Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Facebook Sign In with Google Sign In with OpenID Sign In with Twitter

In this Discussion

Frass and some other questions

Hi everyone, I'm new to the forum as of today. I started a little farm in my apartment together with my girlfriend maybe two months ago. Most of the original larvae have now metamorphosed and some small new larvae have been hatched. I'm based in Malmö, Sweden.My first goal is to be able to produce some food for our family, but long term I would love to go industrial with this. My focus is the make the process as efficient as possible.

After going through your great forum I have a few questions.

There seems to be lots of talk about separating frass from substrate/larva. How important is that? We have had our larva in the same substrate (with no removal of frass) for more than a month and they seem to be doing fine. Maybe beginners luck, what do I know.

We use wheat flour for substrate. We started with oats but found it incredibly tim consuming to separate the puppaes. Using flour we can easily sieve out the substrate, remove the pupae and put the larvae back into the substrate. How come no one else seems to use wheat flour?

How long do you keep the beetles? Until they die? Do they mate and lay eggs several times?

About the bag design. That is only for the last stage, right? I see very little about the other stages such as separating pupae from larvae and beetles from larvae. I guess that these stages will have considerable fewer iduviduals though, so maybe it's smart to focus on the last stage. For me who hasn't really reached the last stage yet it all seems to be about seperating pupae and beetles, but I guess that will change...

And one comment in all friendliness. My impression is that you focus to much on cheap material. In the long run labour cost will be the biggest expense. I believe creating a setup that requires minimal work is key to make bug production economically sustainable, even if it takes a little investment to reach there.

thanks, and hope we'll get to know each other better over the coming period :)


  • Hey Per, welcome to the community and thanks for some really insightful comments!

    I'll try to answer all your questions one by one - if I miss anything, let me know.

    Frass may have some negative impact on worm growth (I seem to recall a research paper demonstrating this, but I don't remember which one). However, any impact is likely to be very low. The only time separation is mandatory is at harvest - and if you can design a system that doesn't require a manual separation step then you save some time and effort.

    We haven't tried wheat flour ourselves, but the reason people focus on other feeds is that it is significantly more expensive than some others (such as wheat bran, which is an inexpensive byproduct of processing wheat). This doesn't matter so much at a small scale, but it could cut heavily into margins in a larger operation. Additionally, wheat bran contains a higher proportion of protein and fibre to carbohydrates than wheat flour, which actually makes it better suited to mealworms' requirements. If the fineness of flour is useful, finely ground bran might be an interesting idea!

    Beetles will start laying eggs 5-10 days after emergence, and die after 1-2 months. Each female will lay dozens of eggs every day, several hundred in total - so how long you keep them around depends on how many offspring you want and how many beetles you have. It's easy to end up with more than you can manage!

    The mealworm "bag" is designed to house worms from shortly after emergence until they are towards the end of their larval stage, ready to harvest. You're right that the other stages involve far fewer (or smaller) individuals. Since you'll only be breeding a relatively small number of adults, they can pupate, emerge, mate, lay and die in smaller breeding trays. You may want to periodically separate out the egg-laden substrate from breeding trays into a "nursery" area, to avoid eggs being eaten by beetles, but it's not necessarily required.

    With regard to separating pupae and beetles - the reason you'd do this is to prevent loss from newly emerged beetles eating pupae. Since this is very time consuming (as you have discovered), it generally makes more sense to start with a larger initial breeding population and not worry too much about the loss from cannibalism - you're losing a few more insects each cycle, but they're worth very little compared to the labor required to separate them.

    We totally agree about the cost of labour - it's definitely the biggest expense associated with farming at a small to medium scale. The "bag" kit designs reduce some of the time associated with processing, since they're filled with the right amount of feed they'll contain nothing but insects and shed exoskeletons at harvest time.

    Since it's worth trading a little yield for a lot of free time (by not separating beetles from pupae), these are the only times you'll need to interact with the farm:

    • Transfer juvenile worms from nursery containers to bags; add feed.
    • Add moisture food (carrots, etc) to each bag a few times per week.
    • Harvest adult worms by pouring them from one end of the bag. You can now use or sell them!
    • Take a small sample of breeders, allow them to metamorphose and breed in a separate container.
    • Remove substrate from breeding tray each week and place in nursery containers; wait to hatch.

    It should be a total of just a few hours of labour per three month cycle, even without any automation.

    I'd love to hear some of your ideas for automating the process further! One thing I like about mealworms is they can be manipulated as part of a fluid consisting of themselves and their substrate. It wouldn't be too difficult to create a completely automated mealworm farm that performs all of these operations without human interaction.

  • edited March 2014

    One place where extra investment can be handy is in climate control. Mealworms will grow fastest at around 80f/26.6c, so most places in the world will require some temperature regulation. If you can keep them at 80f, you can fit in more life cycles per year, which is a great way to increase annual revenue!

  • Mealworm frass has the compound farnesol, which induces the expression of what is called the "juvenile hormone" (hormone keeps larva from adulthood). When the level of juvenile hormone rises then the larva will molt again, but not pupate; in laymans terms too much frass increases the repetition of instars by the larvae. Numerous instars lead to larger larvae moving about & actually they are not likely to go after one another. However, they are prone to cannibalize those in the metamorphosis stage of pre-pupae which are vulnerable because those are not moving (larvae which look kind of crooked & when touch don't respond). Females are more susceptible to juvenile hormone ratios & respond to elevated levels by cycling their number of instars more than males do (in response to high amounts of juvenile hormone). Larval programing, during what is meant to be the last instar coincides with their brain pro-thoracico-tropic hormone still inhibited by juvenile hormone. The juvenile hormone must clear out of the larval haemolymph (insect circulatory fluid) to let the pro-thoracico-tropic hormone become active before metamorphosis is possible into pupae (no more instars).

  • So practically, reducing frass should limit the impact of farnesol on extending life cycle, causing insects to pupate more quickly than otherwise. Interesting - if you're aiming for a higher harvest weight, perhaps frass is a good thing!

  • Frass doesn't delay male larvae from pupating - just females. As discussed elsewhere low substrate oxygen ( rising CO2) induces herd population to undergo a shift in ratio of the sexes to favor a population of larvae that consists of more females (an evolutionary trait to preserve the capability for progeny in a situation threatening hypoxia). In theory, by adding a precisely controlled flow of CO2 ( via metered valve fed by tank of CO2) into the mealworm's environment that is not going to cause anaesthesia (from too high CO2 level) this would cause the herd's female larvae to predominate. Then dansitu's theory of maintaining a maximally tolerable amount of frass in the substrate would provoke those female larvae to hang around instead of metamophosize into pupae & end up weighing more. This would take experimentation of CO2 ppm; but if aiming for an automated system & preparing a closely controlled environment it may be worthwhile determining how much the concept can standardize ones marketable yield. CO2 gas tank refills are cheap (at least in USA), there are propane driven CO2 production systems & if altering substrate gas proportions would contribute to a maximized return may make it profitable to automate mealworm production as Per recommends (for that matter altering the nitrogen gas ratio in coordination with altered CO2 ratios also affects mealworm larvae). Maybe an additional refinement would then be to determine what time cycle introducing CO2 is best; such as "x" hours administered diurnal (per 24 hour period) at "x" concentration, these factors in precise phases. I suspect optimal results, if theory shows merit, would turn out to be more complex than a simple linear tactic.

  • At first glance it seems that larger larvae at harvest might be desirable, but adding instars and drawing out the larval period may lower the FCR, and drawing out the lifecycle could result in requiring more parallel generations developing for continuous production. On the other hand, increased protein content vs. chitin in a larger larva might end up being more value overall, but the calculations would need to be made (or found) on the lifetime FCR.

    One account posits the optimal harvest weight at 100-110 mg / worm for maximum efficiency (A.E.Ghalyand, F.N.Alkoaik, The Yellow Mealworm as a Novel Source of Protein, American Journal of Agricultural and Biological Sciences 4 (4): 319-331, 2009)

  • Automating mealworm production should include mechanical stimulation. This can be provided by strung beads moving across/through the growing mealworms as lie in their substrate; a motor geared up for variable speed can rigged to drag/push/pull the chain of beads. The beads serve as a tactile agent which does 2 things. First it challenges the larvae during the develop of their adaptive innate immunity. Immunological challenges can lead to greater body weight down the line because it causes more eating activity. Second it stimulates the nervous system to produce more juvenile hormone. Then the juvenile hormone acts as a feedback regulator to dampen down the immune reactivity (allays auto-immune flare up). The potential of this tactic integrated into an automated method is to mechanically delay pupation & let the larvae continue to grow up to 50% longer. Of course there is a point of diminishing return once the larvae start to consume their own body fat (lose plumpness & respond sluggishly); then these larvae are also going to be targeted for cannibalism. The diameter of the beads in the chain will make a difference & it should not be over sized in relation to the larvae. Look for 3 - 6 mm diameter beads as most logical sizes to start experimenting with; estimating those size mm beads will be approximately 1/10th to 1/5th the size of average mealworm larvae, non-traumatic & practical to move through the herd without disrupting it's habitat. Duration of each tactile passage of bead chains is another factor; a safe periodicity to begin with is 5 minutes of every hour (not continuous) performed diurnally (24 hrs.) . Female larvae are probably going demonstrate a reaction equal to the males only when they (the females) are stimulated by the bead passes for longer phases of time (each time repeat the stimulation diurnally); but this is conjecture assuming female mealworm larvae begin with a lower base line of movement than the males. Beads should be slow moving to mimic what the larvae would experience from a mother beetle moving through the herd. The idea is every time the motor sweeps/drags the bead chain once an hour the larvae are touched by a bead once, not again & again without an hour's respite; although some experimentation may prove that more than 1 bead nudge hourly sweep is better (at giving larvae more time to fatten). Furthermore, & again these are extrapolations: this mechanical stimulation is likely to improve the responsiveness to growth in a way that compensates for any less than ideal humidity conditions. Plus it will be more relevant than just modulating light/sound &/or shaking/vibrating the larvae alone; however, these could prove to add some benefit when coordinated with tactile (bead chain) stimulation.

  • To be clear, I am proposing an alternative (larval stimulation) to letting frass levels rise & still getting to provide a longer larval growth/weight period (as dansitu observed frass indirectly promotes). Mealworm frass has 18.5% protein making it very subject to micro-organisms proteo-lysis (here a unsuitable protein modification) the longer it is not taken away. Their frass also contains 22% oleic acid, 47% linoleic acid & 19% palmitic acid; which could be isolated to be recycled (purified) in animal feed. Data source also proposes recycling the frass protein. See (2012 International Journal of Industrial Entomology ; 25,1: pgs 93-98 "Nutritional Value of Mealworm, Tenebrio molitor as Food Source" =

  • Installation for automated mealworm larvae would be something like the following... Larva and their immediate substrate are every hour tilted onto a conveyor belt in a thin layer. This moves them slowly from their controlled humidity/temperature environment toward a stable chain of beads that crosses perpendicular to the conveyor belt. One needs to determine the speed at which the larvae are shunted onto the conveyor belt. The speed with which they eventually physically move through the chain of beads should be close to the speed which one's particular strain of female beetles moves along. Similarly one needs to assess the average weight of their female beetles; the weight of different types of beads may vary & instead of just going by bead diameter look for beads that weigh similar to adults, while not being too large in diameter. Where the chain of beads are encountered it will be like a curtain the larvae must swish through & the larvae should be displaced as they pass like an pride of adult female beetles would wedge the larvae aside in a soft substrate. Adjust the temperature level of the sector where larvae on conveyor belt pass through the bead rope/curtain. Use an infra-red heat "gun" to measure what the temperature goes up by in normal circumstances when adult beetles move among larvae - in general movement raises temperature & , of course, this will be to a small degree which is transitory. Recreating the native conditions of larvae amongst adult beetles will also require an analysis of the atmospheric micro-climate (oxygen/CO2/nitrogen levels) when beetles move on larvae. Thus, at the same time temperature is manipulated (upward a fraction?) when larvae on the conveyor belt are inside the sector where they are going to pass through the bead curtain that isolated sector's "air" will be adjusted (lower ratio of oxygen?). O.K. so the larvae have been conveyed to a sector that briefly alters the temperature, air & possibly relative humidity before/while subjected to tactile stimulation from a curtain of chained beads. Look at how packing houses/walk in cool rooms use heavy clear plastic slats hanging down to separate environmental conditions, while facilitating what brushes them aside for getting in & out by simple momentum. Next, the larvae have left the stimulation "room" & the conveyor belt carries them to a sorting sector; the temperature/humidity/atmosphere gas ratios for the sorting sector are re-adjusted to normal in this "room". The larvae are gently/slowly rolled off the edge of the solid conveyor belt bed which has carried them so far onto a different lower conveyor belt that is equally slowly running perpendicular, The top belt turns underneath, for returning to where it regularly picks up larvae & substrate for trips to the bead chain/curtain visit. The curling lip deposits the larvae & substrate onto the sorting conveyor belt, which catches them as they gently slide down/fall from above. Sorting conveyor belt is made with a taunt steel mesh that has precisely graduated spaces (holes). The holes will allow everything except the largest larvae to fall through in a separation process. What falls through are larvae too small to harvest, frass & substrate. These fall onto their own perforated conveyor belt with very small holes so most of the frass & dust are separated out. The too small (unharvested) larvae & substrate are conveyed back to the main growth chamber ("room") where they get a feeding time before next hourly trip on a solid conveyor belt to the bead chain/curtain stimulation room (& heat/oxygen ratio shift if determined to be relevant). Those larvae too big to fall through the holes are conveyed directly to a hopper (collection tub) inside a dedicated chamber where CO2 is pumped in at sufficient ppm concentration to anesthetize them until stop moving. The sized (holed) conveyor belt that dropped them off when it's edge curled under then returns to pick up those larvae which have just come from their bead chain/curtain stimulation. With only culled large larvae knocked out by CO2 they can then be sanitized (washed/irradiated with ultraviolet) & move to be euthanized by method of choice. Once dead the larvae can be collected & stored until enough have accumulated to be worth a production run for an end product. The frass & dust likewise will be shunted to a collection site, accumulated & dealt with. Frass' oil can be liberated easiest by the solvent hexane; this commercial technique is widely used for extracting oil from substrates that have protein (as does frass).

  • Do you have any references for the mechanical stimulation technique? I wonder if similar results could be obtained through vibration, or a periodic jet of air? This would be a lot less mechanically complex.

    The problem with sorting mealworms through any sort of sieve is that they tend to grab on; they grip the edges and block the holes. Also, while a worm might not fit lengthways through a gap, they can crawl through a much smaller hole than they would be able to fall through lengthways.

  • Thanks guys for some good answers!

  • Hi dansitu, ++++ITEM: Regarding "chain" stimulation see Tschinkel & Willsons' experiments & "chart 8" showing vibration lots less effective for what you propose: (1971) "Inhibition of pupation due to crowding in some tenebrionid "; in J. Exp. Zool. 176: 137-146; free full pdf = +++++ITEM: (2005) "Inhibition of pupation due to crowding in some tenebrionid beetle"; Quote: "...Pupation...inhibited by crowding...primary factor ... mechanical stimulation resulting from inter-larval contacts...." = +++++ITEM: regarding difference between handling & tactile strokes on heart see (1992) "A sensorory input inhibiting heart rate in an insect,Rhodnius prolixus"; Quote: "... handling increased contraction ... tactile stimulation of the ventral abdominal cuticle inhibited contraction...." Quote from link's look inside option (top left): "...stroking...ventral lateral abdomen...inhibition...contraction...cardio inhibitory factor...mealworm...." = +++++ITEM: indicating artificial stimulation best done in manner reducing heart rate to avoid losing stored energy, see (1988) "Hyperglycaemic and myoactive factors in the corpora cardiaca of the mealworm, Tenebrio molitor"; Quote: "...corpus cardiacum-corpus allatum complex of Tenebrio molitor contains factors causing hyperglycaemia, stimulating carbohydrate release from the fat body ...endogenous myoactive factor(s) ...increase in heart beat frequency and a decrease in the amplitude of contraction...."; ++++ITEM: possible limitation to air jets if waste larvae's energy in futile motor activity deduced from (1924) "On steriotropism in Tenebrio larvae"; Quote: "...bending..head away from the side stimulated...." =;

  • That's great, thanks for the references!

Sign In or Register to comment.