The integument of the Queensland fruit fly,Ddacus tryoni (Frogg.)

Summary During the pupal stage of Dacus tryoni, the hypodermis of the larva is replaced by an imaginal generation of smaller cells. The hypodermal cells of the tergal glands on the fifth abdominal segment of the adult were examined with the electron microscope; they contain slender, membrane-limited bundles of hollow wax filaments that traverse the cuticle in branched pore canals. Outside the glandular areas, the pore canals are narrower. The cuticle of the adult undergoes its greatest increase in thickness soon after emergence; it becomes sclerotized gradually. No epicuticle was detected with either the light or electron microscopes.Early in adult development, bristles are formed over the general surface of the terga. Most of these are innervated by single, bipolar nerve cells, and have more or less enlarged trichogen cells that appear to secrete wax through pore-plates in the cuticle. The bristles in different regions of the abdomen range in function from pure sensory receptors to pure secretors. The sensory bristles on the tergal glands were examined with the electron microscope.For assistance with the electron microscopy, I thank Mr. Tony Webber and Miss Ann Miller of the Electron Microscopy Unit at Sydney University. — Supported by a C.S.I.R.O. Junior Post-Graduate Studentship.     

Development and ultrastructure of the fat body cells and oenocytes of the Queensland fruit fly,Dacus tryoni (Frogg.)

Summary Half-way through the larval period in Dacus tryoni, the fat body cells begin to accumulate protein in the form of granules. Early in the pupal period, both the fat body cells and oenocytes become free in the body cavity. Meanwhile, an imaginal generation of hypodermal cells, while in the process of displacing the larval hypodermis, gives rise to an imaginal generation of oenocytes. Soon after, imaginal fat body cells also appear. A few days after emergence, the larval fat body cells and oenocytes disintegrate and their imaginal equivalents expand to fill the body cavity.This paper also describes the ultrastructure of the larval and imaginal fat body cells and of the imaginal oenocyte. In all three, tubular invaginations of the plasma membrane occupy the peripheral cytoplasm. At most stages, the fat body cells contain a considerable quantity of slightly distended, rough endoplasmic reticulum, which suggests that when these cells are not sequestering protein, they are secreting it into the blood. The imaginal oenocytes are packed with smooth endoplasmic reticulum, which supports other evidence that they participate in the synthesis of cuticular wax.For assistance with the electron microscopy, I thank Mr. Tony Webber and Miss Ann Miller of the Electron Microscopy Unit at Sydney University. For the loan of some sectioned material, I am grateful to Dr. D. T. Anderson.     

Attractants for the gravid Queensland fruit fly Dacus tryoni

The vapours of certain pure chemicals, typical of ripe fruits, elicited characteristic components of ovipositional behaviour from gravid Dacus tryoni (Froggat) in an olfactometer: the flies walked and flew upwind to the source of the vapour and then probed with their ovipositors. A range of alcohols, acids, ketones and esters having 2–6 carbon atoms were effective (1 and 10% of iso-butyric acid, n-butyric acid, methyl butyrate, ethyl butyrate, 2-butanone, ethyl lactate and ethyl acetate; and 10% concentrations of ethanol and 2-propanone). The most effective were 4–6 carbon acids, esters and ketones. Behavioural threshold for n-butyric acid vapour at 26°C was obtained from a 5×10^–3% dilution in paraffin oil; maximum fly response occurred at about 200 times this concentration. Low concentrations of the 15-carbon sesquiterpene, -farnesene, were also very effective, despite its lower volatility. These results suggest that at least three different types of alfactory sensory neurones are involved in the identification of fruit attractants by gravid D. tryoni.     

Experimental analysis of the pattern of tethered flight in the Queensland fruit fly, Dacus tryoni

ABSTRACT. The tethered flight of the Queensland fruit fly, Dacus tryoni Frogg. (Diptera, Tephritidae), was investigated, and the duration of each flight during a 2-h experimental period was recorded. The pattern of flight was analysed, and related to the age, sex and origin of the specimens, and to the availability of host fruit during the rearing of the adults. The effect of adult crowding on the pattern of flight was also briefly examined. The results indicated that the origin of the flies had little effect on the pattern of flight; male and female flies showed different trends with respect to the proportion of short flights undertaken as the flies matured; and the availability of fruit had a marked effect on the pattern of flight in recently mature flies. These data are discussed with respect to the dispersive/non-dispersive movements of the flies postulated from previously documented field data. It is suggested that there is a characteristic pattern of tethered flight, which can be related to the absence of hosts in the immediate environment, and would be likely to lead to greater dispersal under natural conditions.     

Survival of step and ramp changes of temperature by the Queensland fruit fly, Dacus tryoni

ABSTRACT. Survival time at subzero temperatures is related to both long-term thermal history and the rate at which the insects are cooled. Insects cooled at one degree per hour survive for up to 3 times as long at a given test temperature than do insects cooled at 1C/min. Survival times are significantly shorter than times taken to freeze. Survival time at an extreme high temperature is related to long-term thermal history but the rate of heating makes no difference.     

Mark-recapture estimates of overwintering survival of the Queensland fruit fly,Dacus tryoni,in field cages

1. Four groups of 75 pairs each of marked adult Dacus tryoni that had emerged from field-collected guavas were liberated into cage-covered guava trees at the end of May, 1960 and sampled at weekly intervals to estimate the overwintering survivorship.
2. Estimates of numbers were made using three different methods as described byLeslie (1952) and one byJolly (1965). The Method A, approximate Method B andJolly models gave quite similar results although the Method A model was shown to be more sensitive to stochastic changes in the R/r ratios. The complete Method B model produced much smaller variances, due to its full utilization of the information in the data but it was very sensitive to small biases—even those caused by the restriction that recaptures must be whole numbers—and thus could not quite approach the exact expectations.
3. Although in some instances, statistical tests (Leslie , 1958) showed some significant deviations from randomness in recapture, these were not appreciable and did not affect the estimates.
4. When sufficient food was available, the flies in the field cages exhibited survival rates during the winter months of 1960 quite comparable to those of laboratory flies kept at room temperature with several exceptions: (a) there was a high mortality among the newly emerged flies during the initial week after release, (b) a high mortality during the last week of June and (c) a high mortality during the third week in July and the end of August which can be correlated with lack of rain, low humidity and warm daytime temperatures. Low temperatures apparently had little effect on survival.
5. There was no difference in the survival of males and females.
6. One replicate cage, whose flies had been anesthetized with carbon dioxide prior to release, exhibited significantly higher mortality rates than the other replicates.
7. Data from a fifth cage indicated that females were not inseminated during the winter months and did not produce mature eggs until the beginning of September.

     

The role of adult learning in the acceptance of host fruit for egglaying by the Queensland fruit fly, Dacus tryoni

Evidence presented indicates that exposure of wild Queensland fruit files, Dacus tryoni (Froggatt), to a given host fruit type (e.g. pear) for 3 days causes females to attempt oviposition to a greater degree in that fruit type than in other fruit types (e.g. tomato, grape). The effects of exposure to a particular fruit type proved reversible, suggesting that D. tryoni females were capable of learning. Females exposed to pear for 3 days appeared to retain the effect of such exposure on acceptance of tomato for up to 4 days but appeared to retain the effect on acceptance of grape for less than 2 days. The possible significance of prior experience of females with a particular fruit type on future ability to discriminate among varying-quality specimens of that type is discussed.

---

Résumé Les éléments apportés montrent que l'exposition de D. tryoni Froggat à un type de fruit déterminé (par exemple, la poire) pendant 3 jours, conduit la femelle à tenter de pondre à une plus haute fréquence dans ce type de fruit que dans d'autres (par example; tomate ou raisin). Les effects de l'exposition à un type particulier de fruit étant réversibles, ceci suggère que les femelles de D. tryoni sont capables d'apprentissage. Des femelles exposées à des poires pendant 3 jours ont conservé l'effet d'une telle exposition pendant 4 jours en présence de tomates, mais moins de 2 jours en présence de raisin. La discussion porte sur la signification possible de la connaissance antérieure d'un type particulier de fruit sur l'aptitude ultérieure des femelles à distinguer parmi des spécimens de ce type, mais de qualités différentes.

     

The dynamics of ovarian maturation and oocyte resorption in the Queensland fruit fly, Dacus tryoni, in daily-rhythmic and constant temperature regimes

The rate of oocyte development in constant regimes corresponded to the rates predicted by Pritchard's (1970) relationship which indicates that the lower temperature threshold is 13.5dgC. Rates in fluctuating regimes indicated that development occurs at certain temperatures below the constant threshold if these are alternated with higher temperatures on a daily basis
In certain fluctuating regimes, oocyte development occurred at a rate in excess of 3.5% per day and maturation proceeded to completion. In such regimes resorption occurred only if egg laying was prevented; when this happened the terminal oocytes usually remained intact and the penultimate ones were resorbed. In other fluctuating regimes, oocyte development proceeded at a rate of less than 3.5% per day and no maturation was achieved because the most advanced oocytes were resorbed-either before or during the vitellogenic phase.
The 3.5% threshold corresponds to a value of 2 day-degrees per day (2DDPD) above 13.5C. Resorption, but no development, was observed in partly mature flies around O DDPD, whereas neither occurred in a still colder regime. These results are shown to be in accord with Fletcher's (1975) field data which are also used to discuss the significance of resorption. Finally, the DDPD relationship is used to define those daily temperature profiles which may permit maturation in the field.     

Genetic and molecular markers of the Queensland fruit fly, Bactrocera tryoni

Twenty-six microsatellite markers, along with two restriction fragment length polymorphism (RFLP) markers and three morphological markers, have been mapped to five linkage groups, corresponding to the five autosomes of the Queensland fruit fly, Bactrocera tryoni. All these molecular and genetic markers were genotyped in three-generation pedigrees. Eight molecular markers were also localized to the salivary gland polytene chromosomes by in situ hybridization. This provides a substantial starting point for an integrated genetic and physical map of B. tryoni.     

The effect of acclimation on mating frequency and mating competitiveness in the Queensland fruit fly, Dacus tryoni, in optimal and cool mating regimes

Mating frequency in groups of Dacus tryoni which had been either warm-acclimated or cold-acclimated were compared in temperature regimes ranging from just above mating-threshold to optimal. Cold-acclimation appeared to suppress initial mating ability of mature insects of both sexes to an extent which depended upon the acclimation regime used. The most favourable cold-acclimation regime produced flies which in certain circumstances were able to mate at an initial frequency similar to that of warm-acclimated flies. In no mating regime was initial mating significantly more frequent in any cold-acclimated group than it was in any warm-acclimated group. In most cases warm-acclimated flies in a given regime mated at high frequency for 1–2 days, whereas the cold-acclimated flies mated at low frequency for a greater number of days. In all cases, cold-acclimated flies accumulated a similar or significantly lower total number of matings than warm-acclimated groups. In experiments in which both warm-acclimated and cold-acclimated males competed for cold-acclimated females, the warm-acclimated males always out-competed the cold-acclimated males in two mild (near optimal) regimes. In a relatively harsh (near torpor threshold) regime, there was no significant difference in the competitive abilities of cold-acclimated and warm-acclimated males. The relevance of these results to possible acclimation procedures used in control campaigns involving release of sterile males is discussed.     

Genetic manipulation of the circadian clock's timing of sexual behaviour in the Queensland fruit flies, Dacus tryoni and Dacus neohumeralis

ABSTRACT. Mating in Dacus tryoni is restricted to dusk, whereas that of a sibling species, Dacus neohumeralis , occurs in the middle of the day. The timing of sexual behaviour in both species is determined by an interaction between a circadian clock and light intensity. In D. tryoni peak mating responsiveness is at the time of dusk, and the optimal light intensity for mating is approximately 91x. In D.neohumeralis peak responsiveness is in the middle of the day, and the optimal light intensity for mating is greater than 10 000 lx. The two species were crossed and the time of mating and response to light intensity of F₁, F₂ and backcross progeny determined. The circadian clock set a mating phase ('gate') as narrow in F₁ flies as in their parents, suggesting the circadian timing mechanism to be common between the two species. The results indicate that the genetic mechanism controlling timing is independent of that controlling response to light intensity, and that both genetic mechanisms are complex.     

The effect of photoperiod on the circadian rhythm of mating responsiveness in the fruit fly, Dacus tryoni

ABSTRACT. The daily fluctuation in mating responsiveness of Dacus tryoni was recorded through a range of light cycles. Evidence was obtained of strict control of the fluctuation by a circadian clock. The phase setting of the fluctuation in the different light cycles was in conformity with Aschoff's (1965) generalizations about circadian rhythms, and had no relationship to overt sexual behaviour occurring during the cycles. The amplitude of the fluctuation in mating responsiveness varied between cycles, suggesting possible variation in the amplitude of the oscillation of the circadian clock.     

Times for recovery from cold-torpor in the Queensland fruit fly Dacus tryoni: the relation to temperature during and after chilling

Recovery time after experience of a given minimum temperature below torpor threshold is related to the value of that minimum, the length of time spent at that minimum, and the temperature prevailing during the recovery period above torpor threshold. A model can predict recovery time for flies experiencing a given temperature fluctuation if the length of time spent at the minimum is expressed as a proportion of LE₅₀ at that minimum.The model has applications in defining the optimal protocol for chilling insects for use in the Sterile Insect Release Method. The model was confirmed by experiments showing that it is likely that flies will recover from non-lethal frosts before ant predators become active.

---

Résumé Le temps de récupération après avoir subi une température minimal située au-dessous du seuil d'engourdissement dépend de la valeur de ce minimum, du temps passé à ce minimum, et de la température au-dessus du seuil d'engourdissement pendant la période de récupération. Un modèle mathématique permet d'estimer le temps de récupération après avoir subi une chute de température déterminée, en fonction du temps passé au minimum thermique exprimé comme une fraction du LE₅₀ (temps nécessaire pour tuer 50% des mouches) à ce minimum.Ce modèle s'est trouvé étayé par des observations montrant qu'il est probable que les mouches se remettent des gelées sublétales avant la reprise d'activité des fourmis prédatrices. Ce modèle peut être utilisé pour définir les conditions optimales de refroidissement des insectes utilisés lors de la libération d'individus stériles.

     

Fecundity,fertility and reproductive recovery of irradiated Queensland fruit fly Bactrocera tryoni

Pupae of the Queensland fruit fly or Q‐fly Bactrocera tryoni (Froggatt) are irradiated routinely to induce reproductive sterility in adults for use in sterile insect technique programmes. Previous studies suggest that adult sexual performance and survival under nutritional and crowding stress are compromised by the current target dose of radiation for sterilization (70–75 Gy), and that improved mating propensity and survival under stress by irradiated males may be achieved by reducing the target sterilization dose without reducing the level of induced sterility. This raises the question of the amount by which the irradiation dose can be reduced before residual fertility becomes unacceptable. The present study measures the levels of residual fertility in male and female irradiated Q‐flies at different irradiation doses (20, 30, 40, 50, 60 and 70 Gy), and investigates the possibility that fecundity and fertility increase between 10–15 and 30–35 days post emergence. Male flies require a higher dose than females to induce sterility, with no residual fertility found in females irradiated at doses of 50 Gy or above, and no residual fertility found in males irradiated at doses of 60 Gy or above. Irradiated females are more fecund at 30–35 days post emergence than at 10–15 days. However, fertility does not increase between 10 and 15 days post emergence and 30–35 days, even at doses below 50 Gy. The present study shows that there is scope to reduce the target sterilization dose for Q‐flies below that of the current dose range (70–75 Gy) at the same time as retaining an adequate safety margin above radiation doses at which residual fertility can be expected.     

Upwind anemotaxis in response to cue-lure by the Queensland fruit fly, Bactrocera tryoni

Movements of mature male Bactrocera tryoni (Froggatt) (Diptera: Tephritidae) were observed individually in a wind tunnel under conditions of ‘cue-lure with wind’, ‘cue-lure with no wind’, ‘wind only’ and ‘no wind or cue-lure’. Further observations were made using a dense foliage array in the wind tunnel and a structured plume of cue-lure. Patterns of walking or flying were essentially the same in all of the first four treatments except that in the ‘cue-lure with wind’ treatment, over half of the flies moved in a consistent track upwind for at least 400 mm at some time during the first 5 min of observation. With clean wind, only 10% of the flies did this. The result was that mean net upwind displacement after 5 min in the ‘cue-lure with wind’ treatment significantly exceeded that in the other three treatments, the results of which did not differ significantly from each other. The upwind tracks were accomplished by either walking or flying (with or without stops) or by a combination of both. When the wind tunnel was filled with a dense foliage array, the results with cue-lure laden wind were similar to those obtained with the equivalent treatment without foliage, except that upwind tracks were predominantly in short stages. When flies were exposed to a structured plume of cue-lure odour (without foliage present), they did not apparently alter their behaviour on leaving or entering the plume, but some did make consistent upwind tracks while they were in the plume.     

Allozyme variation in the fruit flies Dacus tryoni and Dacus neohumeralis (Tephritidae)

Electrophoretic variation in four specific proteins, an alcohol dehydrogenase, and an octanol dehydrogenase from adult body homogenates, an esterase from adult heads, and a tyrosinase from larval hemolymph, is described for laboratory populations of two sibling species, Dacus tryoni and Dacus neohumeralis. For each enzyme, a number of test crosses between different electrophoretic forms provided evidence that the observed variation was due to segregation of alleles at one structural gene locus.     

Oviposition responses of Queensland fruit fly (Bactrocera tryoni) to mineral oil deposits on tomato fruit

Behavioural responses of wild and laboratory‐culture females of Queensland fruit fly, Bactrocera tryoni (Froggatt) (Diptera: Tephritidae), to mineral oil deposits on tomato fruit dipped in aqueous oil emulsions were assessed in a no‐choice test and three choice tests. The oils were two commercial products used to manage plant pests and diseases, Ampol D‐C‐Tron NR and SK EnSpray 99, one distillation fraction of the base oil of the former, and four distillation fractions of the base oil of the latter. The initial and final boiling points of the fractions were equivalent to those of n‐alkanes with chain lengths of C20–22 (Ampol), and C21–C23, C22–23.5, C22–24, and C22–24.5 (SK). For both fly types in the no‐choice test, numbers of punctures and eggs per fruit declined strongly as concentrations of the nC20–22 Ampol fraction in emulsions rose from 0.25 to 2% (vol/vol). Fly type affected the extent of responses but there was no significant interaction for fly type*oil concentration. Responses of laboratory‐culture females in the choice tests also declined as concentrations of SK and the four fractions of its base oil in emulsions rose from 0 to 0.25%. The SK nC22–24 and nC22–24.5 fractions had least impact. Responses of laboratory‐culture flies to 0.5% emulsions of the nC20–22 Ampol fraction and the nC21–23 SK fraction in choice tests were not significantly different. Likewise, responses of laboratory‐culture flies to 0.5% emulsions of the two commercial products were not significantly different. Emulsifier type did not affect numbers of punctures or eggs per fruit in choice responses of laboratory‐culture flies to 0.5% emulsions of the Ampol nC20–22 fraction or 0.5% emulsions of the SK nC21–23 fraction. If the equivalence of no‐choice and choice responses in the laboratory were to hold in the field, then unsprayed ‘sacrificial’ plants would not be necessary and oil emulsions could be used as cover sprays.     

Evaluation of yeasts in gel larval diet for Queensland fruit fly,Bactrocera tryoni

Queensland fruit fly, Bactrocera tryoni (“Q‐fly”), is Australia’s most economically important insect pest of horticultural and commercial crops especially in the eastern regions. The sterile insect technique (SIT) has been adopted as an environmentally benign and sustainable approach for management of Q‐fly outbreaks. High‐performance larval diets are required to produce the millions of flies needed each week for SIT. Yeast products contribute amino acids (protein) to fruit fly larval diets, as well as carbohydrate, fat and micronutrients, but there can be substantial variation in the nutritional composition and suitability of yeast products for use in larval diets. Gel larval diets have recently been developed for large‐scale rearing of Q‐fly for SIT, and composition of these diets requires optimization for both performance and cost, including choice of yeast products. We assessed performance of Q‐flies reared on gel larval diets that contained debittered brewer’s yeast (Lallemand LBI2240), hydrolysed yeast (Lallemand FNILS65), inactivated brewer’s yeast (Lallemand LBI2250) and inactivated torula yeast (Lallemand 2160‐50), including blends. Q‐flies performed poorly when reared on diets containing only or mostly hydrolysed yeast in terms of pupal number, pupal weight and percentage of fliers. Performance was also poor on diets containing high proportions of torula yeast. Overall, debittered brewer’s yeast is recommended as the best option for Q‐fly gel larval diet, as it is cheap, readily available, and produces flies with good performance in quality control assays. Inactivated brewer’s yeast produced flies of comparable quality with only a modest increase in cost and would also serve as an effective alternative.     

Behavioural responses of female Queensland fruit fly, Bactrocera tryoni, to mineral oil deposits

Behavioural responses of Queensland fruit fly, Bactrocera tryoni (Froggatt) (Diptera: Tephritidae), females to fruit dipped in water and fruit dipped in 0.5% (vol/vol) aqueous emulsions of a mineral oil were determined and analysed. The mineral oil was an nC20–22 distillation fraction of the base oil used to produce an nC23 horticultural mineral oil. Females caged with oil‐treated fruit had significantly longer prelanding intervals than females caged with water‐dipped fruit. The latter was attacked immediately or shortly after being caged with flies whereas some oil‐dipped fruit was not attacked within 180 min. The percentage of landings that led to oviposition on water‐ and oil‐treated fruit were 58 and 13%, respectively, and the percentages ovipositing after probing were 74 and 25%, respectively. Likewise, average times spent probing were 7 vs. 31 s whereas average times spent ovipositing were 321 vs. 223 s. Females spent less than half as much time on oil‐treated fruit than on water‐treated fruit. Transition probabilities of rejection, when applied to the behaviour sequence indicated that oil‐treated fruits are about nine times less likely to be infested with B. tryoni.     

Survival of repeated frosts by the Queensland fruit fly, Dacus tryoni: experiments in laboratory simulated climates with either step or ramp fluctuations of temperature

Ability to survive exposure to single or repeated periods at a subzero temperature is related to the temperature experienced, whether it is approached quickly or slowly, the time for which it prevails and the interval between exposures.The severity of any low temperature can be expressed in terms of LE₅₀ (time required to kill 50% of individuals with one exposure). Minima enduring for 35% LE₅₀ do not cause any mortality, even when repeated daily. Minima enduring for 44% LE₅₀ cause ca 14% mortality on the first occurrence but no more if repeated at 3 day intervals, but 3% more per occasion if repeated daily. Minima enduring for 88% LE₅₀ cause ca 40% mortality on first occurrence and an equal amount at each recurrence even if each is 7 days apart. Minima enduring for 125% LE₅₀ and over 150% LE₅₀ cause respectively ca 80% and 100% mortality respectively on first occurrence.The daily maximum temperature (in the range 15°C to 25°C) appears to have little relevance to the mortality caused by a repeated minimum of -5°C. Flies of different ages have a similar ability to survive a repeated minimum of -6°C for up to six exposures, but thereafter old flies are more susceptible than young ones.These results can be related to mortality caused by frosty conditions in the field so long as the time spent at the minimum is known and the temperature on the ground can be measured or calculated.

---

Résumé L'aptitude à survivre à une ou plusieurs expositions à des températures inférieures à zéro dépend à la fois de leur sévérité et du laps de temps entre les expositions. L'effet d'une température minimale déterminée dépend de sa valeur, de sa durée et du type de refroidissement: brutal (step) ou progressif (ramp).Les effets de toute température minimale peuvent être exprimés en termes de LE₅₀ (temps nécessaire pour obtenir une mortalité de 50% avec une exposition unique). Lorsque le temps d'application de la température minimale correspond à 35% du LE₅₀, il n'y a pas de mortalité, de même dans le cas de répétitions quotidiennes. Lorsque le minimum est subi pendant 44% du LE₅₀, il provoque environ 14% de mortalité après la première exposition, avec 3% supplémentaire après chaque exposition quotidienne, mais l'augmentation est nulle si l'exposition au froid n'a lieu que tous les 3 jours. Lorsque le minimum est subi pendant 88% de LE₅₀, la mortalité après la première exposition est d'environ 40% et de même importance à chaque nouvelle exposition, même si elles sont espacées de 7 j. Lorsque la durée d'exposition au minimum correspond à 125% et à 150% de LE₅₀, les mortalités sont respectivement d'environ 80% et 100% à la première exposition.La température maximale quotidienne, entre 15 et 25°C, semble avoir peu d'influence sur la mortalité provoquée par des expositions répétées à-5°C. Les mouches d'âges différents résistent de la même façon jusqu'à 6 expositions répétées à-6°C, mais au-delà les mouches âgées sont plus sensibles que les jeunes.Dans la mesure où, dans la nature, la durée d'exposition à la température minimale est connue et où la température dans le sol peut être calculée ou mesurée, ces résultats peuvent permettre d'interpréter la mortalité provoquée par le gel.