecologically relevant temp erature fluctuations. Temperature Treatments Laboratory experiments took pl ace in environmental contro l chambers. The temperature fluctuation data obtained from the field was used as a template to construct temperature regimes. Average daytime temperatures of 33 C and 38 C were chosen to repr esent overcast and sunny days. The chambers were placed on a 14/10 photop eriod and a 16/8 thermo period with 16 h of respective daytime temperature with the last 2 h hours slowly falling to the 8 h of nighttime temperature at 25 C. After averaging in nighttime temperatures, the two mean temperatures were 29.5 C and 31.5 C, respectively. The mean temperatures of 29.5 C and 31.5 C were assigned different frequencies and amplitudes of temperat ure change. Frequency of temperature changes were 1 and 5 changes per daytime period, while the amplitude of change was 4 C and 8 C. The fifth temperature regime was that of constant temperature at 29.5 C or 31.5 C. This provided each mean temperature with four fluctuating trea tments and one constant temperature treatment for a total of 10 treatments (Fig. 3-1). Unfortuna tely, because of two failed attempts and time and space restrictions, the high frequency treatments were conducted from October 2005 to February 2006, while low frequency and constant temperatur e treatments were conducted from February 2006 to June 2006. During this second time period, the 31.5 C high frequency low amplitude treatment was repeated as a cont rol between the two time periods.

PAGE 70

70 Cages Four replicates were performed per treatment Each replicate consisted of an aluminum cage with dimensions of 30 cm X 30 cm X 30 cm, with a solid aluminum bottom, and screen sides and top, stocked with 15 1st instars hatched w ithin the last 24 hours. Each cage was provided with a petri dish of dry food (wheat flour, wheat bran, soy flour, and tropical fish flakes) and a water supply. Fresh romaine lettu ce was provided daily until reproduction began, at which point lettuce was provided tri-weekly. Cages were rotated daily to compensate for any temperature stratification that might occur within the chambers. Grasshopper frass was removed daily and cages were cleaned as necessary. Nymphal Developmental Time Grasshoppers were monitored daily for molting. Each day, grasshoppers were counted and the number surviving and the number at each stage were recorded. The stage of any dead grasshoppers was also recorded. Data were entere d into a Microsoft Excel spreadsheet. Average nymphal development time for each cage was calc ulated by taking the number of grasshoppers molting to the next stage each day and dividing it by the number surviving, and then adding the products from each day until all grasshoppers had reached adulthood. Body Size and Reproduction Once all the grasshoppers in a treatment reached adult hood, the grasshoppers were removed and the sex, femur length, and overall lengt h of each grasshopper were recorded. Even numbers of females and males were then placed back into the cages. Grasshoppers were then monitored for mating behavior, at which point 32 oz. deli cups (appr oximately 14cm high, 8.9cm in diameter at base, and 11.4cm in diameter at top) filled with moist vermiculite were placed inside the cages as oviposition medium. Each cup was then monitored daily until the first

PAGE 71

71 occurrence of oviposition was recorded. Cages were then monitored tri-weekly and cups were changed weekly. During this time, any egg pods that had been laid outside the cups were recorded. Cups were then stored at 5-10 C until they could be pro cessed, at which time the number of egg pods per cup was recorded, along with the number of eggs in each pod. The number of egg pods laid per cage was divided by th e number of females per cage to arrive at egg pods / female. Those egg pods laid outside of the cups were included in the egg pod mean but not the eggs/pod mean. Unfortunatel y, total reproduction could not be measured due to pesticide residues detected on lettuce fed to the genera l colony during the high frequency treatments. Experimental grasshoppers received the inner po rtions of this lettuce and consequently, all treatments had to be terminated for fear of contamination. Thus, the collection of egg pods was discontinued 115 d after hatch. The same was done for the second set of treatments conducted from February to June. Statistical Analysis Differences in total nymphal development tim e, body size, egg pods per female, eggs per pod, and days to first oviposition were anal yzed using ANOVA and the Least Squares Means (LSM) procedure using a Tukeys adjustment in SAS Analyst 9.0. High frequency treatments conducted from October to February were treated as a separate st udy and analyzed in a separate ANOVA from low frequency and co nstant treatments conducted fr om February to June. An ANOVA was conducted on all fluc tuating treatments as one group, combining differing frequencies and time periods to test for any freque ncy or trial effects such as, difference in food quality or change in biology due to seasonal di fferences. Separate hist ograms for high and low frequency treatments as well as one for combined treatments of nymphal developmental data

PAGE 72

72 were produced using SAS Analyst 9.0. Bar graphs of body size, egg pods per female, eggs per pod, and days to first oviposition we re produced using Microsoft Excel. Results To see if the two laboratory studies coul d be combined, an ANOVA was performed for each parameter on the repeated and original 31.5C high frequency low amplitude treatments. The tests revealed significant differences betw een the two in all parameters. Due to these differences, data were analyzed as two separate studies. Data were also analyzed together to note any effects, regardless of whether they were frequency or trial related. Nymphal development time Amplitude and temperature effects were found to be significant in high frequency treatments (Table 3-1). On average, grasshoppers reared at 31.5C reached adulthood 5.2 d faster than those reared in 29.5C treatments, while those reared in high amplitude treatments reached adulthood 2.1 d faster than grasshoppers reared in low amplitude treatments (Table 3-2). Mean comparisons revealed that the 29.5C high fre quency low amplitude treatment is the only significantly different treatment within high fr equency treatments (Tab le 3-2). Although both 29.5C treatments had longer average nymphal development times, only the low amplitude treatment was found to be significantly different from the 31.5C treatments. This treatment likely caused all of the amplitude and the majority of temperature effect s. The distribution of data for low amplitude treatments shows a distinct separation (Fig. 3-2) which is caused by the 29.5C low amplitude treatment. Grasshoppers in both 31.5C treatments took about 38 d to complete nymphal development, showing no effects of amplitude at this temp erature (Table 3-2). Amplitude, temperature, and amplitude x temper ature effects were found to be significant in low frequency and constant treatments (Table 3-1). On average, grasshoppers reared at 31.5C

PAGE 73

73 reached adulthood 1.6 d faster than those reared in 29.5C treatments (Table 3-2). There was no difference in nymphal development time between grasshoppers reared at high and low amplitude in low frequency treatments. However, those rear ed at alternating temp erature reached adulthood about 2 d faster than grasshoppe rs reared in constant treatme nts (Table 3-2). As in high frequency treatments, mean comparisons revealed only one treatment to be significantly different from the others. The 29.5C constant temperature treatment was significantly different from all treatments except for the 29.5C low frequency lo w amplitude treatment and is the likely the reason there are significant amplitude, temperatur e, and interaction effects (Table 3-2). The distribution of data reveals how similar nymphal development times for low frequency treatments are (Fig. 3-3). The data for constant temperature treatments is very spread out and is likely due to the 29.5C and 31.5C constant temperat ure treatments being so different (Fig. 3-3). Grasshoppers reared in the 31.5C constant treatment showed no delay in nymphal development and even developed faster than one of the alternating 31.5 C treatments. When high and low frequency treatments were combined (minus constant treatments) there were significant frequency or/and trial e ffects (Table 4). In general, low frequency treatments tended to have shorter mean tota l nymphal development times than high frequency treatments, 3.9 d shorter (Table 3-3). This can also be seen in the distribution of the data (Fig. 34). However, this difference is likely caused by the 29.5C high frequency treatments. All other treatments (29.5C low frequency, and both 31.5C high and low frequency treatments) were not significantly different (Table 3-3) When all treatments were combined there were no amplitude effects (Table 3-3).

PAGE 74

74 Body size Sex was the only significant main effect for both femur length and overall length (Table 3-4). In high frequency treatments females had an overall length (in mm, mean SD) of 59.81 1.84 and a femur length (in mm, mean SD) of 26.45 1.13, while males had an overall length of 51.91 1.94 and a femur length of 23.04 1.17. In low frequency and constant treatments females had an overall length of 61.91 1.84 and a femur length of 27.82 2.59, while males had an overall length of 53.97 1.75 and a femur length of 24.29 0.97. Differences in size between male and female grasshoppers were e xpected. There were also several significant interactions involving the sex factor, but a closer look at multiple comparisons revealed no real differences other than sex. When low frequenc y and high frequency treatments were combined, there were both significant sex and frequency or /and trial effects for femur length and overall length (Table 3-4). High frequenc y treatments tended to produce sm aller individuals than did low frequency treatments (Fig. 3-5). The only factor to affect gras shopper body size other than sex was frequency or/and trial effects. Days to oviposition Grasshoppers reared in high frequency trea tments, on average, took between 56 and 70 d from hatch to egg deposition. Temperature was th e only significant factor affecting the number of days to oviposition in high frequency treatmen ts (Table 3-5). Grass hoppers reared in 29.5C high frequency treatments took, on average, a bout 9 d longer to the first occurrence of oviposition (Table 3-6). Mean comparisons reveal ed differences similar to those for nymphal development time. The 29.5C high frequency low amplitude treatment was shown to be the only significantly different treatme nt (Table 3-6) and is likely the sole cause for the significant temperature effect. Grasshoppers reared in low frequency and constant treatments, on average,

PAGE 75

75 took between 56 and 93 d from hatch to egg depo sition (Table 3-6). The results of the ANOVA (Table 3-5) and mean comparis ons (Table 3-6) revealed only the 29.5C constant temperature treatment to be significantly different. This trea tment was so different fr om the others that it likely the reason for both significant temperatur e and amplitude effects. There were no differences between high and low amplitudes of low frequency treatments; however, there were differences between constant trea tments and fluctuating treatments (Table 3-6). Therefore, the amplitude effect was caused by constant temperat ure treatment and is not a true effect of differences in amplitude of temperature change but more of an effect of frequency or the presence or absence of temperature change(d isc). The ANOVA conducted on the combination of high and low frequency treatments (Table 3-5 and Fig. 3-6) revealed there to be no frequency or trial effect on the number of days from hatch to first oviposition, with hi gh frequency treatments averaging 61.8 d 7.3 (mean SD) and lo w frequency treatments averaging 63.1 5.6. Egg Pods / Female The results of the ANOVA (Table 3-5) and mean comparisons (Table 3-6) performed on high frequency treatments for egg pods / female revealed both signi ficant temperature and amplitude effects, and showed that only one treatment was significantl y different from the others, the 29.5C high frequency low amplitude treatment. Grasshoppers reared under this treatment produced an average 2 egg pods / female compared to the average of about 6 egg pods / female for all other treatments (Table 3-6). It is likely that both the amplitude and temperature effects were caused by this treatment. Just as in high frequency treatments, the ANOVA for low frequency treatments (Table 35) revealed significant temperatur e and amplitude effects. However, mean comparisons (Table 36) revealed more than one signi ficantly different treatment within the low frequency treatments.

PAGE 76

76 Grasshoppers reared in 29.5C treatments pr oduced on average 2.25 fewer egg pods / female than those reared in 31.5C treatments. This difference was caused by more than one treatment with all 29.5C treatments bei ng significantly different from all 31.5C treatments but the constant treatment (Table 3-6). The treatment with the fewest egg pods / female produced was the 29.5C constant treatment with only 1.27 egg pods / female (Table 3-6). The amplitude effect seen in low frequency treatments is essentially a frequency effect as there were no differences between low and high amplitude (Table 3-6). Th e only differences were between constant and fluctuating treatments. Females in both constant temperature treatments produced significantly fewer egg pods / females than many of their respec tive alternating treatments (Table 3-6). Unlike high frequency treatments, there is more th an one treatment causing differences between temperature and amplitude. When both high and low frequency treatments were analyzed together there were significant differences be tween the two frequencies (Table 3-5, Fig. 3-6) with females reared in high frequency treatm ents producing 5.10 1.93 (mean SD) egg pods / female and those reared in low frequency treatments producing 4.00 1.48 (mean SD) egg pods / female. Eggs / Pod Both temperature and amplitude were found to be significant factors affecting the number of eggs per pod in high frequency treatments (Table 3-5). Yet again, the only significant treatment of the high frequency treatments wa s the 29.5C low amplitude treatment with an average of 82.46 eggs / pod compared to the average of near 60 eggs / pod for all other treatments (Table 3-6). This treatments was responsible for both temperature and amplitude effects by causing the eggs / pod average to be greater for 29.5C treatments than 31.5C

PAGE 77

77 treatments and low amplitude treatments to average more eggs / pod than high amplitude treatments (Table 3-6). This is in contrast to low frequency treat ments, where grasshoppers reared in 29.5C treatments produced fewer eggs per pod than 31.5C treatments (Table 3-5 and 3-6). The ANOVA conducted on low frequency treatments show s a significant amplitude effect (Table 35). However, after examining the mean comparisons analysis (Table 3-6), it is clear that there are no real differences between high and low amplitudes and that, just as before, the amplitude effect is caused by differences between the constant an d alternating temperature treatments, which is more of a frequency effect. Females in bot h constant temperature treatments produced significantly fewer eggs / pod than did their resp ective alternating treatments. Interestingly, the 29.5C low frequency high amplitude treatment ha d a higher mean number of eggs / pod value than the 31.5C constant treatment and was not c onsidered significantly diffe rent from any of the 31.5C treatments. When both high and low fre quency treatments were combined, the ANOVA revealed no differences between high and low treatments in the number of eggs / pod. Females reared in high frequency treatments produced an average of 62.93 16.71 (Mean SD) eggs / pod and those reared in low frequency treatments produced 63.17 19.55 (Mean SD). Discussion Conformity of laboratory treatments to field study results Mean daily temperatures for laboratory treatments were based on mean daily body temperatures obtained from the field. When able the majority of free roaming grasshoppers maintained mean body temperatures near 38C (Table 2-1). Thus, the initia l temperature used to simulate mean daily temperature to represent ideal or sunny weather conditions was 38C. Free roaming grasshoppers on cloudy or rainy days often had body temperature near 30C (Table 2-

PAGE 78

78 1). However, environmental chamber limitations necessitated the use of a slightly higher temperature and so 33C was chosen to represen t the mean daily temperature for days with adverse weather conditions. In addition to the mean daily temperatures, frequency and amplitude of temperature change were also modeled af ter data obtained from the field. During ideal conditions on predominately sunny days, grasshoppe rs often experienced 01 large fluctuations (>15% change) in body temperature. The low fre quency treatments corresponded well with this and had only 1 fluctuation in temperature per da ytime period (Fig. 3-1). The fluctuation was placed near the end of the day because more of ten then not, cloud cover or rain on such days occurred during the afternoon hour s. During predominately cloudy and rainy days, grasshoppers experienced a mean of 3.67 large fl uctuations per day. In laboratory treatments 5 fluctuations per day were chosen which is not far from the average and likely provided more opportunity for differences between different fr equency treatments than 3 or 4 fluctuations might have. The mean percent change in body temperature of grasshoppers for cloudy and rainy days was 22.34% and 38.83% respectively. The initial amplitude of 8C corresponds well with those percentages for cloudy days. In 31.5C high amplitude treatments an 8C drop in temperature equals a 19% change in temperature. In 29.5C high amplitude treatments the 8C drop equals a 21.6% change in temperature. Low amplitude treatments we re included to complete the study design and provide a measurement of the effect of amplitude on the various parameters being investigated. The low amplitude of 4C was chosen by arbitrarily halving the 8C amplitude. Nymphal development time Total nymphal development time in this study ranged from 35.4 45.0 days, with an average of 38.4 days. The difference between average temperatures (29.5C and 31.5C) for nymphal development time was 5.2 days for high frequency treatments and 1.6 days for low

PAGE 79

79 frequency treatments. These differences are far less than many differences found in other studies (Parker 1930, Whitman 1986, Gndz and Glel 2002). Gndz and Glel (2002) reported a difference in development time of 10 d between 30C and 25C in S. gregaria One reason these results are different may be that the temperat ure difference between treatments was only 2C compared to the 5C or higher difference used in many other studies. An other possible reason is that much lower temperatures, such as 20C and 25C, were used in these other studies. These low temperatures may have been well below opt imal for the grasshoppers being studied, while the higher temperatures tested would have been within their optimal range. As mentioned before, it may not be the temperature difference but where the temperatures fall with respect to optimal temperatures for that species, that matters most. If an increase in 5C caused the new temperature to cross from the suboptimal range into the optima l range, a greater effect would likely be seen than if the 5C change occurred within the same range. This can be seen in a study by Gardner and Thompson (2001) on development time in H. viridis where 5C increases in temperature above 30C did not decrease development time. Th e lower temperatures tested in the current study likely fell closer to optimal fo r this species than those of other studies and their respective species, resulting in diminished responses. Analysis of nymphal development time in high frequency treatments shows the only significantly different treatment as being the 29.5C low amplitude treatment. Willot (1992) reports that for four species of Acrididae, op timum temperature for growth and development is between 35C and 40C. Daytime temperatures within the 29.5C treatments averaged 33C, which is just below the optimal range reported by Willot. Temperatures in the low amplitude treatment ranged from 31C to 35C, while thos e in the high amplitude treatment ranged from 29C to 37C. The body temperatures of grassho ppers reared in the low amplitude treatment

PAGE 80

80 never reached above 35C, while body temperatures of those reared in high amplitude treatments reached levels near the middl e of the optimal range reporte d by Willot (1992). While Willot (1992) does not report optimal body temperatures for S. americana they are likely to at least be within the 35-40C range, if not higher. In contrast to the 29.5C treatments, 31.5C treatments had mean daily temperatures of 38C with lo w amplitude temperatures ranging between 36C and 40C and high amplitude temperatures rang ing between 34C and 42C, all of which fall very near or within the theore tical optimal range. The fact th at only the grasshoppers in the 29.5C low amplitude treatment had a slower de velopment rate, suggests that amplitude of temperature change may only be important for de velopment time at suboptimal temperatures and that it may be possible that just attaining 37C for some portion of the day is beneficial for the development of S. americana Reporting that amplitude only effects nymphal development at suboptimal temperatures entails that there should be a significant inter action between amplitude and temperature. While the intera ction was not considered significant, it was very close with a pvalue equal to 0.06. A situation similar to that of high fre quency treatments occurred in low frequency treatments, with only one treatm ent being significantly different. However, instead of the 29.5C low amplitude treatment being significantly di fferent, it was the 29.5C constant temperature treatment. Development time in the 29.5C lo w amplitude treatment falls between the 29.5C constant treatment and the rest of the treatments, and is not statistically different from any of the treatments. While every treatment received exactly the same number of degree days with respect to their mean temperatures (calculated using da ta from environmental chambers), grasshoppers reared in low frequency treatments experienced the high range of thei r temperature range for longer uninterrupted periods of time than grassho ppers of high frequency treatments. This could

PAGE 81

81 be the reason that nymphal development time in the 29.5C low amplitude treatment was significantly different in high frequency treatments but not in low frequency treatments. The exposure to uninterrupted periods of 35C likely provided a greater benefit than several shorter periods of exposure. If this is true, it suggests that the duration (frequency) of uninterrupted high temperatures is important for development. The significant difference of the 29.5C c onstant treatment causes there to be both significant amplitude and temperature effects. However, there are no differences between high and low amplitude treatments. Therefore, the signi ficant amplitude effect is an artifact of the constant temperature treatments and is essentially a frequency effect manifested as a difference between low frequency and no frequency or the pr esence or absence of temperature change. Data was analyzed this way to preserve the complete block design. Had frequency been analyzed as low frequency and no frequency the model would have lacked high and low amplitudes for the constant treatments since there cannot be amplitude of change if there is no temperature change. As in high frequency treatments, the amplitude fa ctor (or in this case frequency) only had an effect at the lower temperature (Table 3-2), suggesting an interacti on between amplitude and temperature. However, in contrast to the high frequency treatments, the amplitude X temperature interaction is significant (Tab le 3-1). Again, the data suggest that nymphal development is affected by fluctuating temperature only at suboptimal temperatures. Because different frequency treatments we re conducted as different trials, and the treatment repeated between the two trials was found to be signi ficantly different, any significant frequency effects between high and low frequenc y treatments are conf ounded with any trial effects. However, the data can still be analyzed if any effects of fre quency between the two are also attributed to differences between the trials. It was expe cted that low frequency might

PAGE 82

82 develop faster because they spent a longer c ontinuous duration of time at high temperatures, even though they received the same number of de gree days as that of hi gh frequency treatments. While nymphal development time of grass hoppers experiencing 31.5C treatments remain relatively the same between high and low freque ncy treatments, development time of 29.5C treatments is much shorter, 6.5 d shorter (Tab le 3-3), in low frequency treatments when compared to high frequency treatments, suggesti ng again that temperature change only affects nymphal development time at suboptimal temp eratures. However, development times of grasshoppers reared at a constant 29.5C were shorter than that of grasshoppers reared in either 29.5C high frequency treatment (Tab le 3-2). However, it was expect ed that constant temperature treatments would exhibit the slowest development times. This, along with the idea that fluctuations in temperature that allow the attain ment of higher temperatures are more beneficial for nymhpal development at lower mean temperatures leads to the explanation that trial effects, such as seasonal differences, are likely the cause for the faster development times seen in constant temperature treatments when compared to high frequency treatments. However, if the differences between high and low frequency treatments are due to trial affects, then another question arises. Why is there no acceleration of nymphal de velopment time in the 31.5C treatments between high and low frequency treat ments? There are two possible explanations assuming experimental error is not the cause. Tr ial effects are more prominent at suboptimal temperatures or frequency effects are more prominent at suboptimal temperatures. The differences due to a trial effect such as seasona lity would not likely be affected by suboptimal temperatures; therefore, the more plausible explanation is that differences in nymphal development time between high and low frequenc y treatments are due to frequency, whose effects are temperature dependent. This idea is supported by the fact that the same patterns are

PAGE 83

83 seen when frequency treatments are analyzed sepa rately, as discussed earlier. Still, whether or not differences in nymphal development time be tween low and high frequency treatments should be attributed to frequency or tria l effects remains somewhat unclear. Body size Studies in the past have show n that thermoperiod can affect weight and size in insects (Beck 1983). However, no such effect was obs erved in this study. Neither temperature fluctuations nor differences in m ean temperature affected body size in S. americana Fielding (2004) reported that body weight of M. sanguinipes was lower at the low (21C and 24C) and high (39C and 42C) extremes (with respect to optimal for M. sanguinipes ) of the temperatures tested, relative to intermediate temperatures. In the current study, body we ight was not recorded and a lack of difference in size does not necessar ily equate to a lack of difference in weight. There may have been a possibility that there were differences in weight. However, temperatures tested in the current study woul d not be considered extreme, a nd it is likely that there were no differences in weight, just as there were no differences in size. As expected, there were differences between the sexes, but when high and low frequency treatments were combined there were significant differences between the two freque ncies. No other facet of temperature affected body size when the treatments were analyzed separa tely and, therefore, it is unlikely frequency would have an effect and the differences betwee n high and low frequency treatments were most likely caused by trial effects. If tr ial effects affected a parameter not affected by temperature, it may be that the differences between trials also a ffected other parameters mo re so than frequency. This is in contrast to the ny mphal developmental data, which s uggest that differences between high and low frequency treatments mi ght actually be due to frequency.

PAGE 84

84 Days to oviposition In high frequency treatments the 29.5C low amplitude treatment was the only significantly different treatment and there were no differences in sexual maturation between the 29.5C high amplitude treatment and the 31.5C treatments for the number of days to first oviposition. The results suggest that amplitude only has an effect at suboptimal temperatures and that mean body temperatures are likely not as important as maximum body temperatures experienced. Just as in nymphal development time, reaching 37C when in the 29.5C high amplitude treatment provided some benefit. The analysis conducted on low frequency treatments for days to oviposition also reve aled only one significantly different treatment. However, just as for nymphal development time, the treatment was the 29.5C constant treatment and not the 29.5C low amplitude treatment. The reasons for this are likely the same as those given for development time, having to do with the dur ation of continuous tim e spent in the high temperature range of the treatments, suggesting a frequency effect. Together both sets of results suggest that the number of days to oviposition is affected by both amplitude and frequency of temperature change. Because the differences we re only seen in the 29.5C treatments, it is strongly suggested that these differences onl y occur at suboptimal temperatures. The separate analyses conduc ted on both high and lower frequency treatments for days to oviposition revealed results very similar to those for nymphal development time. This is not surprising as both nymphal development time and sexual maturation are both developmental parameters (together equaling total developmen t time) and should be affected in a similar manner. However, when both high and low freque ncy treatments were combined, the analysis revealed the effect of frequency to be non-si gnificant, contrary to what was found for nymphal development time. Why this occurs is uncer tain, but if the difference between nymphal

PAGE 85

85 development times of both high and low frequency trea tments is due to trial effects, it would be very unusual for the frequency or trial effect to be absent in sexual matu ration rate considering the two parameters are closely related. Gndz and Glel ( 2002) reported that S. gregaria needed to reach a critical weight before reproduction began, and that temperature did not affect th is critical weight but only how fast it was attained. It may be that 29.5C hi gh frequency and constant treatments did not maintain high enough temperatures for long enou gh periods of time for grasshoppers to quickly reach this critical weight. One explanation fo r why differences were observed between 31.5C treatments and 29.5C high frequency low amplitude and constant treatments and not the low frequency 29.5C treatments is that the low fr equency 29.5C treatments may have maintained the high limit of the temperature ra nge long enough to increase rate of weight gain. In contrast, the less optimal treatments (29.5 C high frequency low amplitude and 29.5C constant) would have decreased rates of weight gain and de layed sexual maturation. Delayed sexual maturation could have several negative effects on a gr asshopper population. Grasshoppers with delayed sexual maturation are more likely to die of diseas e or predation before be ing able to reproduce. In colder environments, those that survive will have less time to la y eggs, and those eggs that are laid will hatch later and be at a disadvantage co mpared to eggs which were laid earlier in the season. Measuring how many days it took female gra sshoppers to reach sexua l maturity posed a problem in this study and for that reason the accu racy of its measurement may have suffered. All other parameters were taken as cage averages (nymphal development time and pods / female) or as individual measurements of every availabl e grasshopper or egg pod (body size and eggs / pod). The number of days it took to first oviposition was record ed for only one individual per

PAGE 86

86 cage, the first female to lay an egg pod. While it was rather simple to determine total nymphal development time by recording when each grasshopper reached adult hood, it was impossible to record the first oviposition date of every grasshopper in the cage There was no way to determine which female the egg pod came from. Therefore, only 40 of the 600 grasshoppers used in this study were used for recording the number of days from hatch to oviposition. Egg pods / female Mean daily temperature had a significant effect on the number of egg pods laid per female, with 31.5C treatments laying, on average, 2 more egg pods / female than in 29.5C treatments for both high frequency and low fr equency and constant temperature treatments. Females in 29.5C constant treatment laid the lowest average with onl y 1.27 egg pods / female, while 31.5C alternating treatments laid 5-6 egg pods / female. This is a huge difference with respect to overall fecundity. These results ar e similar to those reported by Uvarov (1966a) on C. pellucida where females reared at 27C laid an average of only one egg pod, as opposed to the average of four egg pods laid by fe males reared at 37C. While the results of the an alysis clearly show that temperature has a significant effect on egg pods / female, the effects of amplitude and frequency of temperature change are much hard er to identify. The results from high frequency treatments are similar to those for developmen tal parameters with only the 29.5C low amplitude treatment being significantly different from a ll others, suggesting the same inferences made about nymphal development time might also be ab le to be made about fecundity. However, the results from low frequency and constant tr eatments are very different from those for developmental data with many treatment s being different from each other. The benefits on nymphal development time pr ovided by the longer peri ods of continuous high temperatures in low frequency treatments do not seem to apply to reproductive parameters.

PAGE 87

87 The 29.5C low frequency high amplitude treatment has a much lower average than its high frequency counterpart (Table 3-6). Additiona lly, the 29.5C low frequency treatments are statistically different from the 31.5C treatments, something not seen in developmental data. This suggests that continuous periods of the high range of temperatures have little if any effect on the number of egg pods / female. The notion that temper ature fluctuations have more of an effect at suboptimal temperatures is also harder to see because most of the significant differences are between mean temperatures. While the 29.5C cons tant treatment seems to have produced the fewest egg pods / female, it is not considered statistically different from the other 29.5C treatments or even the 31.5C constant treat ment (Table 3-6). On the other hand, the 31.5C constant treatment is significantly different from one of the 31.5C treatments. This is in contrast to what has been suggested earlier, that temp erature fluctuations are more influential at suboptimal temperatures. These results have ma de it difficult to decipher the effects of temperature, amplitude and frequency on the numb er of egg pods laid per female, especially when one considers that high frequency treatments produced, on average, one more egg pod / female than low frequency treatments and were considered statistically different from low frequency treatments. Eggs / pod Only the higher values of raw data for the number of eggs / pod were within the range of 76-100 eggs per pod reported by Kuitert and Conni n (1952). The means for each treatment, with the exception of one, were never wi thin this range. Why this is remains uncertain, but could be due to a couple of factors. It may be that cond itions in this study were not always optimal and could have caused a reduction in the number of eggs per pod. Additionally, the reduced number

PAGE 88

88 of eggs / pod might be an artifact of prolonge d laboratory colonization. Regardless, there were significant effects of temperature change on the number of eggs laid per pod. The effects of mean temperature on the numbe r eggs / pod are less obvious than those for egg pods / female. In high frequency treatments the females reared in 29.5C treatments produced more eggs / pod than those reared in 31.5C treatments, while the exact opposite occurred in low frequency treatments (Table 36). The differences in amplitude and temperature in high frequency treatments are due to the 29.5C high frequency low amplitude treatments, which is the only significantly different treat ment. Females from the 29.5C high frequency low amplitude treatment produced an abnormally la rge number of eggs / pod when compared to females from the rest of the treatments. This su ggests that suboptimal temperatures might cause an increase in the number of eggs / pod. It was originally hypothesized th at this increase in number of eggs / pod was possi bly a stress induced response to compensate for prolonged development and the reduced number of egg pods laid, as this treatment was always the only significantly different treatment of the high freq uency treatments. However, such compensation is not seen in equivalently affected low fre quency treatments. This leads to the notion that something may have been wrong with one or more elements of the experiment involving the 29.5C high frequency low amplitude treatment. The results for eggs / pod for low frequency treatments were very similar to those for egg pods / female, and many of the same conclusions can be drawn from the results. However, unlike egg pods / female, the 29.5C constant treatment was considered statistically different from other 29.5C treatments and the 31.5C constant treatment (Table 3-6). This exemplifies the negative effects that suboptimal temperatures can have on reproductive parameters. Again, because there are no real differences between high and low amplitu de treatments, the amplitude effect seen in

PAGE 89

89 low frequency treatments is actually caused by the presence or absence of temperature change. Unexpectedly, there was no frequency or trial ef fect on the number of eggs / pod when both high and low frequency treatments were combined, ev en though there were differences in frequency between constant and low frequency treatments. This was puzzling because there was some amount of frequency or trial effect s in every other parameter tested but days to first oviposition.

PAGE 90

90 Table 3-1. ANOVA results for nymphal developmen t time at different treatment combinations. Treatment Source SS df MS F P Amplitude 17.65117.655.34 0.0394 High Frequency Temperature 109.971109.9733.27 <0.0001 Amp x Temp 13.99113.994.24 0.0620 Error 39.66123.30 Amplitude 19.9029.955.59 0.0129 Low Frequency + Temperature 15.60115.608.99 0.0077 Constant Amp x Temp 39.47219.7311.09 0.0007 Error 32.03181.78 Amplitude 11.39111.394.56 0.0431 Frequency 123.871123.8749.58 <0.0001 High Frequency + Temperature 61.70161.7024.70 <0.0001 Low Frequency Freq x Amp 6.5916.592.64 0.1175 Temp x Amp 32.33132.3312.94 0.0014 Temp x Freq 48.66148.6619.48 0.0002 Temp x Freq x Amp 0.1610.160.06 0.8049 Error 59.96242.50

PAGE 91

91 Table 3-2. Mean comparisons of nymphal developm ent time (in days) in high and low treatments (HF = high frequency, LF = low fre quency, HA = high amplitude, LA = low amplitude). Means followed by different lett ers are significantly different at the 0.05 level (LSM, Tukeys adjustment). Treatment Mean SD Amplitude Mean SD Temperature Mean SD 31.5C HF HA 37.68 2.45 A HA 39.36 2.50 A 31.5C 37.80 1.76 A 31.5C HF LA 37.91 1.09 A LA 41.47 4.14 B 29.5C 43.04 2.66 B 29.5C HF HA 41.05 0.98 A 29.5C HF LA 45.02 2.25 B 31.5C LF HA 37.26 1.45 A HA 36.34 1.43 A 31.5C 36.30 1.51 A 31.5C LF LA 35.40 1.34 A LA 36.63 1.88 A 29.5C 37.94 2.45 B 31.5C Constant 36.26 1.48 A Constant 38.40 2.63 B 29.5C LF HA 35.42 0.68 A 29.5C LF LA 37.86 1.56 AB 29.5C Constant 40.54 1.31 B

PAGE 92

92 Table 3-3. Mean comparisons of nymphal development time (in days) in high (HF) and low (LF) frequency treatments at fixed levels of amplitude and temperature (HA = high amplitude, LA = low amplitude). Means followed by different letters are significantly different at the 0.05 level (LSM, Tukeys adjustment). Fixed Parameter Frequency Mean SD HA HF 39.37 2.50A LF 36.34 1.43B LA HF 41.47 4.14A LF 36.63 1.88B 29.5C HF 43.04 2.66A LF 36.64 1.71B 31.5C HF 37.79 1.76B LF 36.33 1.63B All HF 40.42 3.48A All LF 36.48 1.62B

PAGE 93

93 Table 3-4. ANOVA results for body size at different treatment combinations. Femur Length Overall Length Treatment Source SS df MS F P SS df MS F P Amplitude 4.18 1 4.18 3.26 0.0700 0.46 1 0.46 0.13 0.7200 High Frequency Sex 286.34 1 286.34 223.40 <0.0001 1497.47 1 1497.47 414.98 <0.0001 Temperature 0.00 1 0.00 0.00 0.9600 13.79 1 13.79 3.82 0.0540 Temp x Amp x Sex 6.94 1 6.94 5.42 0.0200 Error 119.20 93 1.28 335.59 93 3.61 Amplitude 2.81 2 1.41 0.36 0.7000 8.97 2 4.48 1.95 0.1400 Low Frequency + Sex 817.03 1 817.03 207.64 <0.0001 4102.27 1 4102.27 1786.85 <0.0001 Constant Temperature 1.71 1 1.71 0.43 0.5100 3.01 1 3.01 1.31 0.2500 Temp x Sex 21.47 1 21.47 9.35 0.0030 Error 991.57 252 3.93 578.54 252 2.30 High Frequency + Frequency 97.88 1 97.88 23.48 <0.0001 216.38 1 216.38 11.56 0.0008 Low Frequency Error 1146.62 275 4.17 5145.32 275 18.71

PAGE 94

94Table 3-5. ANOVA results for reproduction at different treatment combinations. Days to Oviposition Egg Pods / Female Eggs / Pod Treatment Source SS df MS F P SS df MS F P SS df MS F P Amplitude 27.56 1 27.56 1.03 0.3300 10.78 1 10.78 24.47 0.0003 3767.46 1 3767.46 16.78 <0.0001 High Frequency Temperature 351.56 1 351.56 13.17 0.0040 19.95 1 19.95 45.30 <0.0001 6914.56 1 6914.56 30.80 <0.0001 Amp x Temp 95.06 1 95.06 3.56 0.0800 19.95 1 19.95 45.30 <0.0001 3652.88 1 3652.88 16.27 <0.0001 Error 320.25 12 26.69 5.29 12 0.44 41089.37 183 224.53 Amplitude 1705.75 2 852.88 12.12 0.0005 17.32 2 8.66 12.09 0.0005 8441.19 2 4220.59 11.96 <0.0001 Low Frequency + Temperature 988.17 1 988.17 14.04 0.0020 32.03 1 32.03 44.71 <0.0001 9780.86 1 9780.86 27.75 <0.0001 Constant Amp x Temp 523.58 2 261.79 3.72 0.0400 1.59 2 0.80 1.11 0.3500 455.09 2 227.54 0.64 0.5200 Error 1266.50 18 70.36 12.89 18 0.72 107602.00 305 352.79 Amplitude 75.03 1 75.03 3.40 0.0800 4.57 1 4.57 9.57 0.0050 Frequency 13.78 1 13.78 0.62 0.4300 9.57 1 9.57 20.01 0.0002 5.68 1 5.68 0.02 0.8966 High Frequency + Temperature 504.03 1 504.03 22.83 <0.0001 45.36 1 45.36 94.87 <0.0001 Low Frequency Freq x Amp 1.53 1 1.53 0.07 0.7900 6.27 1 6.27 13.12 0.0014 Temp x Amp 5.28 1 5.28 0.24 0.6300 14.99 1 14.99 31.34 <0.0001 Temp x Freq 16.53 1 16.53 0.75 0.4000 0.18 1 0.18 0.31 0.5508 Temp x Freq x Amp 132.03 1 132.03 5.98 0.0200 5.98 1 5.98 12.51 0.0017 Error 529.75 24 22.07 11.48 24 0.48 139459.82 415 336.05

PAGE 95

95 Table 3-6. Mean comparisons of reproductive data in high and low frequency treatments (HF = High Frequency, LF = low frequency, HA = high amplitude, LA = low amplitude). Means followed by different le tters are significantly diffe rent at the 0.05 level (LSM, Tukeys adjustment). Treatment Mean SD Amplitude Mean SD Temperature Mean SD 31.5C HF HA 58.25 2.63 A HA 60.50 3.02 A 31.5C 57.13 2.10 A 31.5C HF LA 56.00 0.00 A LA 63.13 10.02 A 29.5C 66.50 7.67 B 29.5C HF HA 62.75 0.96 AB Days to 29.5C HF LA 70.25 9.95 B Oviposition 31.5C LF HA 56.50 1.00 A HA 61.38 7.25 A 31. 5C 62.58 7.88 A 31.5C LF LA 63.25 2.06 A LA 64.88 2.75 A 29.5C 75.42 15.99 B 31.5C Constant 68.00 11.55 A Constant 80.75 18.35 B 29.5C LF HA 66.25 7.63 A 29.5C LF LA 66.50 2.52 A 29.5C Constant 93.50 14.80 B 31.5C HF HA 5.92 0.57 A HA 5.92 0.68 A 31.5C 6.21 0.71 A 31.5C HF LA 6.51 0.78 A LA 4.28 2.45 B 29.5C 3.98 2.16 B 29.5C HF HA 5.92 0.87 A Egg Pods 29.5C HF LA 2.04 0.28 B / Female 31.5C LF HA 4.95 0.97 AB HA 3.94 1.37 A 31.5C 4.56 1.28 A 31.5C LF LA 5.58 0.39 A LA 4.07 1.68 A 29.5C 2.25 1.12 B 31.5C Constant 3.14 0.77 BC Constant 2.20 1.40 B 29.5C LF HA 2.93 0.82 C 29.5C LF LA 2.55 0.54 C 29.5C Constant 1.27 1.28 C 31.5C HF HA 59.12 11.09 A HA 60.88 14.37 A 31.5C 59.22 13.01 A 31.5C HF LA 59.27 13.92 A LA 64.43 18.16 B 29.5C 70.40 20.56 B 29.5C HF HA 62.79 17.17 A Eggs / Pod 29.5C HF LA 82.46 19.98 B 31.5C LF HA 68.11 19.89 A HA 63.73 20.22 A 31.5C 64.13 19.20 A 31.5C LF LA 66.68 18.54 A LA 62.63 18.97 A 29.5C 54.35 19.80 B 31.5C Constant 57.11 17.59 B Constant 53.23 19.38 B 29.5C LF HA 58.20 19.43 AB 29.5C LF LA 55.98 17.94 B 29.5C Constant 40.58 19.97 C

PAGE 96

96 Figure 3-1. Fluctuating laboratory temperatur e treatments for each mean temperature taken from environmental chamber data logge rs over a 24hr time period (resolution = 3min). Y axis is given in C. (HF = high frequency, LF = low frequency, HA = high amplitude, LA = low amplitude).

PAGE 97

97 Figure 3-2. Nymphal development time (d) of high frequency treatments at fixed levels of temperature and amplitude.

PAGE 98

98 Figure 3-3. Nymphal development time (d) of low frequency treatments at fixed levels of temperature and amplitude.

PAGE 99

99 Figure 3-4. Nymphal development time (d) for high frequency and low frequency treatments.

PAGE 100

100 Figure 3-5. Mean ( SD) femur and overall length at high and low frequency treatments (HF = high frequency, LF = low freque ncy). Columns designa ted by a different letter under their respective category ar e considered different at the 0.05 level.

PAGE 101

101 Figure 3-6. Mean ( SD) number of days to first ovi position, egg pods / female, and eggs / pod at high and low frequency treatme nts (HF = high frequency, LF = low frequency). Columns designated by a different letter under their respective category are considered different at the 0.05 level.

PAGE 102

102 CHAPTER 4 CONCLUSIONS Overall, the linear plots of body temperat ure (Figs. 2-2 2-9) show a general dependence of body temperature on s unlight intensity. This relations hip can also be seen in the multiple linear regression analysis of grasshopper body temperatures Data from this study have shown how important sunlight availability can be to S. americana for maintaining body temperatures above ambient and that there is a positive linear relationship between these environmental factors and body temperature. The ra tes of temperature increase measured in this study suggest that grasshoppers have the ability to raise their body te mperature by 2.5C per minute, if not faster, by basking in direct sunlig ht. At this rate, grasshoppers could more than double their body temperature in less than ten minut es and reach optimal temperature in even a shorter period of time. Furthermore, body temper atures of sun constrai ned grasshoppers were 12.4-19.6C above ambient temperature, often around 75% above ambient temperature and even reaching 50C Thus, sunlight plays a vital role in th e achievement and maintenance of optimal body temperatures above ambient temperature. Shade constrained grasshoppers usually ha d body temperatures above ambient, up to 26.9% above and even reaching 41.3% above ambi ent. In addition, body temperatures of all grasshoppers were shown, at times, to be signifi cantly higher than ambient temperature even in the absence of direct sunlight. If this is generally true, it w ould cause problems with phenology modeling and might explain why there often are discrepancies between predicted development rates or population numbers and actual ones. Cu rrent phenology models are often based on mean daily ambient temperature and developmental an d reproductive data obtained at constant temperatures. In order to develop accurate phe nology models, it may be necessary to obtain more

PAGE 103

103 accurate body temperatures either by direct m easurement or by estimation using measurements of sunlight intensity, temper ature and possibly wind speed. As expected, body temperatures of free roaming grasshoppers were often between those of shade and sun constrained grasshoppe rs. This study has shown that behavioral orientation toward direct solar radiation has a significant positiv e effect on body temperature and that there is a positive relations ship between basking behavior and body temperature and as basking increased body temperature increased. On the hottest days tested, when free roaming grasshoppers remained in the shade, their body temp eratures were very similar to those of shade constrained grasshoppers. This further exemplifie s the effects of behavior on body temperature. In addition, grasshopper behavior was found to be affected by ambient temperature and as ambient temperature increased ba sking behavior decrea sed. This helps prove that grasshoppers do not utilize all the heat avai lable to them and that baski ng is only used to raise body temperature when ambient temperatures are not high enough to provide sufficient heat. The differences between free roaming gr asshoppers and constrained grasshoppers, along with the regression analysis conducted on behavioral observati ons, especially the effect of ambient temperature on mean behavioral rati ng (Fig. 2-14), provides evidence showing that grasshoppers are behaviorally th ermoregulating. While the above di scussion might suggest that it would benefit those modeling phenology to meas ure more than just ambient temperature, accurately measuring behavior in the field for the purpose of improving models will likely prove to be a daunting task. A simpler solution may be to monitor environmental conditions to ascertain whether or not optimal body temper ature can be achieved by thermoregulating grasshoppers. It can be assumed that grasshoppe rs will attempt to maximize fitness whenever

PAGE 104

104 possible, and knowing when they are able to do so and how development and reproduction are affected, could help to better predict development time and popul ation numbers in the field. Another important facet of this study, which is directly related to the above discussion, was to examine how adverse weather conditions could affect body temperature and to assess whether even summer conditions in Florida c ould be sub-optimal for the development and reproduction of S. americana This study showed how frequen tly and quickly body temperature in grasshoppers can be affected by adverse weat her conditions, and how mean temperatures of free roaming grasshoppers were significantly redu ced on cloudy or rainy days, compared to those on predominately sunny days. Adverse weather conditions could cause causing up to 38% decreases in body temperature at a rate of up to 0.0269 C/sec and reduce mean body temperatures by up to 8C a day. It is easy to imagine that any prolonge d occurrence of these adverse weather conditions could have seve re negative implications on development and reproduction in S. americana, and grasshoppers in general. In Florida, average ambient temperature is high during the months when gras shoppers are most active, and sunlight is not required to complete development. However, periods of prolonged cloud cover and rain occur during the summer months in Florida, which can prevent grasshoppers from reaching optimal temperatures, and therefore optimal fitness. Knowing how weather affects grasshopper body temperature is part of knowing when grasshoppe rs will be able to optimize body temperature. Including the effects of advers e weather conditions on a grass hoppers ability to optimize body temperature into phenology models would likely help to improv e a models predictive power. At more optimal temperatures, nymphal develo pment time seemed to be little affected by temperature changes in this study, and there wa s little evidence of a Kaufmann effect in S. americana at the 31.5C temperatures tested. The fe w effects temperature fluctuations had on

PAGE 105

105 nymphal development time occurred in the lo wer temperature treatments and at the low amplitude and constant regimes. It is likely th at in 29.5C treatments, te mperature fluctuations provided enough of a rise in temperature to counter act some of the negative effects of rearing at suboptimal temperatures. The fact that the 29.5C constant treatment showed a deceleration in developmental rate compared to other alte rnating 29.5C treatments suggest a nonlinear relationship between body temperature and deve lopmental rate, and s ubsequently a Kaufmann effect at suboptimal temperatures. More tests at suboptimal conditions would be necessary to confirm this. While the results do not provide a clear explanation, what is clea r is that at least at lower temperatures, temperature fluctuations ha ve some significant effect on development time in S. americana even if it may be modest. The mean temperatures of 31.5C and 29.5C chosen for this study were taken from the field portion of this study a nd represent 24hr mean body temperatures for mostly cloudy and mostly sunny days. The small differences in development time detected between the two temperatures, and those caused by temperature fluc tuations, will likely have little effect on phenology modeling for pest management purposes, and for predicting development time in the field. Grasshoppers in the field will not all hatch at the same time, and therefore will not complete development at the same time, making differences in development time of 5 d practically irrelevant. However, it is still suggested that phenology modeling be based on body temperature rather than ambient temperature. Data involving female fecundity were more heavily affected by temperature changes and differences in temperature than development time, and could be more sensitive to small changes in temperature. Together, the data for egg pods / female and eggs / pod give an accurate view of how fecundity might be affected by differences in temperature and temperature fluctuations.

PAGE 106

106 Why reproduction was affected more during the months of February to June than October to February remains unclear. Differences in most of the other parameters showed that differences between frequencies were likely due to trial effects, and suggested that differences in reproduction would also be due to trial effects. One plausible expl anation for the differences is that there was some experimental error involv ed, and there were differences in the rearing methods between the two studies, possibly food quality. Overall, temperature was shown to a ffect nymphal development time, reproductive development time, and female fecundity. With few exceptions, higher temperature treatments increased developmental rate and female fecund ity. Amplitude was also shown to affect nymphal development time, reproductive development time, and female fecundity; however, the effects are less clear. Amplitude caused few differences between 31.5C treatments, with the exception for the 31.5C constant treatment. The effects of amplitude seemed to be more pronounced at the sub-optimal 29.5C treatments. Still, there is no clear pattern. The low amplitude 29.5C treatment within the high frequency treatment s was often the only treatment affected by amplitude, while in low frequency treatments th e 29.5C low amplitude treatment was often not different from the 29.5C high amplitude treatment and it was only the 29.5C constant treatment that was different from the other 29.5C treatment s. Frequency affected all parameters except the number of days to oviposition and eggs / pod, causing increased developmental rates and egg pod production. Again, it must be stressed that thes e effects cannot be separated from those that might be caused by differences between trials. However, there is almost certainly some f actor, likely season, which is interacting with the treatments and producing such puzzling resu lts. High frequency treatments were conducted during the late fall and winter months, while lo w frequency and constant temperature treatments

PAGE 107

107 were conducted during the spring and early summer months. If there was a seasonal effect, one might easily conclude that high fr equency treatments would have laid fewer eggs. However, this is not the case and they seemed to have a hi gher fecundity than those females reared in low frequency treatments. The effect of mean temper ature also seems to be more pronounced in low frequency treatments. If temperature affects reproduction equally, rega rdless of season, the results reported here make little sense. Howeve r, it might be possible that temperature does not affect fecundity equally and that optimal or critical temperatures for reproduction change with regard to season. While this has not been obser ved in grasshoppers it has been observed in termites (Hu and Appel 2004). If the optimal temperature for reproduction in S. americana did change with season, and during th e winter months optimal temper atures were lower than those for summer months, it would easily explain w hy the number of egg pods / females is less affected by the 29.5C temperatures in high fre quency treatments. However, there is little evidence proving season caused differences between th e trials and all that can be stated is that either frequency or an unknown trial effect is responsible. Another possibility is that our lack of knowledge about the specifics of S. americana reproduction may be preventing the correct interpretation of the data. While it is know that there are two generations of S. americana a year in Florida and that th e second generation overwinters as an adult, it is not known if there are any differences in development and fecundity between these two generations. Consider that females reared in high frequency treatments during the winter months could be the laborat ory equivalent of the generation that overwinters as adults. It may be possible that females of S. americana surviving the winter lay more eggs than those females of the second generation as programmed response to reduced population size due to

PAGE 108

108 lower survival rates of overwintering adults. A possible mechanism for such a response could be the reduction in optimal temperature as di scussed above or some other unknown factor The 1991 S. americana outbreak in Florida was attribut ed to a 5 yr drought (Capinera 1993a) which caused available sunshine to be very high and possibly increased grasshopper fitness. The data from this study provide some tangible evidence as to how weather conditions can affect grasshopper fitness and why a 5 yr drought might be benefi cial for grasshopper populations in Florida. Drought conditions would be expected to be close to those represented by July 5 or July 8, when grasshopper body te mperature averaged around 38C, while normal conditions might be represented by days like June 29, with mean body temperatures of free roaming grasshoppers around 30C. O bviously there will be advantag es to extended periods with an 8C difference in daytime body temperatures These extended periods of increased body temperature are an excellent example of how periods of drought might benefit grasshopper development and reproduction l eading to outbreak populations. If drought was thought to cause outbreaks of S. americana at the time of the 1991 outbreak, scientists might have tried to use popul ation modeling to estimate the scale of the outbreak. Using a traditional mode ling method, they would have lik ely used the daily mean for ambient temperatures for the period being monitore d. For supposed drought conditions or in this case, July 5 or July 8, mean ambient temperat ure was around 33C for day time temperatures, and after averaging in nighttime temperatures it would be near 29.5C. Development and reproductive estimates would likely be based on data from laboratory studies conducted at constant temperatures, essentially basing the m odel on something similar to the 29.5C constant treatment from this study. However, gra sshoppers on these days would experience body temperatures which would not be best repr esented by the 29.5C c onstant treatment.

PAGE 109

109 Grasshoppers on these days experienced body temper atures much higher than ambient and with few fluctuations, and instead would be best represented by the 31.5 C low frequency low amplitude treatment. If the phenology model was based on mean ambient temperature and used the data obtained from the 29.5C constant treatment in this study, the population would take 93.5 days to reach reproduction and females would produce 1.3 egg pods per female and 40.6 eggs per pod. In contrast, if the model was based on more accurate body temperatures and used the data obtained from the 31.5C low frequency low am plitude treatment, the population would take 63.3 days to reach reproduction and females would produce 5.6 egg pods per female and 66.7 eggs per pod. S. americana has a spring or early summer generation followed by an autumn generation in Florida. In or der for the population based on am bient temperature and the 29.5C constant treatment to remain stable we must a ssume that 2% of all eggs laid during the early summer generation survive to be reproductive female s. For purposes of this discussion winter generations will be disregarded and assumed to maintain a stable population. The phenology model based on mean ambient temperatures and data from this study would predict that after 5 yr of drought each female from the original populati on before the drought began and her successive offspring would have multiplied into 1.16 female s over the 5 yr period, essentially a static population. If a phenology model was based on more accurate body temperatures and the data from the 31.5C low frequency low amplitude trea tment and the same assumptions were applied, each female and her successive offspring woul d multiply into 22,800 females over the 5 year period of drought. Such an extreme scenario is ve ry unlikely, because initial survival rates are likely to be much lower and would continue to decrease because of in creasing population size

PAGE 110

110 and competition. However, it is easy to s ee how phenology models based on mean daily temperatures and laboratory data collected at constant temperatures could under predict a population outbreak. There are several issues that caused problems with this study. First, the study might have tried to examine too many parameters and sacrifi ced a more in depth study of each individual parameter under more treatments and different te mperatures. Another problem with the study is that collection of reproduction data was discon tinued because of pesticide contamination of grasshopper food and lifetime fecund ity could not be assessed. Life time fecundity data may have been able to better show the e ffects of temperature and temperature fluctuations. However, the biggest problem with this study is the separati on of high and low frequency treatments that caused severe problems when analyzing data an d negated the studys ab ility to analyze any effects of frequency on development and reproduc tion. This study has raised more questions that it has answered. Why does amplitude have a grea ter effect on development and reproduction at lower temperatures? An even more puzzling questi on is, why did grasshoppers lay more eggs per pod at lower temperatures in high frequency treat ments but fewer in low frequency treatments? A more in depth focused study of the affects of temperature fluctuation on development and reproduction in S. americana would answer many of these questions. In conclusion, this study has provided novel da ta from the field relating to grasshopper body temperature, and has shown that temperat ure fluctuations (other than scotophase fluctuations) can have an effect on development (a t least at sub-optimal te mperature at or below 29.5C) and reproduction in S. americana and most likely other grasshoppers. Most importantly, this study has shown that, under most conditi ons, grasshopper body temperatures are very different from ambient temperature. While the study provides no exact method for improving

PAGE 111

111 phenology modeling, it does suggest that the mo st common method currently used can be inaccurate, and that use of more relevant mean body temperatures or data from temperature treatments similar to those temperatures experienced in the field would likely improve our ability to predict population sizes of not only grass hoppers, but other insects as well. While using sunlight intensity to estimate body temperatures for modeling woul d be appropriate for basking insects, such as grasshoppers, it is not appropriate for those insects not utilizing sunlight as their primary heat source. Each insects biology shoul d be considered on an individual basis and an appropriate method to best estimate act ual body temperature should be used.

PAGE 112

112 LIST OF REFERENCES Ashamo, M.O., and O.O. Odeyemi. 2000. Effect of rearing temperature on the fecundity and development of Euzopherodes vapidella Mann (Lepidoptera: Pyralid ae), a pest of stored yam. J. Stored Prod. Res.. 37:253-261. Beck, S.D. 1983. Insect thermoperi odism. Annu. Rev. Entomol. 28:91-108. Begon, M. 1983. Grasshopper populations and weat her: the effects of insolation on Chorthippus brunneus Ecol. Entomol. 8:361-370. Berner, D., C. Krner, and W.U. Blancke nhorn. 2004. Grasshopper populati ons across 2000 m of altitude: is there life history adaptation? Ecography 27: 733-740. Blanford, S., and M.B. Thomas. 2000. Thermal behavior of two acridid specie s: effects of habitat and season on body temperature and potential impact on biological control agents. Environ. Entomol. 29:1060-1069. Brillon, S., Y. Lambert, and J. Dodson. 2005. Egg survival, embryonic development, and larval characteristics of northern shrimp ( Pandalus borealis ) females subject to different temperature and feeding condi tions. Marine Biol. 147:895-911. Capinera, J.L. 1993a. Host-plant selection by Schistocerca americana (Orthoptera: Acrididae). Envir on. Entomol. 22:127-133. Capinera, J.L. 1993b. Differentiati on of nymphal instars in Schistocerca americana (Orthoptera: Acrididae). Florida Entomol. 76:175-179. Capinera, J.L., L.F. Wiener, and P.R. Anamosa. 1980. Behavioral thermoregulation by late-instar range caterpillar larvae Hemileuca oliviae Cockerell (Lepidoptera: Saturniidae). J. Kansas Entomol. Soc.. 53:631-638. Carruthers, R. I., T. S. Larkin, H. Fi rstencel, and Z. Feng. 1992. Influence of thermal ecology on the mycosis of a rangeland grasshopper. Ecology 73:190-204. Casey, T. M. 1981. Behavioral mechanisms of thermoregulation, pp. 79-114. In B. Heinrich, Insect Thermoregulation. J ohn Wiley & Sons, New York, New York. Castillo, J., J.A. Jacas, J.E. Pea, B.J. Ulmer, and D.G. Hall. 2006. Effect of temperature on life history of Quadrastichus haitiensis (Hymenoptera: Eulophidae), an endoparasitoid of Diaprepes abbreviatus (Coleoptera: Curculionid ae). Biol. Contr. 36:189-196. Chappell, M. A., and D. W. Whitma n. 1990. Grasshopper thermoregulation, pp. 143172. In R.F. Chapman and A. Joern (eds.) Biology of Grasshoppers. John Wiley & Sons, New York, New York.

PAGE 113

113 Fielding, D.J. 2004. Developmental time of Melanoplus sanguinipes (Orthoptera: Ac rididae) at high latitudes. Environ. Entomol. 33:1513-1522. Forsman, A. 1997. Thermal capacity of different colour morphs in the pygmy grasshopper Tetrix subulata Ann. Zool. Fennici 34: 145-149. Forsman, A. 2000. Some like it hot: intrapopulation variation in behaivoral thermoregulation in color-polymorphi c pygmy grasshoppers. Evol. Ecol. 14:25-38. Gardner, K.T., and D.C. Thompson. 2001. Deve lopment and phenology of the beneficial grasshopper Hesperotettix viridis Southwest. Entomol. 26:305-313. Gndz, N.E., and A. Glel. 2002. Effect of te mperature on development, sexual maturation time, food consumption and body weight of Schistocerca gregaria Forsk (Orthoptera: Acrididae). Turkish J. Zool. 26:223-227. Harrison, J. F., and J. H. Fewell. 1995. Ther mal effects on feeding behavior and net energy intake in a grassh opper experiencing large diurna l fluctuations in body temperature. Physiol. Zool. 68:453-473. Heinrich, B. 1993. Grasshoppers and other Or thoptera, pp. 143-190. In B. Heinrich, HotBlooded Insects. Harvard Universi ty Press, Cambridge, Massachusetts. Heinrich, B. 1996. The Thermal Warriors. Ha rvard University Press, Cambridge, Massachusetts. Hu, X.P. and A.G. Appel. 2004. Seasonal variatio n of critical thermal limits and temperature tolerance in Formosan and Eastern subterra nean termites (Isoptera: Rhinotermitidae). Physiol. Ecol. 33:197-205. Huey, R.B. and R.D. Stevenson. 1979. Integr ating thermal physiology and ecology of ectotherms: a discussion of approaches. Amer. Zool. 19:357-366. Inglis, G. D., D. L. Johnson, and M. S. Goettel. 1996. Effects of temperature and thermoregulation on mycosis by Beauveria bassiana in grasshoppers. Biol. Contr. 7:131-139. Jordaan, A., S.E. Hayhurst, and L.J. Kling. 2006. Th e influence of temperature on the stage at hatch of laboratory reared Gadus morhuaI and implications for comparisons of length and morphology. J. Fish Biol. 86:7-24. Kemp, W. P. 1986. Thermoregulation in three rangeland grasshopper species. Can. Entomol. 18:335-343.

PAGE 114

114 Kemp, W. P., and B. Dennis. 1989. Developm ent of two rangeland grasshoppers at constant temperatures: development thre sholds revisited. Can. Entomol. 121:363371. Khrt, U., J. Samietz, and S. Dorn. 2005. Thermo regulation behavior in codling moth larvae. Physiol. Entomol. 30:54-61. Kuitert, L.C., and R.V. Connin. 1952. Biology of the American grasshopper in the southeastern United Sates. Florida Entomol. 35:22-33. Lactin, D. J., and D. L. Johnson. 1996. Behavi oral optimization of body temperature by nymphal grasshoppers ( Melanoplus sanguinipes, Orthoptera: Acridida e) in temperature gradients established using incandes cent bulbs. J. Therm. Biol. 21:231-238. Lactin, D. J., and D. L. Johnson. 1997. Response of body temperature to solar radiation in restrained nymphal migratory grasshoppers (Orthoptera: Acridida e): influences of orientation and body size. Physiol. Entomol. 22: 131-139. Lactin, D. J., and D. L. Johnson. 1998a. Convective heat loss and change in body temperature of grasshopper and locust nymphs: relative importance of wind speed, insect size, and insect orientati on. J. Therm. Biol. 23:5-13. Lactin, D. J., and D. L. Johnson. 1998b. Envi ronmental, physical, and behavioral determinants of body temperature in gra sshopper nymphs (Orthopt era: Acrididae). Can. Entomol. 130:551-557. Miles, C.I. 1985. The effects of behaviorally re levant temperatures on mechanosensory neurons of the grasshopper, Schistocerca americana. J. Exp. Biol. 116:121-139. Nakahira, N., R. Nakahara, and R. Arakawa. 2005. Effect of temperature on development, survival, and adult body size of two green lacewings, Mallada desjardinsi and Chrysoperla nipponensis (Neuroptera: Chrysopidae). Appl. Entomol. Zool. 40:615-620. Parker, J.R. 1930. Some effects of temperature and moisture upon Melanoplus mexicanus mexicanus Saussure and Cammula pellucida Scudder (Orthoptera). Mont. Agric. Exp. Sta. Bul. 223:1-132. Petavy, G., J.R. David, P. Gibert, and B. Morete au. 2001. Viability and rate of development at different temperatures in Drosophila : a comparison of constant and alternating thermal regimes. J. Therm. Biol. 26:29-39. Pitt, W. C. 1999. Effects of multiple vert ebrate predators on grasshopper habitat selection: trade-offs due to preda tion risk, foraging, and thermoregulation. Evol. Ecol. 13:499-515. Prange, H.D. 1990. Temperature regulation by re spiratory evaporation in grasshoppers. J. Exp. Biol. 154:463-474.

PAGE 115

115 Putnam, L. G. 1963. The progress of nympha l development in pest grasshoppers (Acrididae) of western Cana da. Can. Entomol. 95:1210-1216. Satar, S., U. Kersting, and M.R. Ulusoy. 2005. Te mperature dependent lif e history traits of Brevicoryne brassicae (L.) (Hom., Aphididae) on white cabbage. Turk. J. Agric. For. 29:341-346. Squitier, J.M., and J.L. Capinera. 2002a. Ob servations on the phenol ogy of common Florida grasshoppers (Orthoptera: Acridi dae). Florida Entomol. 85:227-234. Squitier, J.M., and J.L. Capinera. 2002b. Habitat associations of Florida grasshoppers (Orthoptera: Acrididae). Florida Entomol. 85:235-244. Uvarov, Sir B. 1966a. Grasshoppers and Locu sts: A Handbook of General Acridology, Vol. 1. Cambridge University Press, London, England. Uvarov, Sir B. 1966b. Grasshoppers and Locu sts: A Handbook of General Acridology, Vol. 2. Cambridge University Press, London, England. Whitman, D.W. 1986. Developmental thermal requirements for the grasshopper Taeniopoda eques (Orthoptera: Acrididae). A nn. Entomol. Soc. Am. 79:711-714. Whitman, D. W. 1988. Function and evolution of thermoregulation in the desert grasshopper Taeniopoda eques J. Anim. Ecol. 57:369-383. Willott, S. J. 1997. Thermoregulation in four species of British grasshoppers (Orthoptera: Acrididae). Funct. Ecol. 11:705-713. Woodson, W.D., and J.V. Edelson. 1988. Developmenta l rate as a function of temperature in a carrot weevil, Listronotus texanus (Coleoptera: Curculionida e). Ann. Entomol. Soc. Am. 81:252-254.

PAGE 116

116 BIOGRAPHICAL SKETCH Jason G. Froeba was born on August 20, 1982 in Metairie, Louisiana, where he was also raised with his younger brother. He graduated from Archbis hop Rummel High School in 1999. After high school, he attended the University of New Orleans and graduated with a B.S. in biological sciences in 2003. Hi s undergraduate coursework took him to Costa Rica for 5 weeks of Spanish and Environmental study and leadership. This is where his love for insects first took root. After college he spent a year working fo r Dial One Franklynn Pest control in Metairie, Louisiana. Jason moved to Gainesville, Florida in 2004 to pursue his M.S. in entomology. Jason is currently living in New Roads, Louisiana (n ear Baton Rouge). He currently works for Louisianas Department of Wildlife and Fisherie s, performing data analysis for the Marine Fisheries division. Upon completion of his M.S. degree, Jason will continue to work for Louisianas Department of Wildlife and Fisheries. Jason has been married to Emily Froeba for 3 years. They have a daughter Annemarie, ag e 1, and are expecting a second daughter this summer.


Permanent Link: http://ufdc.ufl.edu/UFE0020960/00001

Material Information

Title: Effects of Weather and Behavior on Body Temperature and the Consequences of Temperature Fluctuations on Development and Reproduction in Schistocerca americana: Implications for Phenology and Population Modeling
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0020960:00001

Permanent Link: http://ufdc.ufl.edu/UFE0020960/00001

Material Information

Title: Effects of Weather and Behavior on Body Temperature and the Consequences of Temperature Fluctuations on Development and Reproduction in Schistocerca americana: Implications for Phenology and Population Modeling
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0020960:00001


This item has the following downloads:


Full Text





EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE AND THE
CONSEQUENCES OF TEMPERATURE FLUCTUATIONS ON DEVELOPMENT AND
REPRODUCTION IN Schistocerca amnericana: IMPLICATIONS FOR PHENOLOGY AND
POPULATION MODELING













BY

JASON G. FROEBA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007



































O 2007 Jason Froeba




























To my Aunt Linda, who always nurtured my scientific interests, and to my wonderful wife
Emily, without whom this accomplishment would not have been possible









ACKNOWLEDGMENTS

I thank my supervisory committee chair, Dr. John L. Capinera, for his advice and

support, his firm motivational style, and for always keeping my best interest in mind. I thank my

other committee members, Dr. Daniel Hahn and Dr. Heather McAuslane, for their mentoring and

constructive criticism, which greatly improved this work. I would like to thank Seth Bybee and

Jennifer Zaspel for their help and anyone else who provided input, support, or materials. Finally,

I would like to thank my wife Emily for her support and encouragement.












TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES ............ ....._._. ...............7.....


LIST OF FIGURES .............. ...............8.....


AB S TRAC T ......_ ................. ............_........9


CHAPTER


1 INTRODUCTION AND LITERATURE REVIEW ................. .............. ......... .....11


2 EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE ....................30


M ethods and M materials .............. ...............30....

Study Site................... ... ... ..... ........3
Measurement of Body Temperature ................. ...............30................
Behavioral Observations .............. ...............33....
Environmental Conditions .............. ...............34....
Statistical Analysis............... ...............34
R e sults............... ... ... .. ........ .. ........... .............3
Measurement of Body Temperature............... ..............3
Environmental and Behavioral Ob servations. .................. ...............36........... ...

Temperature Fluctuations and Adverse Weather Conditions .............. ....................38
D discussion ................. .. ...... ....... ...... ...... .............4
Measurement of Body Temperature............... ..............4
Environmental and Behavioral Ob servations. .................. ...............43........... ...

Temperature Fluctuations and Adverse Weather Conditions .............. ....................47

3 EFFECTS OF TEMPERATURE FLUCTUATIONS ON DEVELOPMENT AND
REPRODUCTION .............. ...............69....


Methods and Materials .............. ...............69....

Temperature Treatments .............. ...............69....
Cages ................. .......... ... ...............70.....
Nymphal Development Time ................. ...............70........... ....
Body Size and Reproduction............... ..............7
Statistical Analysis............... ...............71
Re sults ................. ......... ... ...............72.....

Nymphal Development Time ................. ...............72........... ....
Body Size ................. ...............74.................
Days to Oviposition .............. ...............74....
Egg Pods / Female............... ...............75.
Eg gs/ P od .............. ...............76....












Discussion ................ ... ..... ....... ........ ........ ... ..... ..... ..........7
Conformity of Laboratory Treatments to Field Study Results ................. ............... .....77

Nymphal development time .............. ...............78....
Body size ................. ...............83.................
Days to Ovioposition .............. ...............84....
Egg Pods / Female. ........._..._.._ ...............86....._._ ....
Eggs / Pod ........._..._.._ ...._._. ...............87....

4 CONCLUSIONS .............. ...............102....


LIST OF REFERENCES ........._..._.._ ...._._. ...............112....


BIOGRAPHICAL SKETCH ........._..._.._ ...............116._.._._ ......











LIST OF TABLES


Table Page

2-1 Summary of mean body temperatures, behavior, and environmental data ................... ....49

2-2 Regression statistics for grasshopper body temperature and environmental
param eters. .............. ...............52....

2-3 Rates of temperature decrease due to adverse weather conditions. ................ ................53

3-1 Nymphal development time: ANOVA results ................. ...............90...............

3-2 Mean comparisons of nymphal development time. ................ .............................91

3-3 Mean comparisons of nymphal development time for low and high frequency
treatments combined. ............. ...............92.....

3-4 Body size: ANOVA results............... ...............93

3-5 Reproductive data: ANOVA results .............. ...............94....

3-6 Mean comparisons of reproductive data. .............. ...............95....











LIST OF FIGURES


Figure page

2-1 Wiring method used for recording internal body temperature............... ..............5

2-2 M ay 19 2005. ............. ...............55.....

2-3 June 14 2005 .............. ...............56....

2-4 June 17 2005 .............. ...............57....

2-5 June 22 2005 .............. ...............58....

2-6 June 29 2005 .............. ...............59....

2-7 July 5 2005 ............ _...... ._ ...............60...

2-8 July 8 2005 ............ _...... ._ ...............61...

2-9 November 17 2005............... ...............62..

2-10 Variation in the percent temperature difference between sun constrained
grasshoppers and ambient temperature in relation to sunlight intensity ..........................63

2-11 Regression analysis of sun constrained grasshoppers............... ..............6

2-12 Regression analysis of shade constrained grasshoppers ....._____ .......___ ..............65

2-13 Regression analysis of behavioral data against percent body temperature differences.....66

2-14 Regression of behavioral orientation rating against ambient temperature ........................67

2-15 Regression analysis of free roaming grasshoppers .............. ...............68....

3-1 Fluctuating laboratory temperature treatments .............. ...............96....

3-2 Nymphal development time of high frequency treatments.........._..._.._ ..........._..........97

3-3 Nymphal development time of low frequency treatments ................. .......................98

3-4 Nymphal development time at fixed levels of frequency ............. .....................9

3-5 Mean femur and overall length for high and low frequency treatments ........................100

3-6 Mean number of days to first oviposition, egg pods / female, and eggs / pod for high
and low frequency treatments ............_...... .__ ...............101..









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE AND THE
CONSEQUENCES OF TEMPERATURE FLUCTUATIONS ON DEVELOPMENT AND
REPRODUCTION IN Schistocerca amnericana: IMPLICATIONS FOR PHENOLOGY AND
POPULATION MODELING

By

Jason Froeba

May 2007

Chair: John Capinera
Maj or: Entomology and Nematology

Phenology modeling is central to many pest management strategies and often consists of

modeling temperatures by using ambient temperatures and insect developmental data derived

from constant temperature studies. Not surprisingly, discrepancies between predictions and the

actual timing and size of insect populations can often be attributed to differences between actual

body temperature and the temperature used in the model. In addition, temperature fluctuations

are a potentially important, yet largely ignored area in insect thermal biology and modeling, as

most models are based on data collected at constant temperature. Grasshoppers such as

Schistocerca amnericana are well known for maintaining body temperatures different from

ambient temperature. This, plus the fact that it is one of the few grasshoppers in the eastern USA

that reach damaging levels, make S. amnericana a relevant subj ect on which to assess the effects

of thermoregulation on development and reproduction.

Body temperatures of grasshoppers, along with several environmental parameters, were

monitored continuously to ascertain the effects of weather and behavior, and to determine what

body temperatures grasshoppers could attain and would prefer in the field. Body temperatures

averaged about 30 and 380C during the daylight hours for grasshoppers on overcast and sunny










summer days, respectively, compared to average ambient temperatures of 28 and 320 for

overcast and sunny conditions. Daytime temperatures fluctuated frequently, especially on

overcast days. There were significant positive relationships between body temperature in S.

amnericana and sunlight intensity, ambient temperature and behavior. The rate of increase in

body temperature due to direct solar radiation was as high as 2.50C /min, which greatly elevated

body temperatures, up to 19.60C above ambient. Adverse weather conditions (cloud cover and

rainfall) reduced mean daily body temperatures by up to 80C and caused up to a 38% decrease in

body temperature at a mean rate of 0.9780C/minute and 0.552oC/minute for cloud cover and

rainfall, respectively.

Laboratory studies were conducted at constant temperatures and fluctuating temperatures

of two amplitudes and frequencies for two mean temperatures of 29.50C and 3 1.50C for a total of

ten treatments. A mean daily temperature increase of 20C caused a decrease in mean

development time (up to about 6 d) and increase in mean reproductive output (up to 175 eggs /

female increase). Amplitude of temperature fluctuation was also shown to affect development

rates and fecundity, but more so in the 29.50C treatments than in the 31.50C treatments. The

effects of frequency of temperature change were significant for all parameters tested except days

to oviposition, but could not be separated from effects which may have been caused by the

separation of treatments into two time periods. Overall, temperature fluctuations significantly

affect development and reproduction in S. amnericana. The results suggest that grasshopper body

temperatures are often very different from ambient temperatures, that phenology models based

on ambient temperatures can be inaccurate, and that use of more relevant mean body

temperatures should be considered for species that thermoregulate or attain body temperatures

different than ambient.









CHAPTER 1
INTTRODUCTION

Temperature plays an important role in the life processes of many organisms. This is

especially true in poikilothermic ectotherms, which rely on outside sources of heat to maintain

internal body temperature. Temperature affects the rate at which biochemical processes take

place, and can affect metabolism, developmental rate, reproduction, and body size. The

developmental stage of Atlantic cod, Gadus morhua, at hatch is negatively affected by

suboptimal temperatures, with development being incomplete at both low and high temperatures

(Jordaan et al. 2006). Incubation time in northern shrimp, Pand'alus borealis, was reduced from

214 days at 20C to 123 days at 80C (Brillon et al. 2005). Insects are particularly sensitive to

changes in temperature. Listronotus texanus Stockton (Coleoptera: Curculionidae) exhibited a

decrease in total development time of 35 days from 65 to 30 days when the temperature was

raised from 200C to 27.50C (Woodson and Edelson 1988). Euzopherodes vapid'ella Mann

(Lepidoptera: Pyralidae) showed an increase in mean fecundity from 51.8 eggs per female at

200C to 124.4 eggs per female at 330C. However, wing span and body length was significantly

less at 330C than at lower temperatures tested (Ashamo and Odeyemi 2000). As advantageous as

temperature increases seem to be, there is a limit to the extent temperature increases can be

beneficial.

Extremes in temperature can have severe negative effects on an organism (e.g.

desiccation and deactivation of enzymes) causing thermal death. The temperatures at which

severe negative effects occur define the tolerance range of an organism, which estimates the

range of temperatures over which survival is possible (Huey and Stevenson 1979). While

temperatures outside of the tolerance range lead to death, the effects of less extreme suboptimal

temperatures are usually more subtle (e.g., prolonged developmental time and diapause, and









decreased fecundity and body size). In Mallada111~~~111~~~111~~ desjardinsi (Navas) and Chrysoperla

nipponensis (Okamoto) (Neuroptera: Chrysopidae) body size was reduced at both low and high

suboptimal temperatures (Nakahira et al. 2005). Castillo et al. (2006) showed that developmental

time in Quadra~stichus haitiensis (Gahan) (Hymenoptera: Eulophidae), decreased with increasing

temperature up to a certain point, after which increases in temperature caused increases in

developmental time. This increase in developmental time at suboptimal temperatures is very

common in insects, and the temperatures at which the decrease occurs are often called critical

thermal temperatures.

There are both critical thermal maximum and critical thermal minimum temperatures, and

they are often used to define ecological and behavioral thermal limits (Hu and Appel 2004). The

critical thermal minimum and maximum temperatures help define what is known as the thermal

performance breadth of an organism, which estimates the range of temperatures over which an

organism performs well (Huey and Stevenson 1979). When compared to larger organisms,

insects exhibit a broad thermal performance breadth. This may be because insects will encounter

the same thermal environments as many other organisms but, due to their small size, body

temperature will be more sensitive to small changes in the thermal environment (Heinrich 1993).

In addition to having a broad thermal performance breadth, some insects (e.g., termites) have

been shown to have critical temperatures that fluctuate seasonally (Hu and Appel 2004). Of even

more importance than the thermal performance breadth is the thermal optimum or optimal

temperature at which an insect will experience the highest fitness level possible (Huey and

Stevenson 1979), usually by a combination of increased developmental and reproductive rates.

However, environmental temperatures are often not those that would be considered optimal for

insects, and so body temperature must be controlled in order to optimize fitness.









All insects are poikilothermic and can be ectothermic (relying on external sources of

heat), endothermic (able to use heat generated by muscle activity), or both. In either case, insects

must thermoregulate to maintain functional body temperatures when in a suboptimal thermal

environment. Thermoregulation is the maintenance of an optimal internal body temperature

different from environmental temperatures. Thermoregulation not only includes raising body

temperature above ambient, but also systematically decreasing body temperature in cases where

ambient temperatures become too high. While almost all insects possess some ability to avoid

undesired temperatures (Casey 1981), many insects can maintain body temperatures within an

optimal range by actively altering specific behaviors.

Insect orders in which behavioral thermoregulation has been observed include

Lepidoptera, Odonata, Orthoptera, Coleoptera, Hemiptera, Hymenoptera, and Diptera (Heinrich

1993). Methods of thermoregulation in insects are variable and differ significantly between

orders. Hymenoptera make use of heat produced by the shivering of flight muscles. Honeybees

and bumblebees have been known to use this method to keep nests warm (Heinrich 1996).

Among Lepidoptera, Hemileuca oliviae Cockerell (Lepidoptera: Saturniidae) larvae ascend

vegetation and rest during the heat of the day to stay cool. They shift position so that the ventral

surface of the body is always facing the sun and covered by vegetation (Capinera et al. 1980).

Codling moth larvae, Cydia pomonella (L.) (Lepidoptera: Torticidae), exhibit thermoregulatory

behavior by selecting favorable microhabitats within apples (Kiihrt et al. 2005). Kiihrt showed

that 74% of larvae built larger cavities in the warmer hemispheres of apples closest to the heat

source. Kiihrt also found that, unlike the previous larval stages, when larvae sought to cocoon

there was no difference in distribution of the mature larvae between the temperature gradient and

the control, suggesting little if any preference for temperature at this stage. He concluded that at









least in the codling moth, behavioral thermoregulation is life stage dependent, and that larvae

change their thermoregulation behavior during development based on benefits, needs, and

constraints (Kiihrt et al.2005).

One group in which behavioral thermoregulation has been extensively studied is

Orthoptera, specifically Acrididae, due to their great economic importance. Some Orthoptera,

such as katydids, have been observed shivering flight muscles to raise muscle temperature before

singing, similar to muscle twitching in Hymenoptera (Heinrich 1993). However, Acrididae do

not use their wing muscles to sing, and so far no such warming mechanism has been observed in

the group. Consequently, most Acrididae rely solely on external sources of heat obtained

behaviorally (Casey 1981), primarily radiant heat from the sun, also known as insolation. The

movement into or out of sunlight is the most common form of behavioral thermoregulation in

Acrididae (Lactin & Johnson 1998b). Moving into and remaining in a radiant heat source,

usually direct sunlight, as a way to increase body temperature is known as basking. While there

are other sources of heat, such as radiant heat from the ground and small amounts of metabolic

heat, radiant energy acquired by basking is the most important heat source for any acridid

(Uvarov 1966a). In one study conducted on the genus Locusta, body temperatures rose from

27.7oC to 36oC at a rate of 0.83oC per minute for ten minutes, thereafter rising at a rate of

0. 104oC per minute until reaching 42.7oC, when exposed to radiant heat (Uvarov 1966a). While

metabolic heat may have played a role in causing basal body temperature to be different from

ambient temperature, it is very unlikely to raise body temperature by 0.830C per minute,

especially in a resting grasshopper. Lactin and Johnson (1996) found that nymphs of2~elanoplus

sanguinipes (Fabricius) oriented themselves around artificial heat sources and reached average

body temperatures (near 400C) that were different than those expected if the grasshoppers were









randomly dispersed. They considered this conclusive evidence that M. sanguinipes was actively

thermoregulating. In a later study, Lactin and Johnson (1998b), showed grasshoppers

thermoregulate not only in the laboratory, but also in the field. They estimated the body

temperature of free ranging grasshoppers using a series of complex equations that accounted for

different environmental factors affecting body temperature. They also placed obj ects with

thermodynamic properties similar to those of grasshoppers at random in the field and recorded

the temperature of these obj ects. The estimated body temperatures of the free ranging

grasshoppers differed significantly from those of the obj ects that were placed at random. This

was one of the first field studies that provided strong evidence that grasshoppers actively

thermoregulate in the field by choosing warmer locations and altering body posture.

When trying to understand the thermal biology of an organism it is not only essential to

know if the organism possesses the ability to thermoregulate, but also what other factors might

affect that ability and body temperature. Fortunately, an extensive amount of work has been

conducted for the purpose of understanding the individual mechanisms of heat transfer that affect

a grasshopper' s body temperature and ability to thermoregulate. However, pathways of heat

transfer are very complex and largely independent of each other, making them very hard to

quantify. Some factors affecting these pathways include insect orientation, body texture, color,

and size, radiant intensity, substrate conductivity, wind speed, and turbulence (Chappell and

Whitman 1990).

While basking, acridids will adopt different orientations with relation to the sun's rays

(Uvarov1966b). There are two basic orientations that a grasshopper will adopt, perpendicular or

parallel. In perpendicular basking, the long axis of the body is oriented perpendicular to the sun' s

rays (Uvarov 1966b). While in this position, grasshoppers will often assume a flanking position,









in which the hind leg shaded by the abdomen is raised, and the other is lowered to prevent

shading of the abdomen (Heinrich 1993). This exposes the largest amount of surface area

possible for heating. Parallel posture is one in which the long axis of the body is parallel to the

sun's rays. In most acridid species the parallel posture only exposes about one-fifth the surface

area that perpendicular flanking basking does (Uvarov 1966b).

Body size and color are two other factors that have the potential to affect body

temperature. Rates of temperature increase and final equilibrium temperature (the point at which

temperature no longer increases as a result of a fixed amount of radiant energy) were examined

in 1st and 5th instar stages of Schistocerca gregaria (Forskal) (Uvarov 1966a). First instars had a

higher rate of temperature increase but lower equilibrium temperatures than 5th instars. In

contrast, Willot (1997), while investigating thermoregulation in four species of grasshoppers,

found there to be no significant temperature differences between males and females within

species. Together, the two studies suggest that while large size differences between 1st and 5th

instar nymphs and possibly species affect temperature, the smaller size differences in adults

between males and females has little if any effect on body temperature. Indeed, Lactin and

Johnson (1997) found only a 20C difference in body temperature between a body mass of 0.03g

and 0.30g, a ten-fold difference in weight. Forsman (1997) conducted a study to determine if

differences in color and size would affect body temperature in Tetrix subulata (Linnaeus)

(Orthoptera: Tetrigidae). He found that body size had no effect on body temperature. However,

black individuals achieved higher body temperatures than brown or white individuals. He then

conducted a study to see if these differences in body temperature affected reproductive

performance in females of the T. subulata. Forsman (2000) again found significant variation in

body temperature due to color morphs in females, but no difference in overall performance. He










concluded that even though darker color morphs achieved higher temperatures, there were

possible physiological differences between morphs that caused them to required higher optimum

temperatures. Similarly, in their study on H. oliviae, Capinera et al. (1980) found that artificially

blackened larvae attained higher body temperature than normal larvae, however, performance

implications were not evaluated.

There are several factors that can potentially lower a grasshopper's body temperature,

including convective and evaporative cooling, and long wave radiation, which refers to the fact

that all obj ects lose heat by emission of energy in the form of infrared radiation (Uvarov 1966a).

S. gregaria will climb onto vegetation, increasing convective cooling when ambient temperature

rises to 40oC (Heinrich 1993). Some grasshoppers can extend their legs to raise the body above

the ground, a behavioral mechanism called stilting, which helps the grasshopper to escape high

surface temperatures while increasing convective cooling (Heinrich 1993). There have been

several studies concerned with the effects of convection and evaporative cooling on body

temperature in grasshoppers. Lactin and Johnson (1998a) showed that orientation to the wind had

no effect on body temperature in M~ sanguinipes; however, wind speed did, and as wind speed

increased, body temperature decreased. In a study on the caterpillar H. oliviae, body temperature

was shown to be inversely related to air speed, but body temperature was always above ambient

temperature, even at air speeds of 4m/s (Capinera et al. 1980). This suggests that while wind

cannot prevent temperature increase, it can reduce the maximum attainable temperature and

possibly the rate of increase. Evaporative cooling can become a factor in low humidity

environments, where evaporation rates are high. This can be seen in an experiment in which

nymphs of S. gregaria were held at different humidity levels, and the rate at which thermal

equilibrium occurred was measured (Uvarov 1966a). Nymphs held in high humidity









environments reached equilibrium more quickly than nymphs held in low humidity environments

(Uvarov 1966a). The higher humidity reduced heat loss through evaporation allowing the nymph

to retain more heat and attain the equilibrium temperature faster. Grasshoppers have also been

known to "pant," increasing breathing rate and increasing evaporative cooling, and therefore

lowering body temperature (Casey 1981, Heinrich 1993). Prange (1990) conducted a study on

the effect of respiratory rate on evaporative cooling in three species of grasshoppers Schistocerca

nitens (Thtmnberg), Locusta migratoria (L.), and Tmethis pulchripennis (Bolivar). He showed that

rate of evaporation and ventilation frequency remained relatively constant up to about 450C, after

which both rates increased significantly. Grasshoppers were able to maintain temperatures below

the lethal limit of 480C in air temperature as high as 52-530C. The maintenance of such

temperatures caused water loss in the grasshoppers to reach 8% of their body mass per hour.

Grasshoppers could tolerate a water loss of 33% of their mass, meaning they could maintain sub-

lethal temperatures for about 4 hours. With the exception of deserts, there are few places where

ambient temperature can rise to such lethal levels requiring such an extreme form of behavioral

thermoregulation.

By far the most determining factor of body temperature in grasshoppers is radiant energy

from the sun and is the main reason body color and orientation can affect body temperature. It is

common knowledge that obj ects heat up when exposed to direct sunlight, but more specific

parameters such as maximum attainable temperature and rate of temperature increase differ

depending on the nature of the obj ect. There is also diffuse solar radiation from the sun that

reflects off surrounding obj ects which could possibly affect temperature. Lactin and Johnson

(1997) conducted a study on M~ sanguinipes to determine the effects of direct and diffuse solar

radiation on body temperature. They found that direct solar radiation had a highly significant









effect on body temperature while diffuse solar radiation did not. It is possible that diffuse solar

radiation does not have enough energy to heat obj ects, or perhaps body temperature did not differ

from ambient because diffuse solar radiation was also warming the surrounding environment.

Either way, diffuse solar radiation likely plays little if any part in maintaining body temperatures

different from ambient. In simulations based on this work, Lactin and Johnson found that direct

solar radiation can cause a temperature increase of 0.0080C / W/m2. In another study, Fielding

(2004) placed freshly killed M~ sanguinipes grasshoppers perpendicular to the sun at ground

level. A thermocouple was inserted into the sternum to obtain temperatures under direct

insolation. Fielding found that body temperature could be raised 15-200C above ambient

temperature near the ground when exposed to direct insolation. He also compared predicted

developmental times using ambient temperature and solar adjusted temperatures and showed the

time spent as nymph could be cut in half. While sunlight is the most significant source of

temperature elevation in grasshoppers, it is not always available due to adverse weather

conditions. Because the sun is the most important factor for raising body temperature, adverse

weather conditions or the absence of the sun, is the most important factor hindering the ability to

raise body temperature. In environments such as mountain ranges where ambient temperatures

are low and the warmer seasons are short, cloudy days can prevent grasshoppers from obtaining

the necessary amount of heat to complete development (Berner et al. 2004).

Aside from behavioral responses and abiotic environmental factors, there are ecological

factors which can also play a role in a grasshopper' s ability to thermoregulate. When considering

an insect' s ability to thermoregulate, one must also consider the biotic environment in which the

insect lives. The vegetation surrounding the grasshopper can dramatically change how each of

the abiotic factors mentioned earlier will affect the grasshopper. Vegetation can provide shelter










from sunlight and wind, while at the same time providing a perch that a grasshopper may use to

increase exposure to either element. Vegetation can also produce different thermal environments

with varying temperature and humidity levels. Willot (1997) studied the temperature differences

between swards (large expanses of grass covered soil) of different height. His results showed that

shorter swards reach higher temperatures during the day than taller swards. Differences in food

quality of vegetation in the environment will also play a role in where a grasshopper will spend

its time (Pitt 1999). Along with the surrounding vegetation, other animals can also influence a

grasshopper's ability to thermoregulate. Predators can drive grasshoppers away from optimal

conditions and into suboptimal temperatures and areas of reduced food quality. In a study by Pitt

(1999) on M~ femurrubrum (DeGeer), avian predators were shown to drive grasshoppers down

into vegetation away from high quality food and sunlight into cooler temperatures.

It may seem like grasshoppers go through a large amount of trouble to maintain optimal

temperatures, but there is merit in their effort. Temperature can have profound effects on

physiological processes in acridids and in insects as a whole. Most chemical reactions occur at

faster rates when subj ected to higher temperatures. This generalization can be applied to most

processes in the biological world, and is especially important to poikilothermic organisms and

specifically in grasshoppers. Some processes that benefit from higher temperatures include

speed of muscle contraction, food consumption, metabolism, immune response, development,

and reproduction. Grasshoppers usually need to maintain body temperatures above ambient in

order to optimize these processes.

The speed at which muscle contractions occur is highly dependent on temperature. This

has profound effects on minimum temperatures at which any insect moves, and more

importantly, flies. This dependence on a minimum temperature is so important that some insects









can even shiver their flight muscles to warm up to minimum flight temperatures (Heinrich 1996).

In S. gregaria, the minimum temperature required for flight is around 20oC, while optimum is

around 35oC (Uvarov 1966b). Feeding rates are also positively correlated with temperature; both

chemical digestion and the muscle contractions of the digestive tract are temperature dependent.

Whitman (1988) conducted a study on thermoregulation in Taeniopoda eques (Burmeister), part

of which was measuring feeding rates at different temperatures. The study showed that as

temperature increased, rates of feeding and defecation increased. Harrison and Fewell (1995), in

a similar study on M~ bivittatus (Say), showed that there were minimum thresholds for feeding,

10oC for lab studies and 25oC for field studies. They also showed there to be a maximum rate of

consumption in the laboratory beyond which feeding does not increase even with an increase in

temperature. This suggests an upper feeding threshold at which rate of consumption would

exceed the maximum rate of digestion. In addition, Harrison and Fewell (1995) also showed that

net energy intake increased dramatically from 0.008W at 150C to 0.3 8W at 3 50C. This was also

accompanied by an increase in metabolic rate.

Giindiiz and Giilel (2002) conducted a study on S. gregaria investigating food

consumption and body weight in relation to two different temperatures, 250C and 300C. They

found that food consumption increases with each stage and then begins to decline after the end of

the first week of adult life, and that during this time consumption increased with an increase in

temperature. After the first week of adult life, temperature increases resulted in decreases in food

consumption. Giindiiz and Giilel reported high weight gain from first instar to the first week of

adult life in high temperature treatments relative to low temperature treatments. However,

average weight at the end of the experiment was not significantly different between treatments.

They concluded that the high weight gain was due to increased food consumption rates, that









there was some critical weight which was needed for reproduction to begin, and that temperature

does not affect the critical weight needed but how fast it is attained. In M~ sanguinipes, body

weight was lower at the low (210C and 240C) and high (390C and 420C) extremes of the

temperatures tested relative to intermediate temperatures (Fielding 2004). It is possible that body

weight was affected in this study and not the other because this study used sub-optimal

temperatures.

Temperature also plays an important role in a grasshopper' s ability to fight infection.

Several studies have investigated levels of mycosis at different temperatures in grasshoppers. In

M. sanguinipes, continuous exposure to high temperatures was detrimental to the mycosis of

Beauveria ba~ssiana. Nymphs that were inoculated with B. ba~ssiana and then placed on a heat

gradient remained in warmer areas of the gradient (Inglis et al. 1996). Carruthers (1992)

performed a study on the mycosis ofEntomophaga grylli in the clear winged grasshopper,

Camnnula pellucida (Scudder). The study showed that exposure to temperatures of 380C-40oC for

more than four hours each day was detrimental to survival of E. grylli. The ability of high

temperatures to fight infection gives grasshoppers one more reason to maintain body

temperatures well above ambient.

The effects of temperature on development and reproduction in grasshoppers can be

substantial and have been well documented. A study by Begon (1983) showed that when 4th

instars of Chorthippus brunneus (Thunberg) were exposed to a radiant heat source,

developmental rate could be 5.6 times greater than when the grasshoppers were not exposed. The

same study also showed that adult females held in cages with longer periods of radiant heat laid

more egg pods than females in cages with shorter periods of radiant heat. Putnam (1963)

conducted a study that shows development time in three species of grasshopper ranging from 53










days at 24oC, to 17 days at 3 80C. In another study by Willot (1992) on four species of Acrididae,

development and reproduction were either zero or very low at temperatures below 25oC, while

the optimum temperature for growth and development was between 35oC and 40oC. In S.

gregaria, nymphal development was 10 days faster and sexual maturation 19 days earlier at

300C when compared to 250C (Giindiiz and Giilel 2002). Taeniopoda eques required 60 days

from nymph to adult at 250C and only 35 days at 300C (Whitman 1986). Parker (1930) showed

an increase from 270C to 320C reduced length of larval development by 27 days in M~ mexicanus

Saussure.

While these studies show that temperature increases as small as 40C can cut development

time in half, it is likely that the more important parameter is how close the temperature

approximates the optimal temperature. An Arizona population of Hesperotettix viridis (Thomas)

did not develop at 150C and 200C and developed slowly at 230C. At higher temperatures of

300C, 350C, and 400C, development was the same, approximately 40 days (Gardner and

Thompson 2001). Here, small increases of 50C did not cause an increase in developmental time.

Almost all studies show that increases in temperature can increase development, but the amount

of increase and actual benefits associated with these increases depend highly on the thermal

biology of the organism in question.

Temperature increases development by increasing biological processes in general, but

this can also have a negative side effect. Increased metabolic rates tend to shorten an organism's

life span. In one study conducted on C. pellucida, adults survived for 16 days at 37oC and 32.6

days at 27oC (Uvarov 1966a). At first glance one might conclude that those grasshoppers living

at 27oC have a higher fitness. However, at 27oC females laid an average of only one egg pod, as

opposed to the average of four egg pods laid at 37oC (Uvarov 1966a). Even though the longevity









of the grasshopper had been reduced, its fecundity had quadrupled. These results were similar to

those of Begon (1983).

Overall, basking can help a grasshopper maintain an optimal body temperature and

increase development rate and egg production. Increases in development rate are beneficial to

most grasshoppers, as adult grasshoppers have fewer natural enemies than nymphs, and the faster

one reaches adulthood, the sooner one can leave behind certain risk factors (Kemp 1986).

Increases in developmental rate due to temperature are also essential to the survival of some

species of grasshoppers that live in cooler climates which could not complete their life cycle in a

single season if it were not for the ability to bask and increase their developmental rate (Heinrich

1996). In a study on T. eques, Whitman (1988) found that this species required 850 degree-days

to complete its life cycle, while the air temperature of the environment only provided 692

degree-days. The deficit in the supply of heat was made up by the grasshopper' s ability to

thermoregulate.

Most of the studies mentioned above were conducted at constant temperatures. While this

simplifies the execution and analysis of experiments, it fails to mimic natural conditions. In

nature, terrestrial organisms are submitted to daily fluctuations in temperature, often exceeding a

100C difference (Petavy et al. 2001). Daily temperature fluctuation, which is most commonly

thought of as a day / night cycle, is referred to as thermoperiod. Beck (1983) conducted a review

on thermoperiod in insects and found that thermoperiod is known to affect development,

fecundity, circadian rhythms, diapause, and biological clocks in insects. The effects of

thermoperiod can differ between different insects; under cyclic temperatures some species will

develop more rapidly, others will show no difference, and even a few will develop more slowly

(Beck 1983). For example, the European corn borer, Ostrinia nubilalis Hiibner (Lepidoptera,










Pyralidae), exhibited no difference in development time under thermoperiodic conditions when

compared to constant temperatures, but did exhibit larger late instar larvae under thermoperiodic

conditions (Beck 1983). On the other hand the pitch plant mosquito, Wyeomyia smithii

(Coquillett), developed more slowly under thermoperiodic conditions. However, larvae produced

larger, heavier, and more fecund mosquitoes when compared to larvae reared under constant

conditions (Beck 1983). While the benefits of thermoperiodic conditions in these two species

may not be readily apparent in developmental data, their increased fitness becomes apparent in

reproductive aspects of their life history. On the other hand, the benefits of thermoperiodic

conditions may be blatantly obvious. Satar et al. (2005) tested the effects of thermoperiod on the

aphid Brevicoryne brassicae (L.). At alternating temperatures of 250C and 300C, developmental

time decreased, mortality decreased, longevity increased, and reproduction increased when

compared to those reared at a constant 300C even though they had received less overall heat.

Many studies regarding thermoperiod in insects have yielded ambiguous results and those

that have produced clear results often conflict with results from other studies (Beck 1983).

Studies dealing with thermoperiod often compare fluctuating treatments against a constant

temperature at the average or midpoint temperature of the alternating treatment. The premise for

this is that if development is the same under both alternating temperature and the midpoint

temperature, then the relationship between development time and temperature can be assumed to

be linear. However, the relationship is usually not linear and there is often a deceleration or

acceleration of development rate which is called the Kaufmann effect (Petavy et al. 2001). In

such situations, midpoint temperatures are not suitable for predicting development times under

alternating conditions. An alternative must be used and is called the equivalent development

temperature (EDT), which is the constant temperature that provides the same developmental time









as that observed under a given alternating temperature (Petavy et al. 2001). Petavy et al. (2001)

conducted a study on relatively simple thermoperiods in Drosophila. They consisted of two 12 hr

phases (day and night) and total of 14 treatments with mid range temperatures from 100C to

270C and amplitudes of 60C to 220C. Overall, there was a decrease in development time with

increasing temperature. Mortality was 100% at alternating temperatures of 40C/260C, 90C/330C,

and 210C/340C, and temperatures above 280C caused increases in developmental time. As

expected, mortality was caused by extremes in temperature, while suboptimal temperatures

caused less severe deleterious effects. They determined that the relationship between

development time and temperature is a positive function of amplitude and a negative function of

midpoint. The study showed that development time under alternating temperatures can be up to

20% superior or inferior to the expected development time at the midpoint. Depending on the

species, a given amount of heat from alternating temperature could provide some developmental

advantage over the same amount of heat provided at constant temperature.

This poses a problem for the common practice of using degree-day accumulation based

on average ambient temperature, or the midpoint, and known developmental rates obtained at

constant temperatures to develop phenology models for pest management purposes. This method

assumes there to be a linear relationship between life history traits and accumulated degree days.

For many insects the relationship is not linear and helps explain why Kithrt et al. (2005) report

that in the codling moth and other insects there are time discrepancies between predicted

population numbers and actual numbers observed in the field. Grasshopper phenology is a

foundation of grasshopper management (Gardner and Thompson 2001) and understanding the

effects of temperature on phenology is essential to developing reliable pest management

techniques.









It is known that grasshoppers develop faster and in general are healthier when reared

under day / night temperature cycles (Uvarov 1966b). More recently, in his study on M~

sanguinipes, Fielding (2004) found that grasshoppers reared at alternating temperatures

developed faster than those reared at the midpoint temperature. As with any terrestrial organism,

grasshoppers are affected by daily temperature cycles. However, grasshoppers behaviorally

thermoregulate by exposing themselves to sunlight, and changes in this behavior or availability

of sunlight will affect the cycle of temperatures they experience, specifically daily fluctuations in

temperature other than the day / night cycle most commonly investigated. There has been little

field work conducted to see what types of temperature fluctuations grasshoppers experience

under natural conditions, and investigated even less are the effects these fluctuations may have

on life history traits. If development time is affected by the way in which heat is obtained and not

just the absolute amount of heat obtained, then other factors such as fecundity and body size

might also be affected.

The grasshopper S. amnericana (Drury) is commonly the most economically important

grasshopper in Florida and has been known to cause severe damage to citrus and ornamental

crops (Capinera 1993). In Florida it is known to have a spring or early summer generation

followed by an autumn generation (Kuitert and Connin 1952). This coincides with the report that

the numbers of nymphs increase in early summer and late September (Squitier and Capinera

2002a). The eggs of overwintering adults lack a prolonged diapause and hatch in the year they

were deposited. Consequently, in Florida, S. amnericana overwinters as an adult (Squitier and

Capinera 2002a). S. amnericana has adapted to nearly every Florida habitat (Squitier and

Capinera 2002b). Like many Schistocerca species, it is a tree and shrub dweller and can usually

be found in abundance in areas where both food and perches are available. Disturbed habitats









near the edges of crop fields and roads often contain large numbers of S. amnericana. Squitier and

Capinera (2002b) report that abundance in pine plantations is about 20 times greater when pine

trees are young and small than in mature pine stands. This may be due to several factors. One, a

mature pine stand usually has a less diverse plant community as compared to younger stands

(Squitier and Capinera 2002b). Second, fully grown pine trees may provide more shade than S.

amnericana prefers. When Miles (1985) placed adults of S amnericana in thermally heterogeneous

environments, he found that on average they spent about 90% of their time in locations where

thoracic temperature was 320C 440C. Like many of the grasshoppers mentioned earlier, S.

amnericana prefers to maintain high internal body temperatures and does so by basking. Hence,

shaded habitats, such as mature pine stands, would likely not be preferred.

In 1991 there was an outbreak of S amnericana in Florida. Considerable damage occurred

in Pasco, Polk, Sumter, and Hernando counties (Capinera 1993b). According to Capinera

(1993a), the outbreak was likely due to a 5-year drought which provided an abundance of

sunshine and increased the ambient temperature. This was coupled with a mild winter in 1990-91

which allowed for high overwintering adult survival rates and abundance of suitable habitat. All

of these combined factors may have induced a gradual population increase over several years to

produce an outbreak population. Most grasshopper research is conducted on species that occur in

the western United States where outbreak populations frequently occur. The concentration of

work on other species and lack of frequent outbreaks of S. amnericana has caused this species to

have not been well studied (Capinera 1993a). However, its close relation to two species well

known for outbreaks, S. gregaria in Africa and S. piceifr~ons in Mexico, along with the 1991

outbreak underscores their potential to reach outbreak populations and warrants further study.










In preliminary laboratory testing, S. amnericana adults oriented themselves around a

radiant heat source and reached temperatures of over 40oC for short periods of time when

allowed to position themselves in front of 100w incandescent light bulbs. Achieving

temperatures this high in the field is very unlikely without the presence of direct sunlight. During

hours when a heat source (in this case an incandescent light bulb) was provided, S. amnericana

adults often did not maintain a constant temperature, but allowed their body temperature to

fluctuate between 30oC and 41oC. This corresponds with the idea suggested by Kemp (1989),

that grasshoppers will only use a portion of the heat available to them on a given day. These

preliminary tests, along with observations that S. amnericana adults do not bask all hours of the

day, suggest that sunlight is only required for a portion of the day. This is relevant in Florida,

where during the summer there are very few days with no cloud cover, allowing for a full day of

sunlight. Also, there are times when cloud cover can remain for several days, which could be

detrimental to grasshopper development and fecundity.

Our first obj ective was to assess the effects of sunlight on the body temperature of S.

amnericana in the field, to determine how adverse weather conditions such as rain and cloud

cover may affect body temperature, and to ascertain what sorts of temperature fluctuations a

grasshopper might experience throughout the course of a day. The second obj ective of this study

was to determine the effects ecologically relevant cycles of frequency and amplitude of

temperature change on development and fecundity by developing standardized alternating

temperature regimes based on what grasshoppers experience in the field. Such information will

allow for development of more accurate phenological models that are not based on ambient

temperature, but more accurate body temperatures, not only for this species but for others as

well.









CHAPTER 2
EFFECTS OF WEATHER AND BEHAVIOR ON BODY TEMPERATURE

Methods and Materials

The purpose of this portion of the study was to assess the effects of sunlight, adverse

weather, and behavior on the body temperature of S. amnericana in the field, and more

specifically, tries to determine what kinds of temperature fluctuations a grasshopper might

experience due to these factors. To accomplish this, grasshoppers were wired to thermocouples

and placed in the field where observations of body temperature, behavior and environmental

conditions were recorded.

Study Site

Members of the genus Schistocerca are shrub and tree dwellers. Many of the shrubs and

trees used as perches are also host plants, such as citrus. In an effort to promote natural behavior,

citrus trees were used as the site of the field study. Three potted grapefruit trees (1.5 2m high)

were placed on a light gray gravel surface which was assumed to have similar thermal and

reflective properties as sandy soil, the normal substrate for citrus. The trees were arranged as if

they were a single tree with a dense canopy and several protruding branches. Trees were watered

as needed and occasionally sprayed with a copper solution to control scale insects (Hemiptera :

Margarodidae). Trees were also fitted with ant barriers (Line Guard Inc., Elyria, OH) to deter

predation by ants. Observations were made irregularly from mid-May to mid-November of 2005.

Measurement of Body Temperature

Grasshoppers used in the field study were pre-reproductive adult females ofS. amnericana

taken from a lab colony at the University of Florida. Only pre-reproductive females were used to

prevent any differences in behavior that may be due to necessity of heat intake, which could

differ between stage and sex. In order for continuous body temperatures to be taken,










grasshoppers were fitted with a permanent thermocouple (36 gauge, Teflon insulated,

thermocouple wire, part# TT-T-36-SLE, Omega Engineering, Inc., Stamford, CT) (Fig. 2-1). The

thermocouple consists of a separate copper and constantan wire, with a Teflon coating. One end

of the pair was connected to a male terminal, copper to positive and constantan to negative (Sub

Mini T/C Connector, Part # SMP-T-M, Omega Engineering, Inc., Stamford, CT). The copper

side of the other end was threaded through a sewing needle (size 28), which was then inserted

laterally into the central mesothorax, and pulled through so that the end of the wire runs

completely through the thorax. The copper wire was then soldered to the constantan wire

forming the site of temperature measurement, which was then pulled back into the center of the

thorax. Hot glue (SuperPower Slow Setting Hot Melt Glue, Model No. BSS6-4, Arrow Fastener

Co. Inc., Saddle Brook, NJ) was used to seal off the wounds and to hold the wire in place. There

seemed to be no behavioral effects due to the process, and grasshoppers lived for more than a

week in this state. Similar results have also been noted in other studies where thermocouple

wires were inserted into the thorax of grasshoppers (Carruthers et al. 1992, Lactin and Johnson

1996). Later dissection of the wired grasshoppers showed there to be no damage to internal

organs. This method allows for continuous temperature readings to be taken and is much more

accurate than the common "grab and stab" method used in many other studies (Beck 1983,

Begon 1983, Kemp 1986, Willot 1997, Blanford and Thomas 2000,).

Four grasshoppers were wired with thermocouples each morning the tests were to take

place. One grasshopper was hot-glued to a branch near the center of the canopy to provide

continuous shade (shade constrained grasshoppers), allowing the measurement of minimum

temperatures that could be encountered by a grasshopper on that particular day. This could then

be used as a relative ambient temperature measurement of the grasshopper and be compared with









readings for actual ambient air temperature. A second grasshopper was hot-glued to a green pipe

cleaner which was attached to a branch fully exposed to sunlight and positioned so that the long

axis of the body was perpendicular to the sun' s rays (sun constrained grasshoppers), theoretically

allowing the measurement of maximum temperatures that may be encountered. After a few

minutes, both stationary grasshoppers quit struggling to get free, therefore minimizing any

effects of metabolic heat. The fully exposed grasshopper only began struggling again when it

began reaching lethal temperatures. The remaining two grasshoppers were marked with either

red or blue lettering enamel (Sign Painters' 1 Shot Paint Peinture Lettering Enamel, 153-L

Process Blue and 165-L Rubine Red, Consumers Paint Factory, Inc., Gary, IN) on the hind tip of

the forewings. This allowed them to be easily distinguished when behavioral observations were

being recorded. These two grasshoppers were placed on the grapefruit trees to roam freely,

allowing them to behaviorally thermoregulate and to maintain desired body temperatures. Free

roaming grasshoppers were provided with enough thermocouple wire to allow unlimited

movement within the trees. This provided measurements of preferred temperatures in contrast to

the minimum and maximum measurements provided by the two stationary grasshoppers.

Once all four grasshoppers had been placed on the trees, they were connected to data

loggers (Easy View Dual Input Thermometer, model# EAl5, Extech Instruments, Waltham,

MA). Each data logger had two terminal inputs; hence, one logger was used for the shaded and

exposed grasshoppers and one for the two free roaming grasshoppers. The data loggers were

housed in plastic containers with a watertight lid. Temperature readings were logged every ten

seconds throughout the duration of the tests to provide a more accurate view of how quickly

internal body temperature responded to weather changes.









Behavioral Observations

After the data loggers had been started and it had been determined that all were working

properly, grasshoppers were allowed to settle for a short period before behavioral monitoring

began. Visual monitoring took place every 15 min. during the time when grasshoppers were most

actively thermoregulating, usually from the start of the experiment (9:00 a.m.) until about 4:00

p.m. Behavioral observations for the free-roaming grasshoppers were recorded as five different

categories: completely shaded, partial sun exposure (about 25-75% of body surface) with a

parallel orientation, partial sun exposure with a perpendicular orientation, complete sun exposure

with a parallel orientation, and complete sun exposure with a perpendicular orientation. These

behavioral orientation responses were then numerically rated as 0, 0.375, 0.50, 0.75, and 1.00,

respectively. Previous research (Uvarov 1966b) has shown that orientation of the long axis of the

body plays an important role in temperature regulation and that a perpendicular orientation

provides for higher attainable temperatures. Therefore, a rating of 1.00 represents a grasshopper

receiving the most solar radiation possible. The 0.75 rating for the parallel orientation was

derived from measurements taken that showed a 25% reduction in body surface area exposed to

sunlight when moving from perpendicular to parallel orientation. Each rating was then arbitrarily

halved for partial exposures. After behavioral observations were made, any tangles that may

have developed in the wires since the last observation were removed. Also, during this time the

two stationary grasshoppers were checked to make sure they remained in position, and that the

exposed grasshopper' s orientation remained perpendicular. Adjustments needed to maintain a

perpendicular orientation were accomplished by repositioning the pipe cleaner. A total of 6 d (2

grasshoppers on each day) yielded behavioral data suitable for statistical analysis.









Environmental Conditions

Measurements taken throughout the test period included sunlight intensity, ambient

temperature, and humidity. Sunlight intensity was taken every second and the average recorded

every 10 sec as W/m2 with a Silicon Pyranometer smart sensor (Part # S-LIB-MOO3, Onset

Computer Corp., Bourne, MA) attached to a Hobo Micro Station (Part # DOC-H21-002, Onset

Computer Corp., Bourne, MA). The light sensor was placed as close to the test trees as possible

without becoming shaded. Temperature and humidity readings were recorded every 10 sec with a

Hobo data logger (Hobo Ul2 Temp/RH/Light/External Data Logger, Part # Ul2-012, Onset

Computer Corp., Bourne, MA) placed on top of a plastic plate supported by cork legs, which was

then placed in the shade of a building away from any vegetation.

Statistical Analysis

Descriptive statistics for field data were calculated using Microsoft Excel (Microsoft

Corporation, Redmond, WA). A simple linear regression was performed on sunlight intensity

and percent temperature difference between sun constrained grasshoppers and ambient

temperature. Simple linear regressions were also performed on mean behavior and ambient

temperature, and temperature differences between constrained grasshoppers and free roaming

grasshoppers and behavior. Body temperatures, ambient temperature, % RH, and sunlight

intensity for each day were plotted. Rates of temperature increase and decrease were calculated

using the linear portions of body temperature plots. SAS Analyst 9.0 (SAS Institute Inc., Cary,

NC) was used to perform multiple linear regression analysis and plot partial regressions on the

combined data of body temperatures and behavior, ambient temperature, and sunlight intensity

from June 29, July 5, and July 8, 2005.












Measurement of Body Temperature

Body temperature, ambient temperature, % RH, and sunlight intensity were plotted for

each of the 7 d tested (Fig. 2-2 2-9). For each of the 7 d tested, mean values of body

temperature were calculated (Table 2-1). Because of the very large sample sizes (>2500), even

extremely small differences between body temperatures (such as 0.050C) were considered

significant by ANOVA, causing every temperature to be considered different. Therefore, this

information was omitted from the table.

As expected, sun constrained grasshoppers attained the highest maximum body

temperature of all the grasshoppers tested, with the exception of one day tested in November

(Table 2-1). What was not expected was how high maximum body temperature reached. On

three of the 7 d tested, internal body temperatures rose above 500C, causing thermal death.

Increased respiration rates were observed in sun constrained grasshoppers at such high

temperatures, possibly as a cooling mechanism (Heinrich 1993, Casey 1981). Unfortunately, this

behavior could not be maintained indefinitely and eventually some of the grasshoppers reached

lethal temperatures. Body temperatures at their maximum were 12.4-19.60C above ambient

temperature, often around 75% above ambient temperature. Sun constrained grasshoppers were

meant to represent the highest attainable temperature for that day. For the most part this was true,

but there were factors acting on sun constrained grasshoppers, such as wind and rain, that the

free roaming grasshoppers could avoid. Consequently, mean body temperatures of sun

constrained grasshoppers were not always the highest throughout the day.

Shade constrained grasshoppers tended to have the lowest average body temperatures for

each day, with the exception of the two hottest and sunniest days, July 5 and July 8, where the


Results










body temperature of the shade constrained grasshopper was very similar to those of the free

roaming grasshoppers (Table 2-1). On these days, free roaming grasshoppers spent most of their

time shaded from direct sunlight and consequently had body temperatures similar to those of the

shade constrained grasshopper. On five of the seven days tested, shade constrained grasshoppers

had a mean body temperature greater than ambient temperature, with one day having a maximum

body temperature that was 26.9% above ambient temperature. The highest body temperature

reported in a shade constrained grasshopper was on the hottest day tested, July 8, where body

temperature reached 41.30C.

Mean body temperatures for free roaming grasshoppers were somewhat harder to

generalize, but for the most part they were between those for shade and sun constrained

grasshoppers (Table 2-1). With the exception of November 17, maximum body temperature of

free roaming grasshoppers never surpassed the maximum for sun constrained grasshoppers. On

all days but November 17, free roaming grasshoppers raised their body temperature above 40oC

for short periods of time, which corresponds well with what was found in preliminary laboratory

tests.

On several occasions, night-time body temperatures were also recorded for all four

grasshoppers. For the most part, all grasshoppers maintained body temperatures very similar to

that of ambient, usually within loC above or below ambient.

Environmental and Behavioral Observations

For each of the days where data were available, mean values of sunlight intensity,

ambient temperature, relative humidity, behavior, as well as differences between constrained

grasshopper body temperatures and ambient temperature were calculated (Table 2-1).

Measurements of sunlight intensity were available for four of the days tested. The higher the









mean sunlight intensity the higher the mean percent difference between body temperatures of sun

constrained grasshoppers and ambient temperature. Simple linear regression analysis between

percent difference in temperature and sunlight intensity produced an R2 Value Of 0.99 (Fig. 2-10).

A multiple linear regression analysis with sun constrained body temperature and ambient

temperature and sunlight intensity revealed significant positive linear relationships between body

temperature and both ambient temperature and sunlight intensity (Table 2-2, Fig. 2-11). Not

surprisingly, sunlight was the most significant factor, accounting for 0.8664 of the total r 2. The

data tend to be clustered in the partial regression plots because most body temperatures of sun

constrained grasshoppers were on the higher end and at high levels of sunlight. A multiple linear

regression analysis with the body temperature of shade constrained grasshoppers and ambient

temperature and sunlight intensity also revealed significant positive linear relationships between

body temperature and both ambient temperature and sunlight intensity (Table 2-2, Fig. 2-12). In

contrast to the regression analysis for the sun constrained grasshopper, ambient temperature was

the most significant factor, accounting for 0.9539 of the total r 2. The models given for predicting

body temperature of sun and shade constrained grasshoppers (Figs. 2-11c, 2-12c) were found to

be reliable.

Unlike the constrained grasshoppers, free roaming grasshoppers were allowed to

thermoregulate behaviorally. Both parallel and perpendicular orientations to the sun were

observed, but no grasshopper was ever observed in the flanking position. As a general rule,

grasshoppers tended to begin basking immediately in the morning sun after being placed on the

test trees. In all but one instance, the grasshopper with the higher mean behavioral orientation

rating (the one that spent more time basking) of the two free roaming grasshoppers had the

higher mean daily body temperature (Table 2-1). Increases in mean behavioral orientation rating









for different days reflected increases in the mean difference between the body temperature of

free roaming grasshoppers and those of shade constrained grasshoppers (Fig. 2-13a). Conversely,

increases in mean orientation behavior caused decreases in the mean difference between body

temperature of free roaming grasshoppers and those of sun constrained grasshoppers (Fig. 2-

13b). Mean daily behavioral orientation rating decreases with increases in mean daily ambient

temperature (Fig. 2-14).

A multiple linear regression analysis between the body temperatures of free roaming

grasshoppers and ambient temperature, sunlight intensity, and behavioral orientation rating

showed significant positive linear relationships between body temperature and all three

parameters (Table 2-2, Fig. 2-15). All three parameters were found to be significant contributors

to the variation in body temperature, with ambient temperature being the most significant factor,

accounting for 0.7819 of the total r 2 while behavior only accounted for 0.0063 of the total r 2

The relationship between body temperature and ambient temperature and sunlight intensity for

free roaming grasshoppers follows somewhat similar patterns to those of the constrained

grasshoppers (Figs. 2-15a, 2-15b). Unlike the data for constrained grasshoppers, the data for free

roaming grasshoppers are much more dispersed along the y-axis, indicating some effect of

behavior. The partial regression plot for body temperature and behavior (Fig. 2-15c) takes on a

columnar appearance. However, a pattern still emerges; as behavioral orientation rating increases

there is a slight increase in body temperature. The model for free roaming grasshoppers (Fig.

2-15d) is not as accurate as those for constrained grasshoppers.

Temperature Fluctuations and Adverse Weather Conditions

Rates of body temperature increase in sun constrained grasshoppers were calculated from

linear portions of body temperature plots where sunlight intensity was 800 W/m2 Or greater










(sunny conditions). Sunlight intensity fluctuated too rapidly to allow for rates of temperature

increase to be measured at a constant intensity. The mean (A SD) rate of body temperature

increase for sun constrained grasshoppers was 1.240C/min & 0.009, with a maximum of

2.50C/min, and a minimum of 0.710C/min.

Large fluctuations in body temperature (> 15% change in temperature) occurred several

times a day, averaging 3.67 & 1.50 (Mean & SD) changes per day, and were often the direct

consequence of changes in weather or sunlight intensity (Figs. 2-2 -2-9), with a few due to

behavioral thermoregulation. We used a 15% change in temperature to define a large fluctuation.

This eliminated from the calculations, any small changes that might have been caused by very

brief changes in sunlight intensity, where the full effect of the change could not have been

observed. Additionally, body temperatures of sun and shade constrained grasshoppers show a

high frequency of lower amplitude temperature changes not caused by cloud cover or rain which

were also eliminated by the use of the > 15% rule. The rates of change and amplitude (%

temperature change) were calculated for large temperature fluctuations caused by cloud cover or

rainfall (Table 2-3). Cloud cover caused a higher mean rate of temperature decrease than did

rainfall, but had a lower mean % decrease. The frequency and amplitude of temperature

fluctuations for the laboratory portion of this study were based on these observations.

The lowest temperature recorded as a result of adverse weather conditions, 20. 10C, was

due to rain on June 22 and was 7.6% below ambient temperature. On June 22 and June 29, when

frequent cloud cover and rainfall were recorded, mean body temperatures were up to 8oC lower

for all grasshoppers than on predominately sunny days (Table 2-1).









Discussion


Measurement of Body Temperature

Sun constrained grasshoppers achieved body temperatures that were 12.4-19.60C above

ambient. Fielding (2004) reported similar results of 15-200C above ambient temperature.

However, Fielding's grasshoppers were at ground level and dead, reducing both convective and

evaporative cooling. The grasshoppers in the current study may have reached even higher

temperatures had they been treated in a similar manner. Uvarov (1966a) reports that in L.

migratoria maximum temperatures reach 42.7oC, but does not specify the source of radiant heat.

S. amnericana exhibited maximum temperatures several degrees higher (Table 2-1), even in free

roaming grasshoppers. Though sun constrained grasshoppers did not always attain the highest

body temperatures, they provided a reasonable representation of maximum body temperatures

that may be attainable by a grasshopper on a summer day in Florida. The body temperatures of

sun constrained grasshoppers reported in this study were often around 75% above ambient

temperature. Theoretically, grasshoppers with the ability to thermoregulate could also reach such

temperatures and, in fact, they do. The fact that grasshoppers possess the ability to raise their

body temperature so high above ambient, puts into question the method of using mean ambient

temperature to model phenology and population sizes. Surprisingly, body temperatures of sun

constrained grasshoppers attained lethal levels ( > 500C) when exposed to direct sunlight,

emphasizing the need for cooling. To the author' s knowledge there is no other study that

documents thermal death in Schistocerca species from exposure to direct sunlight.

Shade constrained grasshoppers, for the most part, had the lowest mean daily body

temperature. However, there were 5 d when mean daily body temperatures were above ambient,

sometimes as much as 26%. The likely explanation for shade constrained grasshoppers having a










higher body temperature than ambient is that the grasshoppers were located within the grapefruit

trees and surrounded by vegetation. Sheltering vegetation can provide some amount of insulation

by reducing air flow and preventing heat escape, causing air temperature within the trees to be

different from that measured out in the open air. The surface of the vegetation can also reflect

light back into the canopy and possibly increase temperature even more. The thermal and

reflective properties of the gravel also likely added to the increased temperature. In addition,

when night time temperatures were recorded they were shown to be very similar to ambient,

suggesting that temperature differences between ambient and shade constrained grasshoppers

were caused by radiant energy from the sun being trapped within the tree canopy. These data

show that even when completely shaded, grasshoppers will experience body temperatures much

higher than that of ambient temperature (up to 26%) due to differences in microclimate, and that

measurement of microclimatic conditions would likely be more appropriate for modeling

purposes.

As expected, the mean daily body temperatures of free roaming grasshoppers were

usually between the body temperatures of the constrained grasshoppers, which represent the

extremes in temperature that could be experienced. Kemp (1989) suggested that grasshoppers do

not utilize all of the available heat in a day and that sunlight is not required all hours of the day

for optimal body temperature to be maintained. The data from this study correspond with this on

days when ambient temperature is high and sunlight is available for most of the day. Data from

July 5 and 8 shows that mean body temperatures of free roaming grasshoppers were well below

those of sun constrained grasshoppers, the theoretical maximum (Table 2-1). On days when

ambient temperature was cooler or sunlight was limited, grasshoppers seemed to be attempting to

utilize most of the heat available to them. Data from June 14, 17, 22, and 29 shows increased









basking behavior and body temperatures of free roaming grasshoppers very close to, and in two

instances higher than, those of sun constrained grasshoppers (Table 2-1). However, free roaming

grasshoppers did not bask every moment of the day even on cooler days. All of this suggests that

there is a certain amount of heat a grasshopper needs obtain to achieve optimal fitness and that

they will not utilize more heat than is necessary to do so. High body temperatures do have

negative side effects and would be avoided when there is no longer a benefit provided by

increasing body temperature. July 5 and 8 would represent days when more than enough heat

was available, while June 14, 17, 22, and 29 were days where heat was more limited and

grasshoppers took advantage of most of the heat that was available.

November 17, 2005 was the only day tested during the winter months. November 17

provided some interesting data. As mentioned earlier, sun constrained grasshoppers attained the

highest maximum body temperature of all the grasshoppers tested, with the exception of this day,

when free roaming grasshoppers attained the highest maximum and mean temperatures.

November 17 was the coolest day tested, and was characterized by a stiff breeze and low

sunlight intensity. Because there were no rainfall events, the most likely explanation for lower

sun constrained body temperatures is that the brisk breeze prevented the exposed sun constrained

grasshopper from reaching higher temperatures, while free roaming grasshoppers might have

been able to position themselves to bask while avoiding exposure to the wind. What is also

interesting about November 17 is that the mean body temperature of the shade constrained

grasshopper was below mean ambient temperature, and is probably a result of the high wind

speed and low sunlight intensity. This helps support the idea that, on dates other than November

17, radiant energy from sunlight trapped by the vegetation was causing shade constrained

grasshopper temperatures to rise above ambient.









Environmental and Behavioral Observations

A simple linear regression revealed a very significant positive relationship (R2=0.99)

between mean sunlight intensity and the mean difference between body temperatures of sun

constrained individuals and ambient temperature. As expected, the higher the mean sunlight

intensity, the higher the body temperature rose above ambient temperature. When a multiple

regression analysis was conducted on the raw data, accounting for the effects of both sunlight

and ambient temperature on body temperature, sunlight was the most significant factor,

accounting for 0.8664 of the total r 2 (Table 2-2). Such a dependency of body temperature on

sunlight was expected in sun constrained grasshoppers, and shows how influential sunlight can

be on body temperature. These results are in agreement with many other studies. Lactin and

Johnson (1997) found direct sunlight to significantly affect grasshopper body temperature.

Fielding (2004) also reported a very strong relationship between grasshopper body temperature

and sunlight. The maj ority of data points shown in the plot of predicted body temperatures vs.

actual body temperatures given by the multiple regression analysis (Fig. 2-11c) follow the

general linear pattern predicted by the model. A model predicting the body temperature of

constrained grasshoppers is not likely to prove useful in trying to model the phenology and

population of free moving grasshoppers; however, the model does provide some interesting

information. It provides a method for calculating a rough estimate of maximum grasshopper

body temperature when the ambient temperature and sunlight intensity are known. Additionally,

there are some outlying data points which seem to reduce the model's predictive power. These

points are artifacts of the delayed response of body temperature to changes in sunlight intensity.

Changes in sunlight intensity were recorded immediately, while the effects these changes had on

body temperature took some time to manifest themselves in the body temperature recordings.









In contrast to sun constrained grasshoppers, the body temperature of shade constrained

grasshoppers was much less dependent on sunlight intensity and more dependent on ambient

temperature. The multiple regression analysis showed ambient temperature having a linear

relationship with body temperature and accounting for 0.9539 of the total r 2 (Table 2-2). This

result was expected, because shade constrained grasshoppers were denied access to direct

sunlight. Therefore, the body temperature of shade constrained grasshoppers should be very

dependent on ambient temperature. This is partially supported by the fact that the data for shade

constrained grasshoppers in the partial regression plot between body temperature and sunlight

intensity (Fig. 2-12b) are much more dispersed along the y-axis than the data for the equivalent

sun constrained plot (Fig. 2-11Ib). In addition, the plot of predicted vs. actual body temperatures

(Fig. 2-12c) does not exhibit nearly as many outlying values as that of the plot for sun

constrained grasshoppers (Fig. 2-11c). This further shows the reduced effect of sunlight on body

temperature in shade constrained grasshoppers. It is suspected that the effect of sunlight on body

temperature and the positive relationship between the two (Fig. 2-12b) are due to the partial

dependency of ambient temperature on sunlight intensity (even though tests for co-linearity were

negative) and differences in microclimatic conditions which would also be partially dependent

on sunlight intensity.

Blanford and Thomas (2000), among many others, report a non-linear relationship

between ambient temperature and body temperature in grasshoppers. They report that body

temperatures reach equilibrium at higher ambient temperatures. There is a simple explanation for

the discrepancy between their study and this study. They were trying to show behavioral

thermoregulation and at very high temperatures grasshoppers were actively cooling their bodies,

while at cooler temperatures grasshoppers were basking to raise their body temperature. In the









current portion of this study, behavior has been removed from the equation for constrained

grasshoppers and body temperature is solely dependent on environmental conditions. For this

reason there is a linear relationship between body temperature of shaded grasshoppers and

ambient temperature.

Free roaming grasshoppers adopted both parallel and perpendicular orientations. The fact

that grasshoppers were never observed in a flanking position (when a grasshopper raises one

hind leg and lowers the other while in a perpendicular orientation to maximize exposure) is not

entirely surprising. Flanking has never been reported in S. amnericana. Unfortunately, the

experiment was not designed to investigate how body orientation and body temperature interact,

and the duration of orientation was never recorded. Such information is necessary to determine

how orientation directly affects body temperature in real time and vice versa. However, mean

behavioral ratings were used to determine the effects of behavior on mean daily body

temperature. Increased mean behavioral orientation rating reflected increases in the mean

difference between the body temperature of free roaming grasshoppers and those of shade

constrained grasshoppers (Fig. 2-13a) and decreases in the mean difference with sun constrained

grasshoppers (Fig. 2-13b). As grasshoppers spend more time basking, their mean body

temperatures rise above those of shaded grasshoppers and toward those of sun constrained

grasshoppers which proves behavior has an effect on body temperature in S. amnericana.

Additionally, of the two grasshoppers tested each day, the one with the higher mean behavioral

orientation rating always had a higher mean body temperature for that day. This shows that the

behavioral orientation ratings used in this study were suitable for describing grasshopper

behavior with respect to sunlight absorption. The fact that mean daily behavioral orientation

rating decreases with increases in mean daily ambient temperature (Fig. 2-14) corresponds with










the previous idea that grasshoppers do not use all the heat available to them on very warm days.

It also shows that grasshopper behavior is dependent on environmental conditions. The more

ambient heat that is available to a grasshopper, the less heat it must obtain through basking.

Behavior obviously has an effect on body temperature, and the linear plots of body temperature

(Figs. 2-2 2-9) reinforce this fact even further by showing differences in body temperature

between the two different free roaming grasshoppers for each day which experienced the same

exact environmental conditions. The differences alone do not provide enough evidence for

behavioral thermoregulation, but help make a strong case in conjunction with the other

behavioral data, especially the comparison between mean behavioral rating and mean ambient

temperature (Fig. 2-14).

The multiple regression analysis for free roaming grasshoppers revealed results similar to

those for constrained grasshoppers. Both sunlight and ambient temperature had significant

positive linear relationships. Again, the relationship between body temperature and ambient

temperature remained linear. Behavior was also found to be a significant factor. However,

behavior only accounted for a small portion of the total r 2 (Table 2-2). The data plotted for the

partial regression plot between behavioral orientation and body temperature (Fig. 2-15c) tries to

show the direct effects of behavior on body temperature. However, this data takes on a columnar

appearance because the times of data collection for a continuous variable, body temperature, do

not perfectly coincide the times of data collection for a discontinuous variable, behavior. As

mentioned before, the duration of behavioral orientations were not recorded and behavior was

only recorded every 15 min, then paired with the corresponding 15 min of temperature data. This

data cannot account for any changes in behavior between the 15 min intervals. Even so, a general

trend can be seen, that as behavioral orientation rating increases so does body temperature (Fig.









2-15c). Unlike the data for constrained grasshoppers, the data for free roaming grasshoppers are

much more dispersed along the y-axis (Figs. 2-15a and 2-15b). This is most likely because

behavior allowed grasshoppers to maintain similar body temperatures over a broader range of

environmental conditions and shows how behavior can affect body temperature by controlling

for environmental conditions.

Why ambient temperature is the most significant factor contributing to the variation in

body temperature in free roaming grasshoppers is uncertain. It could be that the behavioral data

are discontinuous and do not accurately represent how behavior affects body temperature. It is

more likely, though, that the effects of ambient temperature on body temperature are not affected

by grasshopper behavior (unlike the effects of sunlight which can be avoided by seeking shade)

and, therefore, ambient temperature is the most consistent factor affecting body temperature. The

fact that the model for free roaming grasshoppers (Fig. 2-15d) is not as accurate as those for

constrained grasshoppers is likely due to the time discrepancies and discontinuous nature of the

behavioral observations.

Temperature Fluctuations and Adverse Weather Conditions

Rates of body temperature increase in sun constrained grasshoppers averaged 1.240C/min

& 0.009 (mean & SD). These results are somewhat similar to the rate of increase of 0.8280C/min

in Locusta reported by Uvarov (1966a). Lactin and Johnson (1997) provide a measurement of

overall temperature increase per W/m2, but give no measurement of rate. The maximum rate

recorded, 2.50C/min, is very high and at such rates grasshoppers could reach optimal body

temperature in a very short period of time. Rates of decrease and % decrease caused by adverse

weather conditions (Table 2-3) show just how significant the impact of cloud cover or rain can

be on grasshopper body temperature. Unexpectedly, cloud cover caused a faster rate of decrease









than did rainfall. However, rates were calculated using the temperature when the event started

and the lowest temperature recorded within the decrease. Rainfall caused much larger percentage

decreases in temperature than did cloud cover, and at lower temperatures the rate of body

temperature decrease tapered off, therefore causing decreases due to rainfall to have a reduced

rate. There were many small fluctuations in grasshopper body temperature throughout the day.

These changes are possibly due to several factors, including small changes in sunlight intensity,

changes in evaporative cooling, and changes in wind speed. The timing of these small changes in

body temperature is similar between sun and shade constrained grasshoppers, but the amplitude

of change is often larger in sun constrained grasshoppers than in shade constrained grasshoppers.

Sun constrained grasshoppers were more exposed to small changes in both wind speed and

sunlight intensity, whereas shade constrained grasshoppers were sheltered from such changes by

vegetation. All of these observations suggest that these small temperature changes are due

mainly to small changes in abiotic factors.

Aside from limiting a grasshoppers ability to achieve optimal body temperature, adverse

weather conditions caused grasshoppers to experience an average of 3.67 large fluctuations in

body temperature (> 15%) during daylight hours. When trying to model insect populations, it is

common to use data collected from subjects reared under constant conditions. Whether or not it

is proper to use such data should be investigated on an individual basis. Depending on the

answer, it might be possible to use daily averages for environmental parameters to model insect

populations, or it may be necessary to record and include fluctuations that occur in

environmental parameters. The next part of this study attempts to determine if there are

developmental and reproductive consequences associated with daily fluctuations in temperature,

such as those observed in the field portion of this study











Table 2-1. Grasshopper body temperatures and behavioral orientation ratings, and environmental
data (mean + SD) obtained from field studies. Differences between constrained
grasshoppers and ambient temperature are given in percent difference. See text for
explanation of treatments and environmental parameters.
June 14 2005 June 17 2005 June 22 2005
Parameter Mean Min Max Mean Min Max Mean Min Max
Free Roaming Blue ( C) 37.03 + 2.99 29.80 44.40 37.37 + 2.16 32.90 43.40 28.19 + 7.42 20.90 47.20
Blue Behavior 0.56 + 0.43 0.38 + 0.41 0.73 + 0.40
Free Roaming Red ( C) 35.61 + 2.61 30.60 42.70 41.71 + 1.84 34.70 44.90 27.45 + 6.25 20.80 43.80
Red Behavior 0.36 + 0.35 0.78 + 0.23 0.46 + 0.45
Shade Constrained ( C) 33.61 + 2.52 28.20 38.80 33.82 + 1.87 28.10 36.10 26.20 + 4.27 21.10 37.10
Sun Constrained ( C) 39.31 + 3.11 33.20 45.80 40.36 + 3.30 32.60 45.70 28.13 + 7.77 20.20 50.40
Ambient Temperature ( C) 30.41 + 1.77 27.80 33.40 28.95 + 1.45 26.60 31.40 27.68 + 1.37 25.10 30.80
Shade Const. Ambient (oo) 10.44 + 2.91 1.40 18.30 16.86 + 4.67 -0.60 26.90 8.76 + 6.26 -7.90 22.10
Sun Const. Ambient (oo) 29.64 + 12.44 2.69 62.60 39.35 + 8.58 14.94 55.90 30.26 + 18.22 -7.60 76.40


oo RH
Sunlight Intensity (w mZ)


58.43 + 8.66 36.80 76.70 68.19 +


5.91 54.30 79.90











Table 2-1. Continued.

July 5 2005 July 8 2005
Parameter Mean Min Max Mean Min Max
Free Roaming Blue (oC) 38.17 1.45 33.80 41.60 37.40 1.85 32.50 41.30
Blue Behavior 0.12 0.24 0.06 0.13
Free Roaming Red (oC) 38.32 1.87 32.10 42.90 37.35 2.09 31.60 42.20
Red Behavior 0.13 0.25 0.07 0.15
Shade Constrained (oC) 37.68 1.68 32.80 40.10 37.76 2.27 32.10 41.30
Sun Constrained (oC) 45.90 2.96 33.20 52.10 46.13 3.42 33.50 51.60
Ambient Temperature (OC) 32.20 1.04 28.50 33.30 32.58 1.25 28.70 34.50
Shade Const. Ambient (%) 17.00 2.29 9.56 21.35 15.83 3.55 2.90 22.90
Sun Const. Ambient (%) 42.47 6.08 16.60 57.70 41.66 9.90 2.40 56.20
% RH 51.07 3.28 46.50 63.70 56.19 6.52 47.80 74.10
Sunlight Intensity (w /m2) 801.40 99.50 420.60 979.40 805.70 204.80 0.60 1014.40











Table 2-1. Continued.


June 29 2005
Mean
30.30 &
0.60 +
30.88 &
0.60 +
28.46 &
31.84 &
27.29 &
4.13 &
16.41 +
81.18 &


November 17 2005
Mean


Parameter

Free Roaming Blue (oC)
Blue Behavior
Free Roaming Red (oC)
Red Behavior
Shade Constrained (oC)
Sun Constrained (oC)
Ambient Temperature (OC)
Shade Const. Ambient (%)
Sun Const. Ambient (%)
% RH
Sunlight Intensity (w /m2)


Min Max


Min Max


3.62
0.40
3.92
0.37
2.29
4.57
1.09
4.79
13.55
4.32


25.30


25.10


25.40
25.30
25.00
-5.70
-6.60
71.50


40.40 27.21 4.28 17.30 38.40


41.40 27.63


5.12 18.70 37.60


33.60
47.80
29.10
17.20
66.60
87.20


21.47
26.35
21.84
-0.90
21.58
29.29


2.60
3.94
1.43
15.19
21.79
2.18


13.80
15.80
20.70
-40.90
-29.70
24.60


24.80
33.10
27.40
16.33
55.90
33.90


316.80 223 .40 39.40 1164.40 460.20 196.80 51.90 731.90















Table 2-2. Regression statistics for multiple regressions between grasshopper body temperatures

and environmental parameters from June 29, July 5, and July 8 2005.


Parameter
Estimate t value P


-6 6499 -17 12 <0 0001

1 2070 84 03 <0 0001

0 0165 138 46 <0 0001



-13 9944 -98 69 <0 0001

1 5149 288 82 <0 0001

0 0033 74 59 <0 0001



-14235 -4 14 <0 0001

1 0921 86 68 <0 0001

0 0049 46 30 <0 0001

1 2106 20 68 <0 0001


r2 Variable


Type II SS Partialr 2


SS F-value P


Sun
Constramned Model 489263 50939 <0 0001 0 93 Intercept

Error 37531 Amblent Temperature

Sunhight Intensity


Shade
Constramned Model 180960 141308 <0 0001 0 97 Intercept

Error 5003 Amblent Temperature

Sunhight Intensity


Free
Roammg Model 192781 18664 <0 0001 0 82 Intercept

Error 42494 Amblent Temperature

Sunhight Intensity
Behavior


0 0644

0 8644





0 9539

0 0192





0 7819

0 0312

0 0063











Table 2-3. Rate of temperature decrease and percent temperature decrease (mean + SD) of body
temperature in sun constrained grasshoppers under cloudy or rainy weather
conditions.
Condition Rate of Decrease Min Max % Decrease Min Max
Cloud Cover 0.0163 OC/sec & 0.0063 0.0074 OC/sec 0.0269 OC/sec 22.34 & 3.98 17.60 31.20
Rain Fall 0.0087 OC/sec & 0.0058 0.0031 OC/sec 0.0146 OC/sec 38.83 & 9.49 30.20 49.00















Point of temperature measurement


L~ ~f
I~
vl
O
U










Tetaninal.


Figure 2-1.


Diagram of wiring method for recording internal body temperature of
grasshoppers showing how to connect copper and constantan wires.


UU







































0000000000000000000000000000000oo



Time


Linear plot of continuous body temperatures of free roaming grasshoppers and
cloud cover from May 19 2005. Cloud cover is represented by the discontinuous
bars in the upper portion of the plot.


45



S40

e

C 35


30



25



20


2-2.


0






Figure :
















50



45



H 40



35



30



25
55



50



45



E 40



35



30


cooococooooooconococoQooocoooooooom


Time


Figure 2-3.


Linear plot of continuous body temperatures of (a) constrained grasshoppers and
cloud cover and (b) free roaming grasshoppers and cloud cover from June 14
2005. Cloud cover is represented by the discontinuous bars in the upper portion of
the plot.















SO



415



40



~35



30



25



20
55



50



45



k40



* 35







25



20


coo



Tirre


coo


Figure 2-4.


Linear plot of continuous body temperatures of (a) constrained grasshoppers with
ambient temperature and cloud cover and (b) free roaming grasshoppers with
ambient temperature and cloud cover from June 17 2005. Cloud cover is
represented by the discontinuous bars in the upper portion of the plot.












55



50



45



40

35

30



2 5







20
55



60



45



40



~35



30







20







Figure 2-5.


IOOO OOOQQQQOQQQQ QOOQQ00Time0

Linear plot o cotnous odytmeauef()cnsriershpeswt



aminenrplt o otnosb temperatures and lou cove ad(b fe raingd grasshoppers with

ambient temperature and cloud cover from June 22 2005. Cloud cover is
represented by the discontinuous bars in the upper portion of the plot.




































d-ulg _secon...













~ngnn94x~wsgqsa~g~~x~g~~x~~gn1






14284~~ b~4~~aP~fa4iEjS~~F~~l~


100


ibleril Temperalue
-S6 Italaine Humia~y


S40

35


X89XSXBS~SX~X~SXSXWSWSTime X


Figure 2-6.


Linear plots of continuous body temperatures and environmental parameters from June 29 2005. a) Constrained

grasshoppers. b) Free roaming grasshoppers. c) Ambient temperature and relative humidity. d) Sunlight intensity.
Secondary axis of c given in %RH.


























































fme Time


Sree Rosmiq Blua
IFree Rosmiq Rd


~wSgM Inlarlny


01BBW~9;3
s~a~wisst
u~ii ~iri


Figure 2-7.


Linear plots of continuous body temperatures and environmental parameters from July 5 2005. a) Constrained

grasshoppers. b) Free roaming grasshoppers. c) Ambient temperature and relative humidity. d) Sunlight intensity.
Secondary axis of c given in %RH.











gg


X~W~XSBPTimeSXsO$ BXPXs ThieSRSXWXWSXSXWSWSg


Figure 2-8.


Linear plots of continuous body temperatures and environmental parameters from July 8 2005. a) Constrained
grasshoppers. b) Free roaming grasshoppers. c) Ambient temperature and relative humidity. d) Sunlight intensity.
Secondary axis of c given in %RH.














































E 2s


20





15





Figure 2-9.


:5xa~s~sxn~xasxnsxa~sqsxa~xgsx
:$~8t~BB=BR~X=~~$8sl~~8PAB~~X


xasxsssnss~qs~~sasxsssnqg~qs~sxasx
n~sn~qea,4~~ RX~sN~R~8z~~~aPW~88~4
mY~~~o~- NNNN~MMMnYYtYY~~~n~~~~


Linear plots of continuous body temperatures and environmental parameters from November 17 2005. a) Constrained

grasshoppers. b) Free roaming grasshoppers. c) Ambient temperature and relative humidity. d) Sunlight intensity.
Secondary axis of c given in %RH.












S45
r 40

30
S25
S20
J '15
E 10
-1 5 y=0.054x 1936 r =0.99
0
0 200 400 600 800 1000

Sunlight Intensity wlm2


Figure 2-10. Variation in the percent temperature difference between the body temperature of
sun constrained grasshoppers and ambient temperature in relation to mean
sunlight intensity from June 29, July 5, July 8, and November, 17 2005
(F=255.13, P=0.004 df = 1, 2).





Pi "
3
E

E 01


















a
rrr
a,
a
E O~
a,
r-

o
m -d~



--20




56-


e! 5~-
a
e
n.
E
40-


35-
m

30.


Ambient 'Temperature


b


BOO -700 -60 -siO -400 rOD -20 -0 o 1D 200 COD 400 BO 600 700 BO

Sunlight Intesity


~70
O
O


O
,,


-~,L6o O ~sl

P rg
0~8~ Oo ao ~7
oo ~tti~~B$~Bb O~ gr O o O
o~~o o o

,","E~i
,os~


~Lb~C~ y= 1.2070t + 0.0165s -6.6499



Predicted Body Temperature


Figure 2-11.


Regression analysis of sun constrained grasshoppers from June 29, July 5, and

July 8 2005. a) Partial regression plot of body temperature and ambient

temperature. Y axis represents residuals of regression between body temperature
and sunlight intensity. X axis represents residuals of regression between ambient

temperature and sunlight intensity. b) Partial regression plot of body temperature
and sunlight intensity. Y axis represents residuals of regression between body

temperature and ambient temperature. X axis represents residuals of regression
between sunlight intensity and ambient temperature. c) Plot of actual vs.

predicted body temperature (t = ambient temperature, s = sunlight intensity).


























--6 --5 --4 --3 --2 --1 0 1 2 3 4 5 6
Ambient Temperature


P d~bo dB ~yqB O
oo
on olipo ~ o
~-- P"?-~Y~~~~
~ip ,,

O~e~


-coo3 -700 -00 --on --400 --000 -200-m0 0 100 200 soo 400 50 000 700 co
Sunlight Intesity


42.5





so.o


24 26 28 30) 32 34 GB
Predicted Body Temperature


38 40 42


Figure 2-12.


Regression analysis of shade constrained grasshoppers from June 29, July 5, and
July 8 2005. a) Partial regression plot of body temperature and ambient
temperature. Y axis represents residuals of regression between body temperature
and sunlight intensity. X axis represents residuals of regression between ambient
temperature and sunlight intensity. b) Partial regression plot of body temperature
and sunlight intensity. Y axis represents residuals of regression between body
temperature and ambient temperature. X axis represents residuals of regression
between sunlight intensity and ambient temperature. c) Plot of actual vs.
predicted body temperature (t = ambient temperature, s = sunlight intensity).











Sy =6.5165x -0.5737 R = 0.57















*



0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Mean Behavioral Orientation Rating

y =-13.6376x +9.1419 R = 0.94















0.1 0.2 0.3 0.4 0.5 0.6 0I .8


Mean Behavioral Orientation Rating


7







1

0
-1
10


Figure 2-13.


Regression analysis of behavioral data from June 14, June 17, June 22, June 29,
July 5, and July 8 2005. a) Mean behavioral orientation rating plotted against the
difference between free roaming body temperature and shade constrained body
temperature in (F=13.14, P=0.005, df =1, 10). b) Mean behavioral orientation
rating plotted against the difference between free roaming body temperature and
sun constrained body temperature (F=168.85, P=<0.0001, df =1, 10).













y = -0.1035x + 3.4932 R = 0.73
0.8
w*



.2 0.5






S0.1
I +
272 9303 23
Amben Tepeaur(C

Figre21.Rgesoanlssfrmabeairloinainrtnplteagisam et
te prtr (F 2.8 =.04 f=,1)













C


j
pn rl






--08 -UI --ae -as -u4 --al -a2 -ar ao ar 02 ul a4 as ae ar as as
Behavior


1CI




t
B
E ra
ar

$ -,


Fk~ d

,"i"~
~u 9


Ambient Temperature


""5-l'i~;


ef~6
3
e
a
E PD
Q
"
P -,


me ~ y = 1.09Y2lt + 0.0049)s + 1,2106b 1,4235


. -


Sunlight Intesity


3m*m5m6mmO


Predicted Body Tertmprtr


1...............................1.......
9 XI lr


Figure 2-15.


Regression analysis of free roaming grasshoppers from June 29, July 5, and July 8 2005. a) Partial regression plot of
body temperature and ambient temperature. Y axis represents residuals of regression between body temperature and
sunlight intensity and behavioral orientation rating. X axis represents residuals of regression between ambient
temperature and sunlight intensity and behavioral orientation rating. b) Partial regression plot of body temperature and
sunlight intensity. Y axis represents residuals of regression between body temperature and ambient temperature and
behavioral orientation rating. X axis represents residuals of regression between sunlight intensity and ambient
temperature and behavioral orientation rating. c) Partial regression plot of body temperature and behavioral orientation
rating. Y axis represents residuals of regression between body temperature and ambient temperature and sunlight
intensity. X axis represents residuals of regression between behavioral orientation rating and sunlight intensity and
ambient temperature. d) Plot of actual vs. predicted values for body temperature (t = ambient temperature, s = sunlight
intensity, b = behavioral orientation rating).


~- --









CHAPTER 3
EFFECTS OF TEMPERATURE FLUCTUATIONS ON DEVELOPlVENT AND
REPRODUCTION

Methods and Materials

This portion of the study was conducted in the laboratory in rearing chambers to

determine how development time, adult body size, and female fecundity is affected by

ecologically relevant temperature fluctuations.

Temperature Treatments

Laboratory experiments took place in environmental control chambers. The temperature

fluctuation data obtained from the field was used as a template to construct temperature regimes.

Average daytime temperatures of 33 oC and 38oC were chosen to represent overcast and sunny

days. The chambers were placed on a 14/10 photoperiod and a 16/8 thermoperiod with 16 h of

respective daytime temperature with the last 2 h hours slowly falling to the 8 h of nighttime

temperature at 25oC. After averaging in nighttime temperatures, the two mean temperatures were

29.5 oC and 3 1.5 oC, respectively. The mean temperatures of 29.5 oC and 3 1.5 oC were assigned

different frequencies and amplitudes of temperature change. Frequency of temperature changes

were 1 and 5 changes per daytime period, while the amplitude of change was 4oC and 8oC. The

fifth temperature regime was that of constant temperature at 29.5oC or 31.5oC. This provided

each mean temperature with four fluctuating treatments and one constant temperature treatment

for a total of 10 treatments (Fig. 3-1). Unfortunately, because of two failed attempts and time and

space restrictions, the high frequency treatments were conducted from October 2005 to February

2006, while low frequency and constant temperature treatments were conducted from February

2006 to June 2006. During this second time period, the 31.5oC high frequency low amplitude

treatment was repeated as a control between the two time periods.









Cages


Four replicates were performed per treatment. Each replicate consisted of an aluminum

cage with dimensions of 30 cm X 30 cm X 30 cm, with a solid aluminum bottom, and screen

sides and top, stocked with 15 1st instars hatched within the last 24 hours. Each cage was

provided with a petri dish of dry food (wheat flour, wheat bran, soy flour, and tropical Eish

flakes) and a water supply. Fresh romaine lettuce was provided daily until reproduction began, at

which point lettuce was provided tri-weekly. Cages were rotated daily to compensate for any

temperature stratifieation that might occur within the chambers. Grasshopper frass was removed

daily and cages were cleaned as necessary.

Nymphal Developmental Time

Grasshoppers were monitored daily for molting. Each day, grasshoppers were counted

and the number surviving and the number at each stage were recorded. The stage of any dead

grasshoppers was also recorded. Data were entered into a Microsoft Excel spreadsheet. Average

nymphal development time for each cage was calculated by taking the number of grasshoppers

molting to the next stage each day and dividing it by the number surviving, and then adding the

products from each day until all grasshoppers had reached adulthood.

Body Size and Reproduction

Once all the grasshoppers in a treatment reached adulthood, the grasshoppers were

removed and the sex, femur length, and overall length of each grasshopper were recorded. Even

numbers of females and males were then placed back into the cages. Grasshoppers were then

monitored for mating behavior, at which point 32 oz. deli cups (approximately 14cm high, 8.9cm

in diameter at base, and 11.4cm in diameter at top) Eilled with moist vermiculite were placed

inside the cages as oviposition medium. Each cup was then monitored daily until the first









occurrence of oviposition was recorded. Cages were then monitored tri-weekly and cups were

changed weekly. During this time, any egg pods that had been laid outside the cups were

recorded. Cups were then stored at 5-10.C until they could be processed, at which time the

number of egg pods per cup was recorded, along with the number of eggs in each pod. The

number of egg pods laid per cage was divided by the number of females per cage to arrive at egg

pods / female. Those egg pods laid outside of the cups were included in the egg pod mean but not

the eggs/pod mean. Unfortunately, total reproduction could not be measured due to pesticide

residues detected on lettuce fed to the general colony during the high frequency treatments.

Experimental grasshoppers received the inner portions of this lettuce and consequently, all

treatments had to be terminated for fear of contamination. Thus, the collection of egg pods was

discontinued 115 d after hatch. The same was done for the second set of treatments conducted

from February to June.

Statistical Analysis

Differences in total nymphal development time, body size, egg pods per female, eggs per

pod, and days to first oviposition were analyzed using ANOVA and the Least Squares Means

(LSM) procedure using a Tukey's adjustment in SAS Analyst 9.0. High frequency treatments

conducted from October to February were treated as a separate study and analyzed in a separate

ANOVA from low frequency and constant treatments conducted from February to June. An

ANOVA was conducted on all fluctuating treatments as one group, combining differing

frequencies and time periods to test for any frequency or trial effects such as, difference in food

quality or change in biology due to seasonal differences. Separate histograms for high and low

frequency treatments as well as one for combined treatments of nymphal developmental data









were produced using SAS Analyst 9.0. Bar graphs of body size, egg pods per female, eggs per

pod, and days to Birst oviposition were produced using Microsoft Excel.

Results

To see if the two laboratory studies could be combined, an ANOVA was performed for

each parameter on the repeated and original 31.50C high frequency low amplitude treatments.

The tests revealed significant differences between the two in all parameters. Due to these

differences, data were analyzed as two separate studies. Data were also analyzed together to note

any effects, regardless of whether they were frequency or trial related.

Nymphal development time

Amplitude and temperature effects were found to be significant in high frequency

treatments (Table 3-1). On average, grasshoppers reared at 31.50C reached adulthood 5.2 d faster

than those reared in 29.50C treatments, while those reared in high amplitude treatments reached

adulthood 2.1 d faster than grasshoppers reared in low amplitude treatments (Table 3-2). Mean

comparisons revealed that the 29.5oC high frequency low amplitude treatment is the only

significantly different treatment within high frequency treatments (Table 3-2). Although both

29.5oC treatments had longer average nymphal development times, only the low amplitude

treatment was found to be significantly different from the 31.5oC treatments. This treatment

likely caused all of the amplitude and the maj ority of temperature effects. The distribution of

data for low amplitude treatments shows a distinct separation (Fig. 3-2), which is caused by the

29.5oC low amplitude treatment. Grasshoppers in both 31.5oC treatments took about 38 d to

complete nymphal development, showing no effects of amplitude at this temperature (Table 3-2).

Amplitude, temperature, and amplitude x temperature effects were found to be significant

in low frequency and constant treatments (Table 3-1). On average, grasshoppers reared at 31.50C









reached adulthood 1.6 d faster than those reared in 29.50C treatments (Table 3-2). There was no

difference in nymphal development time between grasshoppers reared at high and low amplitude

in low frequency treatments. However, those reared at alternating temperature reached adulthood

about 2 d faster than grasshoppers reared in constant treatments (Table 3-2). As in high

frequency treatments, mean comparisons revealed only one treatment to be significantly different

from the others. The 29.50C constant temperature treatment was significantly different from all

treatments except for the 29.50C low frequency low amplitude treatment and is the likely the

reason there are significant amplitude, temperature, and interaction effects (Table 3-2). The

distribution of data reveals how similar nymphal development times for low frequency

treatments are (Fig. 3-3). The data for constant temperature treatments is very spread out and is

likely due to the 29.50C and 31.50C constant temperature treatments being so different (Fig. 3-3).

Grasshoppers reared in the 31.5oC constant treatment showed no delay in nymphal development

and even developed faster than one of the alternating 3 1.5oC treatments.

When high and low frequency treatments were combined (minus constant treatments)

there were significant frequency or/and trial effects (Table 4). In general, low frequency

treatments tended to have shorter mean total nymphal development times than high frequency

treatments, 3.9 d shorter (Table 3-3). This can also be seen in the distribution of the data (Fig. 3-

4). However, this difference is likely caused by the 29.50C high frequency treatments. All other

treatments (29.50C low frequency, and both 31.50C high and low frequency treatments) were not

significantly different (Table 3-3). When all treatments were combined there were no amplitude

effects (Table 3-3).










Body size

Sex was the only significant main effect for both femur length and overall length (Table

3-4). In high frequency treatments females had an overall length (in mm, mean & SD) of 59.81 +

1.84 and a femur length (in mm, mean & SD) of 26.45 A 1.13, while males had an overall length

of 51.91 + 1.94 and a femur length of 23.04 & 1.17. In low frequency and constant treatments

females had an overall length of 61.91 + 1.84 and a femur length of 27.82 & 2.59, while males

had an overall length of 53.97 & 1.75 and a femur length of 24.29 & 0.97. Differences in size

between male and female grasshoppers were expected. There were also several significant

interactions involving the sex factor, but a closer look at multiple comparisons revealed no real

differences other than sex. When low frequency and high frequency treatments were combined,

there were both significant sex and frequency or/and trial effects for femur length and overall

length (Table 3-4). High frequency treatments tended to produce smaller individuals than did low

frequency treatments (Fig. 3-5). The only factor to affect grasshopper body size other than sex

was frequency or/and trial effects.

Days to oviposition

Grasshoppers reared in high frequency treatments, on average, took between 56 and 70 d

from hatch to egg deposition. Temperature was the only significant factor affecting the number

of days to oviposition in high frequency treatments (Table 3-5). Grasshoppers reared in 29.50C

high frequency treatments took, on average, about 9 d longer to the first occurrence of

oviposition (Table 3-6). Mean comparisons revealed differences similar to those for nymphal

development time. The 29.50C high frequency low amplitude treatment was shown to be the

only significantly different treatment (Table 3-6) and is likely the sole cause for the significant

temperature effect. Grasshoppers reared in low frequency and constant treatments, on average,









took between 56 and 93 d from hatch to egg deposition (Table 3-6). The results of the ANOVA

(Table 3-5) and mean comparisons (Table 3-6) revealed only the 29.50C constant temperature

treatment to be significantly different. This treatment was so different from the others that it

likely the reason for both significant temperature and amplitude effects. There were no

differences between high and low amplitudes of low frequency treatments; however, there were

differences between constant treatments and fluctuating treatments (Table 3-6). Therefore, the

"amplitude" effect was caused by constant temperature treatment and is not a true effect of

differences in amplitude of temperature change but more of an effect of frequency or the

presence or absence of temperature change(disc). The ANOVA conducted on the combination of

high and low frequency treatments (Table 3-5 and Fig. 3-6) revealed there to be no frequency or

trial effect on the number of days from hatch to first oviposition, with high frequency treatments

averaging 61.8 d & 7.3 (mean & SD) and low frequency treatments averaging 63.1 & 5.6.

Egg Pods / Female

The results of the ANOVA (Table 3-5) and mean comparisons (Table 3-6) performed on

high frequency treatments for egg pods / female revealed both significant temperature and

amplitude effects, and showed that only one treatment was significantly different from the

others, the 29.50C high frequency low amplitude treatment. Grasshoppers reared under this

treatment produced an average 2 egg pods / female compared to the average of about 6 egg pods

/ female for all other treatments (Table 3-6). It is likely that both the amplitude and temperature

effects were caused by this treatment.

Just as in high frequency treatments, the ANOVA for low frequency treatments (Table 3-

5) revealed significant temperature and amplitude effects. However, mean comparisons (Table 3-

6) revealed more than one significantly different treatment within the low frequency treatments.









Grasshoppers reared in 29.50C treatments produced on average 2.25 fewer egg pods / female

than those reared in 31.50C treatments. This difference was caused by more than one treatment

with all 29.50C treatments being significantly different from all 31.50C treatments but the

constant treatment (Table 3-6). The treatment with the fewest egg pods / female produced was

the 29.50C constant treatment with only 1.27 egg pods / female (Table 3-6). The amplitude effect

seen in low frequency treatments is essentially a frequency effect, as there were no differences

between low and high amplitude (Table 3-6). The only differences were between constant and

fluctuating treatments. Females in both constant temperature treatments produced significantly

fewer egg pods / females than many of their respective alternating treatments (Table 3-6). Unlike

high frequency treatments, there is more than one treatment causing differences between

temperature and amplitude. When both high and low frequency treatments were analyzed

together there were significant differences between the two frequencies (Table 3-5, Fig. 3-6)

with females reared in high frequency treatments producing 5.10 & 1.93 (mean & SD) egg pods /

female and those reared in low frequency treatments producing 4.00 & 1.48 (mean & SD) egg

pods / female.

Eggs / Pod

Both temperature and amplitude were found to be significant factors affecting the number

of eggs per pod in high frequency treatments (Table 3-5). Yet again, the only significant

treatment of the high frequency treatments was the 29.5oC low amplitude treatment with an

average of 82.46 eggs / pod compared to the average of near 60 eggs / pod for all other

treatments (Table 3-6). This treatments was responsible for both temperature and amplitude

effects by causing the eggs / pod average to be greater for 29.5oC treatments than 31.5oC









treatments and low amplitude treatments to average more eggs / pod than high amplitude

treatments (Table 3-6).

This is in contrast to low frequency treatments, where grasshoppers reared in 29.5oC

treatments produced fewer eggs per pod than 31.5oC treatments (Table 3-5 and 3-6). The

ANOVA conducted on low frequency treatments shows a significant amplitude effect (Table 3-

5). However, after examining the mean comparisons analysis (Table 3-6), it is clear that there are

no real differences between high and low amplitudes and that, just as before, the amplitude effect

is caused by differences between the constant and alternating temperature treatments, which is

more of a frequency effect. Females in both constant temperature treatments produced

significantly fewer eggs / pod than did their respective alternating treatments. Interestingly, the

29.50C low frequency high amplitude treatment had a higher mean number of eggs / pod value

than the 3 1.50C constant treatment and was not considered significantly different from any of the

31.50C treatments. When both high and low frequency treatments were combined, the ANOVA

revealed no differences between high and low treatments in the number of eggs / pod. Females

reared in high frequency treatments produced an average of 62.93 A 16.71 (Mean & SD) eggs /

pod and those reared in low frequency treatments produced 63.17 & 19.55 (Mean & SD).

Discussion

Conformity of laboratory treatments to field study results

Mean daily temperatures for laboratory treatments were based on mean daily body

temperatures obtained from the field. When able, the maj ority of free roaming grasshoppers

maintained mean body temperatures near 380C (Table 2-1). Thus, the initial temperature used to

simulate mean daily temperature to represent ideal or sunny weather conditions was 380C. Free

roaming grasshoppers on cloudy or rainy days often had body temperature near 300C (Table 2-










1). However, environmental chamber limitations necessitated the use of a slightly higher

temperature and so 330C was chosen to represent the mean daily temperature for days with

adverse weather conditions. In addition to the mean daily temperatures, frequency and amplitude

of temperature change were also modeled after data obtained from the field. During ideal

conditions on predominately sunny days, grasshoppers often experienced 0-1 large fluctuations

(>15% change) in body temperature. The low frequency treatments corresponded well with this

and had only 1 fluctuation in temperature per daytime period (Fig. 3-1). The fluctuation was

placed near the end of the day because more often then not, cloud cover or rain on such days

occurred during the afternoon hours. During predominately cloudy and rainy days, grasshoppers

experienced a mean of 3.67 large fluctuations per day. In laboratory treatments 5 fluctuations per

day were chosen which is not far from the average and likely provided more opportunity for

differences between different frequency treatments than 3 or 4 fluctuations might have. The

mean percent change in body temperature of grasshoppers for cloudy and rainy days was 22.34%

and 38.83% respectively. The initial amplitude of 80C corresponds well with those percentages

for cloudy days. In 31.50C high amplitude treatments, an 80C drop in temperature equals a 19%

change in temperature. In 29.50C high amplitude treatments the 80C drop equals a 21.6% change

in temperature. Low amplitude treatments were included to complete the study design and

provide a measurement of the effect of amplitude on the various parameters being investigated.

The low amplitude of 40C was chosen by arbitrarily halving the 80C amplitude.

Nymphal development time

Total nymphal development time in this study ranged from 35.4 45.0 days, with an

average of 38.4 days. The difference between average temperatures (29.50C and 31.50C) for

nymphal development time was 5.2 days for high frequency treatments and 1.6 days for low










frequency treatments. These differences are far less than many differences found in other studies

(Parker 1930, Whitman 1986, Giindiiz and Giilel 2002). Giindiiz and Giilel (2002) reported a

difference in development time of 10 d between 300C and 250C in S. gregaria. One reason these

results are different may be that the temperature difference between treatments was only 2oC

compared to the SoC or higher difference used in many other studies. Another possible reason is

that much lower temperatures, such as 20oC and 25oC, were used in these other studies. These

low temperatures may have been well below optimal for the grasshoppers being studied, while

the higher temperatures tested would have been within their optimal range. As mentioned before,

it may not be the temperature difference but where the temperatures fall with respect to optimal

temperatures for that species, that matters most. If an increase in 50C caused the new temperature

to cross from the suboptimal range into the optimal range, a greater effect would likely be seen

than if the 50C change occurred within the same range. This can be seen in a study by Gardner

and Thompson (2001) on development time in H. viridis where 50C increases in temperature

above 300C did not decrease development time. The lower temperatures tested in the current

study likely fell closer to optimal for this species than those of other studies and their respective

species, resulting in diminished responses.

Analysis of nymphal development time in high frequency treatments shows the only

significantly different treatment as being the 29.5oC low amplitude treatment. Willot (1992)

reports that for four species of Acrididae, optimum temperature for growth and development is

between 35oC and 40oC. Daytime temperatures within the 29.5oC treatments averaged 33oC,

which is just below the optimal range reported by Willot. Temperatures in the low amplitude

treatment ranged from 31oC to 35oC, while those in the high amplitude treatment ranged from

290C to 37oC. The body temperatures of grasshoppers reared in the low amplitude treatment









never reached above 3 50C, while body temperatures of those reared in high amplitude treatments

reached levels near the middle of the optimal range reported by Willot (1992). While Willot

(1992) does not report optimal body temperatures for S. amnericana, they are likely to at least be

within the 3 5-400C range, if not higher. In contrast to the 29.50C treatments, 3 1.50C treatments

had mean daily temperatures of 380C with low amplitude temperatures ranging between 360C

and 400C and high amplitude temperatures ranging between 340C and 420C, all of which fall

very near or within the theoretical optimal range. The fact that only the grasshoppers in the

29.50C low amplitude treatment had a slower development rate, suggests that amplitude of

temperature change may only be important for development time at suboptimal temperatures and

that it may be possible that just attaining 37oC for some portion of the day is beneficial for the

development of S. amnericana. Reporting that amplitude only effects nymphal development at

suboptimal temperatures entails that there should be a significant interaction between amplitude

and temperature. While the interaction was not considered significant, it was very close with a p-

value equal to 0.06.

A situation similar to that of high frequency treatments occurred in low frequency

treatments, with only one treatment being significantly different. However, instead of the 29.50C

low amplitude treatment being significantly different, it was the 29.50C constant temperature

treatment. Development time in the 29.50C low amplitude treatment falls between the 29.50C

constant treatment and the rest of the treatments, and is not statistically different from any of the

treatments. While every treatment received exactly the same number of degree days with respect

to their mean temperatures (calculated using data from environmental chambers), grasshoppers

reared in low frequency treatments experienced the high range of their temperature range for

longer uninterrupted periods of time than grasshoppers of high frequency treatments. This could









be the reason that nymphal development time in the 29.50C low amplitude treatment was

significantly different in high frequency treatments but not in low frequency treatments. The

exposure to uninterrupted periods of 350C likely provided a greater benefit than several shorter

periods of exposure. If this is true, it suggests that the duration (frequency) of uninterrupted high

temperatures is important for development.

The significant difference of the 29.50C constant treatment causes there to be both

significant amplitude and temperature effects. However, there are no differences between high

and low amplitude treatments. Therefore, the significant amplitude effect is an artifact of the

constant temperature treatments and is essentially a frequency effect manifested as a difference

between low frequency and no frequency or the presence or absence of temperature change. Data

was analyzed this way to preserve the complete block design. Had frequency been analyzed as

low frequency and no frequency the model would have lacked high and low amplitudes for the

constant treatments since there cannot be amplitude of change if there is no temperature change.

As in high frequency treatments, the amplitude factor (or in this case frequency) only had an

effect at the lower temperature (Table 3-2), suggesting an interaction between amplitude and

temperature. However, in contrast to the high frequency treatments, the amplitude X temperature

interaction is significant (Table 3-1). Again, the data suggest that nymphal development is

affected by fluctuating temperature only at suboptimal temperatures.

Because different frequency treatments were conducted as different trials, and the

treatment repeated between the two trials was found to be significantly different, any significant

frequency effects between high and low frequency treatments are confounded with any trial

effects. However, the data can still be analyzed if any effects of frequency between the two are

also attributed to differences between the trials. It was expected that low frequency might









develop faster because they spent a longer continuous duration of time at high temperatures,

even though they received the same number of degree days as that of high frequency treatments.

While nymphal development time of grasshoppers experiencing 3 1.5oC treatments remain

relatively the same between high and low frequency treatments, development time of 29.5oC

treatments is much shorter, 6.5 d shorter (Table 3-3), in low frequency treatments when

compared to high frequency treatments, suggesting again that temperature change only affects

nymphal development time at suboptimal temperatures. However, development times of

grasshoppers reared at a constant 29.5oC were shorter than that of grasshoppers reared in either

29.5oC high frequency treatment (Table 3-2). However, it was expected that constant temperature

treatments would exhibit the slowest development times. This, along with the idea that

fluctuations in temperature that allow the attainment of higher temperatures are more beneficial

for nymhpal development at lower mean temperatures, leads to the explanation that trial effects,

such as seasonal differences, are likely the cause for the faster development times seen in

constant temperature treatments when compared to high frequency treatments. However, if the

differences between high and low frequency treatments are due to trial affects, then another

question arises. Why is there no acceleration of nymphal development time in the 31.5oC

treatments between high and low frequency treatments? There are two possible explanations

assuming experimental error is not the cause. Trial effects are more prominent at suboptimal

temperatures or frequency effects are more prominent at suboptimal temperatures. The

differences due to a trial effect such as seasonality would not likely be affected by suboptimal

temperatures; therefore, the more plausible explanation is that differences in nymphal

development time between high and low frequency treatments are due to frequency, whose

effects are temperature dependent. This idea is supported by the fact that the same patterns are









seen when frequency treatments are analyzed separately, as discussed earlier. Still, whether or

not differences in nymphal development time between low and high frequency treatments should

be attributed to frequency or trial effects remains somewhat unclear.

Body size

Studies in the past have shown that thermoperiod can affect weight and size in insects

(Beck 1983). However, no such effect was observed in this study. Neither temperature

fluctuations nor differences in mean temperature affected body size in S. amnericana. Fielding

(2004) reported that body weight of M sanguinipes was lower at the low (210C and 240C) and

high (390C and 420C) extremes (with respect to optimal for M~ sanguinipes) of the temperatures

tested, relative to intermediate temperatures. In the current study, body weight was not recorded

and a lack of difference in size does not necessarily equate to a lack of difference in weight.

There may have been a possibility that there were differences in weight. However, temperatures

tested in the current study would not be considered extreme, and it is likely that there were no

differences in weight, just as there were no differences in size. As expected, there were

differences between the sexes, but when high and low frequency treatments were combined there

were significant differences between the two frequencies. No other facet of temperature affected

body size when the treatments were analyzed separately and, therefore, it is unlikely frequency

would have an effect and the differences between high and low frequency treatments were most

likely caused by trial effects. If trial effects affected a parameter not affected by temperature, it

may be that the differences between trials also affected other parameters more so than frequency.

This is in contrast to the nymphal developmental data, which suggest that differences between

high and low frequency treatments might actually be due to frequency.










Days to oviposition

In high frequency treatments the 29.50C low amplitude treatment was the only

significantly different treatment and there were no differences in sexual maturation between the

29.50C high amplitude treatment and the 31.50C treatments for the number of days to first

oviposition. The results suggest that amplitude only has an effect at suboptimal temperatures and

that mean body temperatures are likely not as important as maximum body temperatures

experienced. Just as in nymphal development time, reaching 370C when in the 29.50C high

amplitude treatment provided some benefit. The analysis conducted on low frequency treatments

for days to oviposition also revealed only one significantly different treatment. However, just as

for nymphal development time, the treatment was the 29.50C constant treatment and not the

29.50C low amplitude treatment. The reasons for this are likely the same as those given for

development time, having to do with the duration of continuous time spent in the high

temperature range of the treatments, suggesting a frequency effect. Together both sets of results

suggest that the number of days to oviposition is affected by both amplitude and frequency of

temperature change. Because the differences were only seen in the 29.50C treatments, it is

strongly suggested that these differences only occur at suboptimal temperatures.

The separate analyses conducted on both high and lower frequency treatments for days to

oviposition revealed results very similar to those for nymphal development time. This is not

surprising as both nymphal development time and sexual maturation are both developmental

parameters (together equaling total development time) and should be affected in a similar

manner. However, when both high and low frequency treatments were combined, the analysis

revealed the effect of frequency to be non-significant, contrary to what was found for nymphal

development time. Why this occurs is uncertain, but if the difference between nymphal









development times of both high and low frequency treatments is due to trial effects, it would be

very unusual for the frequency or trial effect to be absent in sexual maturation rate considering

the two parameters are closely related.

Giindiiz and Giilel (2002) reported that S. gregaria needed to reach a critical weight

before reproduction began, and that temperature did not affect this critical weight but only how

fast it was attained. It may be that 29.50C high frequency and constant treatments did not

maintain high enough temperatures for long enough periods of time for grasshoppers to quickly

reach this critical weight. One explanation for why differences were observed between 31.50C

treatments and 29.50C high frequency low amplitude and constant treatments and not the low

frequency 29.50C treatments is that the low frequency 29.50C treatments may have maintained

the high limit of the temperature range long enough to increase rate of weight gain. In contrast,

the less optimal treatments (29.50C high frequency low amplitude and 29.50C constant) would

have decreased rates of weight gain and delayed sexual maturation. Delayed sexual maturation

could have several negative effects on a grasshopper population. Grasshoppers with delayed

sexual maturation are more likely to die of disease or predation before being able to reproduce.

In colder environments, those that survive will have less time to lay eggs, and those eggs that are

laid will hatch later and be at a disadvantage compared to eggs which were laid earlier in the

season.

Measuring how many days it took female grasshoppers to reach sexual maturity posed a

problem in this study and for that reason the accuracy of its measurement may have suffered. All

other parameters were taken as cage averages nymphall development time and pods / female) or

as individual measurements of every available grasshopper or egg pod (body size and eggs /

pod). The number of days it took to first oviposition was recorded for only one individual per










cage, the first female to lay an egg pod. While it was rather simple to determine total nymphal

development time by recording when each grasshopper reached adult hood, it was impossible to

record the first oviposition date of every grasshopper in the cage. There was no way to determine

which female the egg pod came from. Therefore, only 40 of the 600 grasshoppers used in this

study were used for recording the number of days from hatch to oviposition.

Egg pods / female

Mean daily temperature had a significant effect on the number of egg pods laid per

female, with 31.50C treatments laying, on average, 2 more egg pods / female than in 29.50C

treatments for both high frequency and low frequency and constant temperature treatments.

Females in 29.50C constant treatment laid the lowest average with only 1.27 egg pods / female,

while 31.50C alternating treatments laid 5-6 egg pods / female. This is a huge difference with

respect to overall fecundity. These results are similar to those reported by Uvarov (1966a) on C.

pellucida, where females reared at 27oC laid an average of only one egg pod, as opposed to the

average of four egg pods laid by females reared at 37oC. While the results of the analysis clearly

show that temperature has a significant effect on egg pods / female, the effects of amplitude and

frequency of temperature change are much harder to identify. The results from high frequency

treatments are similar to those for developmental parameters with only the 29.50C low amplitude

treatment being significantly different from all others, suggesting the same inferences made

about nymphal development time might also be able to be made about fecundity. However, the

results from low frequency and constant treatments are very different from those for

developmental data with many treatments being different from each other.

The benefits on nymphal development time provided by the longer periods of continuous

high temperatures in low frequency treatments do not seem to apply to reproductive parameters.









The 29.50C low frequency high amplitude treatment has a much lower average than its high

frequency counterpart (Table 3-6). Additionally, the 29.50C low frequency treatments are

statistically different from the 31.50C treatments, something not seen in developmental data. This

suggests that continuous periods of the high range of temperatures have little if any effect on the

number of egg pods / female. The notion that temperature fluctuations have more of an effect at

suboptimal temperatures is also harder to see because most of the significant differences are

between mean temperatures. While the 29.50C constant treatment seems to have produced the

fewest egg pods / female, it is not considered statistically different from the other 29.50C

treatments or even the 31.50C constant treatment (Table 3-6). On the other hand, the 31.50C

constant treatment is significantly different from one of the 3 1.50C treatments. This is in contrast

to what has been suggested earlier, that temperature fluctuations are more influential at

suboptimal temperatures. These results have made it difficult to decipher the effects of

temperature, amplitude and frequency on the number of egg pods laid per female, especially

when one considers that high frequency treatments produced, on average, one more egg pod /

female than low frequency treatments and were considered statistically different from low

frequency treatments.

Eggs / pod

Only the higher values of raw data for the number of eggs / pod were within the range of

76-100 eggs per pod reported by Kuitert and Connin (1952). The means for each treatment, with

the exception of one, were never within this range. Why this is remains uncertain, but could be

due to a couple of factors. It may be that conditions in this study were not always optimal and

could have caused a reduction in the number of eggs per pod. Additionally, the reduced number










of eggs / pod might be an artifact of prolonged laboratory colonization. Regardless, there were

significant effects of temperature change on the number of eggs laid per pod.

The effects of mean temperature on the number eggs / pod are less obvious than those for

egg pods / female. In high frequency treatments the females reared in 29.50C treatments

produced more eggs / pod than those reared in 31.50C treatments, while the exact opposite

occurred in low frequency treatments (Table 3-6). The differences in amplitude and temperature

in high frequency treatments are due to the 29.5oC high frequency low amplitude treatments,

which is the only significantly different treatment. Females from the 29.50C high frequency low

amplitude treatment produced an abnormally large number of eggs / pod when compared to

females from the rest of the treatments. This suggests that suboptimal temperatures might cause

an increase in the number of eggs / pod. It was originally hypothesized that this increase in

number of eggs / pod was possibly a stress induced response to compensate for prolonged

development and the reduced number of egg pods laid, as this treatment was always the only

significantly different treatment of the high frequency treatments. However, such compensation

is not seen in equivalently affected low frequency treatments. This leads to the notion that

something may have been wrong with one or more elements of the experiment involving the

29.50C high frequency low amplitude treatment.

The results for eggs / pod for low frequency treatments were very similar to those for egg

pods / female, and many of the same conclusions can be drawn from the results. However, unlike

egg pods / female, the 29.50C constant treatment was considered statistically different from other

29.50C treatments and the 31.50C constant treatment (Table 3-6). This exemplifies the negative

effects that suboptimal temperatures can have on reproductive parameters. Again, because there

are no real differences between high and low amplitude treatments, the amplitude effect seen in









low frequency treatments is actually caused by the presence or absence of temperature change.

Unexpectedly, there was no frequency or trial effect on the number of eggs / pod when both high

and low frequency treatments were combined, even though there were differences in frequency

between constant and low frequency treatments. This was puzzling because there was some

amount of frequency or trial effects in every other parameter tested but days to first oviposition.










Table 3-1. ANOVA results for nymphal development time at different treatment combinations.

Treatment Source SS df MS F P


Amplitude
Temperature
Amp x Temp
Error


Amplitude
Temperature
Amp x Temp
Error


Amplitude
Frequency
Temperature
Freq x Amp
Temp x Amp
Temp x Freq
Temp x Freq x Amp
Error


17.65
109.97
13.99
39.66

19.90
15.60
39.47
32.03

11.39
123.87
61.70
6.59
32.33
48.66
0.16
59.96


17.65
109.97
13.99
3.30

9.95
15.60
19.73
1.78

11.39
123.87
61.70
6.59
32.33
48.66
0.16
2.50


5.34
33.27
4.24



5.59
8.99
11.09



4.56
49.58
24.70
2.64
12.94
19.48
0.06


0.0394
<0.0001
0.0620



0.0129
0.0077
0.0007



0.0431
<0.0001
<0.0001
0.1175
0.0014
0.0002
0.8049


High Frequency





Low Frequency +
Constant






High Frequency +
Low Frequency











Table 3-2. Mean comparisons of nymphal development time (in days) in high and low treatments
(HF = high frequency, LF = low frequency, HA = high amplitude, LA = low
amplitude). Means followed by different letters are significantly different at the 0.05
level (LSM, Tukey's adjustment).


Treatment
31.50C HF HA
31.50C HF LA
29.50C HF HA
29.50C HF LA
31.50C LF HA
31.50C LF LA
31.50C Constant
29.50C LF HA
29.50C LF LA
29.50C Constant


Mean & SD
37.68 & 2.45
37.91 + 1.09
41.05 & 0.98
45.02 & 2.25
37.26 & 1.45
35.40 & 1.34
36.26 & 1.48
35.42 & 0.68
37.86 & 1.56
40.54 & 1.31


Amplitude
HA
LA


Mean & SD
39.36 & 2.50 A
41.47 & 4.14 B


Temperature
31.50C
29.50C


Mean & SD
37.80 & 1.76 A
43.04 & 2.66 B


HA
LA
Constant


36.34 & 1.43 A
36.63 A 1.88 A
38.40 & 2.63 B


31.50C
29.50C


36.30 & 1.51 A
37.94 & 2.45 B










Table 3-3. Mean comparisons of nymphal development time (in days) in high (HF) and low (LF)
frequency treatments at fixed levels of amplitude and temperature (HA = high
amplitude, LA = low amplitude). Means followed by different letters are significantly
different at the 0.05 level (LSM, Tukey's adjustment).
Fixed Parameter Frequency Mean & SD
HA HF 39.37 & 2.50 A
LF 36.34 & 1.43 B
LA HF 41.47 &4.14 A
LF 36.63 A 1.88 B
29.5oC HF 43.04 & 2.66 A
LF 36.64 & 1.71 B
31.5oC HF 37.79 & 1.76 B
LF 36.33 A 1.63 B
All HF 40.42 & 3.48 A
All LF 36.48 & 1.62 B













Table 3-4. ANOVA results for body size at different treatment combinations.
Femur Length Overall Length
Treatment Source SS df MS F P SS df MS F P

Amplitude 4.18 1 4.18 3.26 0.0700 0.46 1 0.46 0.13 0.7200
High Frequency Sex 286.34 1 286.34 223.40 <0.0001 1497.47 1 1497.47 414.98 <0.0001
Temperature 0.00 1 0.00 0.00 0.9600 13.79 1 13.79 3.82 0.0540
Temp x Amp x Sex 6.94 1 6.94 5.42 0.0200
Error 119.20 93 1.28 335.59 93 3.61


Amplitude 2.81 2 1.41 0.36 0.7000 8.97 2 4.48 1.95 0.1400
Low Frequency + Sex 817.03 1 817.03 207.64 <0.0001 4102.27 1 4102.27 1786.85 <0.0001
Constant Temperature 1.71 1 1.71 0.43 0.5100 3.01 1 3.01 1.31 0.2500
Temp x Sex 21.47 1 21.47 9.35 0.0030
Error 991.57 252 3.93 578.54 252 2.30


High Frequency + Frequency 97.88 1 97.88 23.48 <0.0001 216.38 1 216.38 11.56 0.0008
Low Frequency Error 1146.62 275 4.17 5145.32 275 18.71














Table 3-5. ANOVA results for reproduction at different treatment combinations.

Days to Oviposition Egg Pods Female Eggs Pod
Treatment Source SS df MS F P SS df MS F P SS df MS F P


Amplitude
Temperature
4mp x Temp
Error

Amplitude
Temperature
Amp x Temp
Error

Amplitude
Frequency
Temperature
Freq x Amp
Temp x Amp
Temp x Freq
Temp x Freq x Amp


27.56 1 27.56 1.03
351.56 1 351.56 13.17
95.06 1 95.06 3.56
320.25 12 26.69

1705.75 2 852.88 12.12
988.17 1 988.17 14.04
523.58 2 261.79 3.72
1266.50 18 70.36
75.03 1 75.03 3.40
13.78 1 13.78 0.62
504.03 1 504.03 22.83
1.53 1 1.53 0.07
5.28 1 5.28 0.24
16.53 1 16.53 0.75
132.03 1 132.03 5.98


0.3300 10.78 1 10.78 24.47
0.0040 19.95 1 19.95 45.30
0.0800 19.95 1 19.95 45.30
5.29 12 0.44

0.0005 17.32 2 8.66 12.09
0.0020 32.03 1 32.03 44.71
0.0400 1.59 2 0.80 1.11
12.89 18 0.72
0.0800 4.57 1 4.57 9.57
0.4300 9.57 1 9.57 20.01
<0.0001 45.36 1 45.36 94.87
0.7900 6.27 1 6.27 13.12
0.6300 14.99 1 14.99 31.34
0.4000 0.18 1 0.18 0.31
0.0200 5.98 1 5.98 12.51

11.48 24 0.48


0.0003
=0.0001
<0.0001


0.0005
<0.0001
0.3500


0.0050
0.0002
<0.0001
0.0014
<0.0001
0.5508
0.0017


3767.46
6914.56
3652.88
41089.37

8441.19
9780.86
455.09
107602.00


3767.46
6914.56
3652.88
224.53

4220.59
9780.86
227.54
352.79


16.78 <0.0001
30.80 <0.0001
16.27 <0.0001


11.96 <0.0001
27.75 <0.0001
0.64 0.5200


High Frequency





Low Frequency
Constant






High Frequency
Low Frequency


1 5.68 0.02 0.8966


Error 529.75 24 22.07


139459.82 415 336.05













Table 3-6. Mean comparisons of reproductive data in high and low frequency treatments (H =

High Frequency, LF = low frequency, HA = high amplitude, LA = low amplitude).
Means followed by different letters are significantly different at the 0.05 level (LSM,

Tukey's adjustment).
Treatment Mean + SD Amplitude Mean + SD Temperature Mean + SD


31.50C HF HA 58.25 + 2.63 A HA 60.50 + 3.02 A 31.50C 57.13 + 2.10 A
31.50C HF LA 56.00 + 0.00 A LA 63.13 + 10.02 A 29.50C 66.50 + 7.67 B
29.50C HF HA 62.75 +0.96 AB

Days to 29.50C HF LA 70.25 +9.95 B

Oviposition 31.50C LF HA 56.50 +1.00 A HA 61.38 + 7.25 A 31.50C 62.58 + 7.88 A
31.50C LF LA 63.25 + 2.06 A LA 64.88 + 2.75 A 29.50C 75.42 + 15.99 B
31.50C Constant 68.00 + 11.55 A Constant 80.75 + 18.35 B
29.50C LF HA 66.25 +7.63 A
29.50C LF LA 66.50 +2.52 A
29.50C Constant 93.50 +14.80 B

31.50C HF HA 5.92 + 0.57 A HA 5.92 + 0.68 A 31.50C 6.21 + 0.71 A
31.50C HF LA 6.51 + 0.78 A LA 4.28 + 2.45 B 29.50C 3.98 + 2.16 B
29.50C HF HA 5.92 +0.87 A

Egg Pods 29.50C HF LA 2.04 + 0.28 B
/Female 31.50C LF HA 4.95 +0.97 AB HA 3.94 + 1.37 A 31.50C 4.56 +1.28 A
31.50C LF LA 5.58 + 0.39 A LA 4.07 + 1.68 A 29.50C 2.25 + 1.12 B
31.50C Constant 3.14 + 0.77 BC Constant 2.20 + 1.40 B
29.50C LF HA 2.93 +0.82 C
29.50C LF LA 2.55 +0.54 C

29.50C Constant 1.27 + 1.28 C


31.50C HF HA
31.50C HF LA
29.50C HF HA

Eggs! Pod 29.50C HF LA
31.50C LF HA
31.50C LF LA
31.50C Constant
29.50C LF HA
29.50C LF LA
29.50C Constant


59.12
59.27
62.79
82.46
68.11
66.68
57.11
58.20
55.98
40.58


60.88 + 14.37 A
64.43 + 18.16 B




63.73 + 20.22 A
62.63 + 18.97 A
53.23 + 19.38 B


31.50C
29.50C




31.50C
29.50C


59.22 + 13.01 A
70.40 + 20.56 B




64.13 + 19.20 A
54.35 +19.80 B


HA
LA
Constant















31.5HFHA


31.5HFLA


31.5LFLA


29.5LFHA


29.5LFLA


Figure 3-1.


Fluctuating laboratory temperature treatments for each mean temperature taken
from environmental chamber data loggers over a 24hr time period (resolution =
3min). Y axis is given in oC. (HF = high frequency, LF = low frequency, HA =
high amplitude, LA = low amplitude).





III
43 45 47


401


~X)i
LU
d


35 37 39 41


20

315

10 1









515
10

35
30


36 33 40 42 44 464 48

Nymphal Development (d)


Figure 3-2.


Nymphal development time (d) of high frequency treatments
of temperature and amplitude.


at fixed levels



































































SI I


40~












40








30







O x,


OO -
50




O


42 5


37 5


33


34.5 38 37 5 39
Nymphal Development (d)


40.5 42


Figure 3-3.


Nymphal development time (d) of low frequency treatments at fixed levels
of temperature and amplitude.





















































Nymphal development time (d) for high frequency and low frequency
treatments.


60

50

40



20

10


70

60


50-





20

1-


33 38


39 42

Nymphal Development (d)


Figure 3-4.












O HF
ALF


60



50



~40



3 30



20



10



0


Femur Length


Overall Length


Figure 3-5.


Mean (+ SD) femur and overall length at high and low frequency treatments
(HF = high frequency, LF = low frequency). Columns designated by a different
letter under their respective category are considered different at the 0.05 level.