Biological Features and Population Growth of Two Southeastern European Tribolium confusum Jacquelin du Val (Coleoptera: Tenebrionidae) Strains.

A study of the biological features and the potential population growth between two laboratory strains of the confused flour beetle, Tribolium confusum Jacquelin du Val (Coleoptera: Tenebrionidae) from Greece and Serbia is conducted on cracked barley and cracked white rice. The results show that, at a species level, T. confusum is able to complete development on cracked barley but not on cracked white rice. Therefore, cracked white rice proves to be an unsuitable commodity for T. confusum. Larval development on cracked barley is significantly shorter for the Serbian compared to the Greek strain (37.7 and 49.7 days, respectively), but pupal development does not differ between the two strains (6.2 days for both strains). Additionally, male longevity does not differ between the Greek and Serbian strains (144.4 and 151.4 days, respectively), while female longevity is significantly shorter for the Serbian (151.7 days) compared to the Greek strain (186.6 days). Fecundity does not differ between the two strains (11.3 and 17.7 eggs/female for the Greek and the Serbian strain, respectively), whilst survival is higher for the Serbian strain on both tested commodities. The values of the net reproductive rate, the intrinsic rate of increase and the finite rate of increase on cracked barley are significantly higher for the Serbian (7.27 females/female, 0.025 female/female/day and 1.026, respectively) compared to the Greek strain (2.91 females/female, 0.014 females/female/day and 1.014, respectively). It therefore is expected that different strains of T. confusum may exhibit variable phenology as well as potential population growth. Additionally, we expect our results to have bearing on the management of this species.


Introduction
The confused flour beetle, Tribolium confusum Jacquelin du Val (Coleoptera: Tenebrionidae) is a long-lived species that can seriously and rapidly infest stored-products [1][2][3]. It infests cereal grains or their products and is usually found in mills, bakeries, warehouses and pet stores [4,5]. It is regarded as a secondary colonizer since it cannot easily develop in sound grain kernels [6,7]. It can also damage dried plants or fruits, dairy products, oilseeds, nuts, animal feed, cotton and spices [8,9]. Adults are approximately 3.5 mm long and have a reddish brown color, while larvae can reach a 6-7 mm length [9].
Insects 2020, 11, 218 3 of 11 Laboratory and kept under the same conditions as the Greek strain for one generation. The founding individuals of the Greek and Serbian strains were originally collected from Greek and Serbian storage facilities, respectively.

Commodities
Clean and free of infestation and pesticides hulless barley and white rice were used in the tests. Prior to experimentation, the moisture of the tested grains was adjusted to 13.5 ± 0.5% by heating them in an oven at 50 • C or by adding distilled water [53,54]. The moisture was measured by a calibrated moisture meter (mini GAC plus, Dickey-John Europe S.A.S., Colombes, France).

Development and Survival of Immatures
Both grains were cracked by an electric grinder Multi 600 (Izzy, Benroubi S.A., Amaroussion, Greece). Subsequently, they were sieved in a standard testing sieve having openings of 2.36 mm (Advantech Manufacturing Inc., New Berlin, WI, USA) and in one with 2.00 mm openings (Retsch GmbH, Haan, Germany) to remove dusts and obtain uniform particles of cracked grains. Taken from each strain, 100 unsexed approximately 7-day-old adults were transferred to two 250 mL-glass vials, each containing 125 g of fresh, pre-sieved hard wheat flour, and then placed into an incubator set at 30 • C, 65% relative humidity and continuous darkness for one day. Afterwards, flour from each vial was sieved to remove adult individuals and laid eggs, using standard testing sieves with 0.85 and 0.25 mm openings (Advantech Manufacturing Inc., New Berlin, WI, USA). The eggs found on the mesh of the sieve with 0.25 mm openings were gently and separately transferred to petri dishes (5.5 cm diameter, 1 cm height) with a fine brush (Cotman 111 No 000, Winsor and Newton, London, UK) that did not contain food. The lids of the dishes had a central circular opening (1.5 cm diameter) which were covered by muslin gauze allowing adequate ventilation inside the dishes. Then, dishes were placed inside incubators set at 30 • C, 65% relative humidity and continuous darkness and inspected every 24 h for duration and survival under a SZX9 Olympus stereomicroscope with 57× total magnification (Bacacos S.A., Athens, Greece). Regarding each strain and commodity, the experiment was initiated with a cohort of 100 eggs. Using a fine brush (Cotman 111 No 000, Winsor and Newton, London, UK), newly hatched T. confusum larvae were very carefully placed separately inside dishes (5.5 cm diameter, 1 cm height) that contained 1 g cracked barley or cracked white rice. All quantities of 1 g were weighed with a Precisa XB3200D compact balance (Alpha Analytical Instruments, Gerakas, Greece). The dishes were placed into incubators set at 30 • C, 65% relative humidity and continuous darkness for the entire experimental period. The duration and survival of larval and pupal stages were inspected every 24 h under the aforementioned stereomicroscope. All larvae of both strains died in cracked white rice before reaching the pupal stage, therefore the experiment was continued with strains on cracked barley.

Adult Longevity and Reproductive Capacity
One-day-old pupae were sexed following Halstead [55]. Each newly emerged pair (a male and a female) was placed in a different dish (5.5 cm diameter, 1 cm height) that contained 1 g of cracked barley then placed into incubators set at 30 • C, 65% relative humidity and continuous darkness. Adult longevity and female fecundity were inspected daily until death of the male and female. Finally, the progeny sex ratio was calculated for each strain on the basis of 100 randomly selected pupae originating from eggs obtained from the coupled females.

Statistical Analysis
The data on the larval and pupal development, female and male longevity, as well as on the female fecundity were examined via the Shapiro-Wilk normality test which indicated departure from a normal distribution. Therefore, all pairwise comparisons were done using the Mann-Whitney Rank Sum Test. The Kaplan-Meier method was used to estimate the survival curves of the two strains fed on cracked barley and cracked white rice. Additionally, the Kaplan-Meier estimate was used to derive Insects 2020, 11, 218 4 of 11 the mean survival times and their 95% confidence intervals (C.I.). All analyses were done using the statistical package SigmaPlot 14.0 [56].
Concerning each T. confusum strain fed on cracked barley, the following demographic parameters were calculated [36]: the net reproductive rate R 0 = (l x × m x ) , i.e., the per capita rate of progeny production in an interval equal to cohort study interval (l x corresponds to the cohort survival to age x and m x the age specific fecundity); the intrinsic rate of increase (r m ) (e r m ×x × l x × m x ) = 1 , i.e., the rate of natural increase in a closed population (that is subjected to constant age-specific schedules of fertility and mortality for a long period); the finite rate of increase λ = e r m , i.e., the rate at which the population will increase in each time step, the mean generation time T = ln R 0 r m , i.e., the time required for the population to increase by a factor equal to the net reproductive rate and the doubling time DT = ln 2 r m , i.e., the time required for the population to double. Significant differences between demographic parameters were tested via the superposition of 95% C.I. (Wald test), which were obtained by bootstrapping in R [57]. Particularly, for each treatment we sampled ten thousand individuals to obtain the 95% confidence intervals. When T. confusum fed on cracked white rice, no demographic analysis was conducted as both strains did not complete their development.

Results
The biological features of the T. confusum strains fed on cracked barley are presented in Table 1. Larval development was significantly shorter for the Serbian compared to the Greek strain (37.7 and 49.7 days, respectively), but pupal development did not differ between the two strains (6.2 days for both strains). Additionally, male longevity did not differ between the Greek and Serbian strains (144.4 and 151.4 days, respectively). While female longevity was significantly shorter for the Serbian (151.7 days) compared to the Greek strain (186.6 days), fecundity did not differ between the two strains (11.3 and 17.7 eggs/female for the Greek and Serbian strain, respectively). When both T. confusum strains fed on cracked white rice they did not complete their development. Table 1. Duration of immature development, female and male longevity in days (mean ± SE, median) and female fecundity (eggs/female) of two strains of Tribolium confusum (mean, 95% C.I.) fed on cracked barley. Medians within a column followed by the same letter are not statistically different (Mann-Whitney Rank Sum Test at a = 0.05). The survival curves for the Serbian and Greek strains on both of the tested commodities are presented in Figure 1. Mean survival times were 120.8 and 21.1 days for the Greek strain when fed on cracked barley and cracked white rice, respectively, where the corresponding values were 169.9 and 41.4 days for the Serbian strain (Table 2). Based on the 95% C.I. criterion, mean survival times were longer for the Serbian strain. The survival curves for the Serbian and Greek strains on both of the tested commodities are presented in Figure 1. Mean survival times were 120.8 and 21.1 days for the Greek strain when fed on cracked barley and cracked white rice, respectively, where the corresponding values were 169.9 and 41.4 days for the Serbian strain (Table 2). Based on the 95% C.I. criterion, mean survival times were longer for the Serbian strain.  The values of the net reproductive rate, the intrinsic rate of increase and the finite rate of increase on cracked barley were significantly higher for the Serbian (7.27 females/female, 0.025 female/female/day and 1.026, respectively) compared to the Greek strain (2.91 females/female, 0.014 females/female/day and 1.014, respectively) ( Table 3). The mean generation time, as well as the doubling time did not differ between the two strains (76.8 and 78.1 days, 41.2 and 27.7 days, for the Greek and Serbian strains, respectively).  The values of the net reproductive rate, the intrinsic rate of increase and the finite rate of increase on cracked barley were significantly higher for the Serbian (7.27 females/female, 0.025 female/female/day and 1.026, respectively) compared to the Greek strain (2.91 females/female, 0.014 females/female/day and 1.014, respectively) ( Table 3). The mean generation time, as well as the doubling time did not differ between the two strains (76.8 and 78.1 days, 41.2 and 27.7 days, for the Greek and Serbian strains, respectively).

Discussion
Our study reveals several clear findings on the performance of T. confusum. This species did not complete the development on cracked white rice in contrast to cracked barley. Previous reports have documented that diet affects several biological features of T. confusum. Adult emergence ranged from 3.8 to 62.4 individuals on various commercial types of milk powder within a period of 65 days [58]. A top patent flour enriched with bran and germ enhanced fecundity of T. confusum non-enriched patent flour (3.2 eggs/day/female) or whole wheat flour (7.0 eggs/day/female) [59]. Among several types of diets, wheat flour containing 5% (w/w) brewer's yeast provided the shortest interval (17 days) at which T. confusum larvae became pupae and led to heavier pupae (3 mg) [60]. Similarly, Prus et al. [61] reported that the absence of yeast from wheat flour resulted to reduced fecundity and reproductive capacity of T. confusum. By testing different combinations of cracked wheat kernels, wheat starch and crude α-amylase inhibitor, Warchalewski et al. [62] altered the larva-to-adult T. confusum period, ranging from 22.1 to 46.3 days, and the percentage of adult emergence, ranging from 56.0 to 76.0%. Grain commodities, e.g., spikelets of emmer and spelt, exhibited different levels of susceptibility to T. confusum infestations with the former favoring its population growth more than the latter [63]. However, non-grain commodities, such as soybean and pigeon pea flours, reduced fecundity, extended the larval or pupal developmental period and reduced the survival of adults compared to wheat flour [64]. Furthermore, they are able to cause variable levels of weight losses to different varieties of stored-maize ranging from 2.56 to 21.63 g maize/100 g maize six months post-treatment [65].
Our study also provides evidence that different strains of T. confusum may exhibit variable biological and demographic performances. Although this species did not complete development when fed on cracked white rice, the Serbian strain showed a higher survival than the Greek strain. Moreover, larvae of the Serbian strain exhibited faster larval development on cracked barley. Additionally, the survival of the Serbian strain was higher on this commodity. The intrinsic rate of increase is an indicator of the potential growth of an insect population [34,66,67]. As this value is higher for the Serbian strain, we expect to increase its population faster compared to the Greek strain. This fact also is depicted on the values of the finite rate of increase and the net reproductive value which are also higher for the Serbian strain. However, the mean generation time did not differ between the two strains, which may appear somewhat surprising. This parameter represents the average time for a population to increase by a factor equal to the net reproductive rate [68]. Therefore, although the Serbian strain has a shorter larval development, a longer adult time period may be needed to reach its higher net reproductive rate compared to the Greek strain.
The fact that both tested strains did not complete development on cracked white rice proves that this commodity is unsuitable for T. confusum. This species can only spend a short time period on cracked white rice, as this commodity can act only as a temporary host for it. This issue could be attributed to the nutritional value of this commodity. A portion of its nutrient ingredients such as protein, fat, vitamins and minerals are lost during the milling process, which includes the removal of bran and germ from the kernels [69,70]. Development and fecundity of T. confusum are increased when thiamine, riboflavin and niacin exist in its diet [59]. However, these vitamins are substantially reduced during processing of rice [70]. Particularly, in white rice, the amount of riboflavin ranges between 0.2 and 0.6 µg/g [69] which is lower than the minimum required level for the development of T. confusum larvae (1-2 µg/g) [59,71]. To contrast, both strains completed their cycle on barley that contained riboflavin ranging between 1.5 and 2.85 µg/g [72,73]. Thus, white rice should be considered as an unsuitable commodity for T. confusum. However, on the basis of our results, white rice can host larvae of both strains for a considerable period of time before they die, i.e., maximally 124 and 61 days for the Serbian and Greek strains, respectively. During these intervals, larvae may be moved through the transfer of commodities between or among storage facilities and through their standard cleaning [74][75][76]. Found in the new storage environment and/or location in the same storage facility, more suitable grain commodities than white rice may exist for T. confusum larval development (e.g., barley, as our study indicates). Considering that stored-product insects, including T. confusum, are attracted variably by the volatile odors of stored-grains [7], new infestations may be initiated. Therefore, white rice could be considered as a vehicle of temporal survival of different T. confusum strains. Whether larvae that have their feeding commodity altered from white rice to a different one would be able to pupate and provide fecund adults merits further experimentation. The discovery of grain or non-grain commodities which marginally allow the development of stored-product insects is important since it illuminates the potential paths they follow for survival and further expansion [54,77].
Based on the values of the demographic parameters, both strains are able to increase their population on cracked barley. Therefore, this commodity can be suitable for T. confusum. One of the most important findings of the current study is that, although both strains successfully completed their cycle on barley, the females of the Serbian strain exhibited a higher net reproductive rate and faster larvae development. Despite the fact that our tests were conducted under the same abiotic and biotic conditions, the Serbian strain had been adapted for >25 years at 25 • C, although we reared it at 30 • C for one generation, while the Greek strain had been adapted for >17 years at 30 • C. This issue might partially explain the obtained differences between the two strains, taking into account that alteration of temperature and relative humidity led to alteration of the relative rate of increase of various lineages of another closely related stored-product pest T. castaneum [78]. Moreover, Kavallieratos et al. [25] found that the increase of temperature increased the food uptake of seven European strains of T. confusum differently. This fact was interpreted as a variable susceptible to natural insecticides, i.e., diatomaceous earths (DEs) as wheat protectants. Consequently, the increase in temperature may have activated the Serbian strain to become more fecund. Although the tested strains originated from neighboring countries (Greece and Serbia), they performed differently on barley on the basis of the calculated values of their life history parameters. Similarly, Vayias et al. [79] found significant differences in the mortality levels of adults between a Danish strain and a German strain of T. confusum seven days post-exposure on wheat treated with DEs. Later, Athanassiou et al. [26] showed that spinosad killed significantly more adults and larvae of a T. confusum strain which originated from Italy compared to a strain from Greece on wheat after 14 and 21 days of exposure. To contrast, according to Wade [29], two wild strains of T. confusum from the U.S.A. and Spain did not exhibit significant differences in the mean rates of increase of their populations over 14 generations on wheat flour.

Conclusions
Our study shows that different T. confusum strains may exhibit significant differences in their biological characteristics and demographic traits, with regard to the feeding commodity. However, it should be noted that the obtained differences also could be due to various causes such as rearing abiotic (e.g., temperature) or biotic (e.g., wheat varieties that form flour) conditions. Therefore, different strains may alter the potential growth and spread of a population, as well as the level of infestation of stored products. Taking a practical point of view, the knowledge of the life histories of certain strains leads to the optimization of their cultures in the insectary for the production of adequate numbers of individuals that are necessary for laboratory tests. Since origin impacts the response of T. confusum strains to insecticides either as grain protectants or as fumigants, or even as topical applications [25,[79][80][81][82][83], and taking into account the results of the current study, further research is needed to shed light on the life history of such strains infesting different types of commodities when they are treated with insecticides. The fact that different T. confusum strains show remarkable differences in their population dynamics when infesting certain suitable grain commodities should not be overlooked because it may lead to considerable losses. The study of their life history parameters on different abiotic scenarios would be a tool that enables the estimation of their temporal population fluctuation, triggering accurate management treatments against T. confusum.