A series of larval diet experiments were initiated with simple low cost ingredients consisting of various cereals and legumes. From experiments on single diet components, shorter larval periods were recorded for corn, chickpea and bovine liver while larvae feeding on the mushroom diet had the longest larval period. Survival of larvae to the pupal and adult stage was lowest (60 and 45% respectively) for larvae feeding on mushroom in the series of diet ingredients tested.
Nutritional requirements of mosquito larvae are known to include at least 14 amino acids, sugars, polyunsaturated fatty acids (PUFAs) especially C18, C20, and C22, [19–24], sterols, and nucleotides for the larval development, survival and adult flight [25, 26], and a minimal concentration of essential vitamins to allow their optimal growth [25, 27]. Studies on the vitamin supplements in Culex pipiens have shown their significant role in normal growth and survival. In related experiments the absence of riboflavin, folic acid, biotin or choline, resulted in few larvae developing to pupae, and most of them died in the 3rd or 4th instar [24, 28].
Cereals and legumes can meet several of the dietary requirements of the mosquito larvae. Corn consists of 1-2% lipid with 72-85% unsaturated fatty acids, primarily, oleic acid and linoleic acid. Corn is also rich in vitamin A and has a high protein content . The chickpeas are rich in poly-unsaturated fatty acids, proteins, carbohydrates, B-group vitamins, and certain minerals [30–34]. Both cereal grains and legumes contain varying amounts of carbohydrate, fat, protein, water and minerals [33, 35, 36]. Thus corn and chickpea alone could support larval development of An. stephensi.
In the second experiment, yeast was added to the individual diet ingredients to check for an improvement in the diet quality. The addition of baker's yeast did not lead to any improvement in the measured parameters. It might have been due to scum formation observed in yeast added diets that caused larval mortality and delayed pupal emergence. Based on these observation yeast was not tested in further experiments.
For further improvement in the diet we prepared mixed diets including the components that performed better in the single diet experiment. Different quantities of the mixtures were administered to see the optimum level of each diet mixture. All the three diet mixtures performed well when used in appropriate quantities. Mixing the plant ingredients compensates for nutritional deficiencies present in individual diet components. For example, cereals are reported to have low quantities of the amino acids tryptophan and methionine , which are present in legumes such as chick-pea and beans. There was an improvement in larval parameters when the amount of diet administered was increased to more than 4 mg (provided on alternate days for the first four days and then daily). Further increases in the amount fed did not result in any significant improvement. We did not find a significant difference in the fecundity for different quantities of the mixture diets. The difference might have been due to the variation resulting from the protocols used. We estimated fecundity from the output of five females due to handling ease but there is a chance of missing individual difference if some of the females did not oviposit at all.
The wing size of male mosquitoes also did not differ for the mixture diets used in the study. However overall wing size was greater as compared with adults reared on two component diets (Table 2). In a related experiment on mixture diets administered to An. arabiensis larvae, the size of L4 larva and male size (wing-length) was reported to be the highest for a mixed diet prepared from natural ingredients . These observations support the hypothesis that single larval diets might be adequate for the completion of larval stages to emergence but may lack appropriate quantities of vitamins and minerals that would positively influence adult size. Other studies confirm the direct effect of parental rearing on the developing embryo  or the influence of dietary restrictions on larval and adult development . Although, the objective of the present study was only to look at the effect of different diets on life parameters of mosquitoes and not the effect of larval rearing on offspring, we observed useful clues on the indirect effect of larval diets on adult size. Therefore, further studies should be performed on the effect of diet quality provided to early stages of mosquitoes and their effect on adult life parameters.
We prepared mixed diets by combining bovine liver with ingredients from plant sources (cereal and legumes) that are rich in essential amino acids, saturated and unsaturated fatty acids, minerals and vitamins [37, 41]. The inclusion of bovine liver, an animal-derived protein, also ensures the presence of at least 14 essential amino acids, asparagine, arginine, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, and valine . The reason that our mixed diets performed better might be due to the high/ balanced nutritive values of the ingredients present in combination diets. Further experiments on the standardization of this diet for mass rearing of several Anopheles and Aedes species need to be carried out prior to its use on a large scale.
Taking into account the life parameters, comparable results were obtained when 5 mg of 4, 5 or 6 ingredient diets were used. Looking at the cost analysis (Table 4) there is a decline in the cost involved with the addition of each plant derived ingredient thus the cost of diet drops from 21.7 US$/ 1.3 kg (with three plant ingredients plus liver) to 14.7 US$/ 1.3 kg (with five plant ingredients plus liver) for rearing one million larvae. This is because the relative cost of plant derived ingredients is much lower than that of bovine liver and addition of more plant derived ingredients reduces the amount of bovine liver used per kg of the final diet mixture. Therefore we recommend the six component diet (BCWPRL) as the most cost effective balanced diet for rearing the larvae of An. stephensi.
Results from related studies on diet development using commercial ingredients for An. arabiensis have demonstrated the usefulness of mixture diets from bovine liver, tuna meal and vitamin mix with an estimated cost of US$ 64.30/ 1.4 kg of diet for production of one million mosquitoes . The diet reported here can be used to rear one million mosquito larvae of An. stephensi at an average cost of US$ 15 only. The prices calculated here are based on small quantity orders and would probably be reduced significantly when ordering the bulk quantities needed in a mass-rearing facility.