Introduction

Around 1900 insect species are used as human food around the world (^{van Huis, 2013}) and contribute significantly to local economies (^{DeFoliart, 1999}). In Mexico, 535 insect species are reported to be edible (^{Ramos-Elorduy et al., 2008}). Among these is Comadia redtenbacheri (Hammerschmidt), a lepidopteran of the Cossidae family, which is one of the four major wood borer families (^{Kalisch and Baxendale, 2010}). C. redtenbacheri is the only species of the family reported in Mexico (^{Brown, 1976}). Its main host species are pulque agaves, Agave salmiana Otto ex Salm, A. mapisaga Trel. and A. atrovirens Karw. ex Salm (^{Camacho et al., 2003}). Female C. redtenbacheri oviposit masses of eggs at the base of the agave leaves. The period of incubation is 30 to 35 d. After hatching, the larvae establish in the agave rhizome where they complete their development (^{Granados 1993}), passing through seven larval instars (^{Hernández-Livera et al., 2005}) over a period of five months in a greenhouse and eight in the field (^{Llanderal-Cázares et al., 2007}). Comadia redtenbacheri larvae, called maguey red worm, is regarded as human food and in the laboratory the pupation period is five months in a 5-cm deep substrate of equal parts of soil and vermiculite (^{Miranda-Perkins et al., 2013}). Adult longevity is only 3 to 5 d because their mouth apparatus is atrophied and they cannot feed (^{Hernández-Livera et al., 2005}; ^{Llanderal-Cázares et al., 2007}).

The fresh maguey red worm in the tourist area of Otumba and Teothihuacán, Estado de Mexico, has a price of $500.00 pesos a kilogram during the collection season and $700.00 during the months of low production (^{Miranda et al., 2011}). But ^{Camacho et al. (2005)} report prices of $800.00 to $1000.00 kg^-1.

Most edible insects are harvested from wild populations (^{van Huis, 2013}), and in the case of the maguey red worm, gatherers take advantage of its natural behavior for harvesting the worm during the rainy season from July to September, when the water stimulates the last instar larvae, ready to pupate, to leave the maguey. But the typical gathering procedure is to locate infested plants and extract the larvae even though they have not finished their development, causing wild populations to diminish (^{Granados, 1993}; ^{Miranda et al., 2011}). The procedure also affects the agave populations; plants are damaged and replanting practices are non-existent (^{Ramos-Elorduy, 2006}; ^{Llanderal-Cázares et al., 2010}).

^{Nolasco et al. (2002)} and ^{Camacho et al. (2005)} attempted establishing C. redtenbacheri at different development stages under controlled conditions. ^{Llanderal-Cázares et al. (2010)} reported establishment and development of C. redtenbacheri larval stages up to the pupation stage on agave plants grown in a greenhouse. In the laboratory, the processes of pupation, mating, oviposition and hatching were obtained (^{Miranda-Perkins and Llanderal-Cázares, 2013}; ^{Miranda-Perkins et al., 2016}).

^{van Huis et al. (2013)} believe that insect rearing under controlled conditions can help to prevent overexploitation in nature and propose that in the tropics emphasis should be on optimizing the productivity of traditional systems by developing procedures for managing the resource in small-scale rearing installations.

Thus, the objective of this study was to establish C. redtenbacheri rearing conditions in pots in which A. salmiana agave plants had been established to determine what number of larvae released per pot would obtain the highest percentages of pupation, adult emergence and oviposition.

Materials and Methods

Results and Discussion

The pupation process of C. redtenbacheri in the nursery began in September 2014 and concluded in March 2015 with the beginning of adult emergence, which lasted until May of the same year. When 600 larvae/pot were released, average emergence of 15.9 % was obtained and this was the best condition (p≤0.001); with the density of 400 larvae/pot, average emergence was 1.1 % and 5.3 % with 200 larvae/pot (Table 1).

Table 1 Average number of C. redtenbacheri adults emerged, by larval density, using Poisson regression test. 

By sexing the pupal exuviae of the emerged adults, we found that a little more than half were females for the three larval densities used: 48 females and 46 males (600), 2 females and 2 males (400) and 6 females and 4 males (200). Considering that each female can produce 104 descendants on average (^{Ramírez-Cruz and Llanderal-Cázares, 2015}), the 48 females in the density of 600 larvae/pot could yield 4992 larvae, sufficient to infest the available agaves.

^{Camacho et al. (2003)}, in an outdoor study, placed C. redtenbacheri larvae to pupate and obtained 15 % emergence; they inferred that the low percentage was due to drastic changes in light and temperature. In our study, the low emergence may have been due to excessive moisture in the pots from water retained by the soil due to rain during the time the experiment lasted, which was difficult to drain because of the size of the pots, and this may have induced the larvae that were inside their cocoons to exit again to search for another site for pupation. ^{Miranda-Perkins et al. (2013)} reported that when moisture was not provided to the larvae, adult emergence increased 50 %. Another cause of low emergence may have been the origin of the larvae. When larvae are from commercial samples, adult emergence is reduced 50 %. In insect rearing direct manipulation of the larvae should be minimal during their entire development to avoid causing stress (^{Madrigal 2001}).

The presence of entomopathogens in the soil used as substrate for pupation may also have had an important role in the low percentage of emergence. When the soil of the nine pots was sifted, we recovered 995 individuals in different stages of development, from larvae, prepupae (the great majority), and pupae to term. It is likely that an attack of entomopathogens would impede larvae from reaching the pupa stage, and if they do pupate, they are prevented from complete development into adults. Of the individuals recovered, 10 samples were analyzed, and fungi of the genera Rhizopus, Penicillium and Aspergillus were found. These are often found as pathogens in insect rearing (^{Shapiro, 1984}). Rhyzopus stolonifer (Ehrenberg: Fries) Vuillemin is a saprophyte in the soil and has been found in diets for silk worms Bombyx mori L. (^{Trivedy et al., 2011}). ^{Miranda-Perkins et al. (2013}) recorded that nearly 40 % of the C. redtenbacheri larvae deposited in the soil for pupation were infected by Beauveria spp., and ^{Hernández-Flores et al. (2015)} reported C. redtenbacheri larvae associated with several genera of bacteria.

Although disease transmission intensity in insect populations is proportional to the number of susceptible individuals and to the increase in stress associated with high population densities (^{Steinhaus, 1958}; ^{Vilaplana et al., 2008}), there are examples of insects that exhibit phenotypic plasticity in response to changes in population density, a phenomenon called denso-dependent prophylaxis (^{Simpson et al., 2001}). For example, when Schistocerca gregaria are found in high densities, they are more resistant to the fungus Metarhizium anisopliae than solitary locusts, due to a potentiated antimicrobial activity in their defense systems (^{Wilson et al., 2002}). This characteristic could have occurred in the case of C. redtenbacheri, since a higher rate of emergence occurred at the highest density (600 larvae per pot).

For weight of oviposited egg masses, the best treatment was the density of 600 larvae (p≤0.013), relative to the other densities evaluated (Table 2), obtaining an average of 0.961 g, which, when divided by the calculated average weight of each egg (0.00056 g), results in an equivalent of 1715 eggs. This translates as the number of potential larvae: almost the same number (95 %) of the larvae obtained initially in the three replications (1800). Under a management system, even if a third of the eggs were not fertile and if the surviving larvae established in the agaves and continued their development, in the following generations, the adults could accomplish a new infestation of the agaves available in the area destined to that end.

Table 2 Average weight of egg masses of C. redtenbacheri, by larval density. 

Values with different letters are statistically different (Tukey; p≤0.05).

Because the highest percentage of adult emergence and oviposition were observed in the treatment with the largest number of larvae (600), it is advisable to increase the density of individuals in other experiments to obtain higher production in the same area. Thus, the rearing system would be more efficient; besides, non-commercial larvae should be used.

Conclusions

Comadia redtenbacheri establishment in the nursery was possible by releasing larvae on agave plants growing in pots and kept outdoors. Pupation, adult emergence and oviposition on the available agave plants was achieved with the densities used. The best result was obtained with the density of 600 larvae per pot. In further studies, the species should continue its development so that in the following generation, the larvae that hatch from the egg masses can establish on their own in the available agave plants. The use of higher densities of larvae per unit of area for initial infestation could result in higher adult emergence and oviposition