As Bt Cotton turns 10, observational data certifies it a Super-Flop: Part III

Has the Bt gene reduced pest infestation and decreased expenditure on pesticides? - The Theory

Resistance is a decrease in pest susceptibility that can be measured over human experience. When you use an insecticide to control a pest, some populations eventually evolves resistance." (Tabashnik).  

Under the stress of intensive chemical spraying the weaker members of the insect populations get weeded out.” (Carson). Evolution of insecticide resistance in >400 species of insects not only confirmed Darwin's survival of the fittest theory besides threatening agriculture and human health worldwide. This was the global experience of using hybrid seeds. Bt Cotton is just an accentuation of this trend as the only change now was usage of the same hybrids with a Bt gene inserted into their germplasm.

Although scientists are trying to directly increase yield of some agronomically important crops using transgenic technology, success are so far eluding them (Jenner 2003). So any direct contribution to yield increase has to be accounted by the (hybrid cotton) germplasm in which the Bt gene is inserted. Accordingly, the Bt gene does not directly increase yields but only does so indirectly by reducing yield loss due to pests.  But the Bt gene is not designed to protect against all pests but only against the Lepidopteran insects that account for 50% of yield reduction attributed to pests in the country. 

So theoretically the Bt gene’s contribution to increased yield is limited to the degree it offers crop resistance to Lepidopteran insects and thereby preventing loss in yields. Actual increase or decrease in Bt yield has to be in relation to the yield loss of the non-transgenic counterparts under the same cropping practice.  But in India presently there are an estimated 680 brands of BT cotton seeds whose vulnerability to different kinds of pests vary greatly, that includes the degree of their vulnerability to Lepidopteran insect pests. 

But because it confers cotton plant resistance to Lepidopteran insects, the Bt lobby further claims that farmers are able to reduce their input costs enabled by a decrease in the volume of insecticides sprayed (and associated costs) which in turn boosts income margins as spraying less insecticide also means lower application costs.  For small-scale farmers who do most farming activities by hand, this means less time for weeding and other management practices.

The history of insecticide resistance however informs us that adaptation by insects could diminish the long-term efficacy of Bt crops. Meihls et al report that resistance evolved quickly without refuges and slower or not at all with refuges. So as a technological fix, a refuge management strategy was mandated in many countries including India, to delay insect resistance, which requires refuges of host plants without Bt toxins near Bt crops to promote survival of susceptible pests.

The concept was based on the assumption that most of the rare resistant pests surviving on Bt crops will mate with abundant susceptible pests from nearby refuges of host plants without Bt toxins. If inheritance of resistance is recessive, the hybrid progeny from such mating will die on Bt crops, substantially slowing the evolution of resistance.

This approach is sometimes also called the “high-dose refuge strategy” because it works best if the dose of toxin ingested by insects on Bt plants is high enough to kill all or nearly all of the aforementioned hybrid progeny. In principle, if a high dose is not achieved, resistance can be delayed by increasing refuge abundance, which lowers the proportion of the population selected for resistance to compensate for survival of hybrid progeny on Bt plants.

NGOs and environmentalists on the other hand ridiculed this theory. They highlighted there is ample evidence that suggests that pests develop resistance to just about any pesticide thrown at them. Bt toxin secreted by the Bt crops is no different from other chemical pesticides. Early on it will cause a temporary reduction of pesticide use (and associated costs), but resistance will eventually develop. They predicted that pest resistance to Bt could appear in the field within three to five years.

Evidence vs Theory

Dr. Kehsav Kranthi, the Director of India’s Central Institute for Cotton Research in a 3 part paper entitled 10 Years of Bt in India provides a review of the Bt gene’s actual performance in the field. This is really a Government of India (GoI) review. Here's some extracts:

"The leaf curl virus started to resurface as a major problem in north India, primarily due to the introduction of more than 270 Bt hybrids in north India, most of which were susceptible to the leaf curl virus. Until 2005, 100% of cotton area in the north was under varieties. Now 95% of the area is under hybrids in Punjab and Haryana, while 40% of the area is under hybrids in Rajasthan. 

Productivity in north India is likely to decline because of the declining potential of hybrids; the emerging problem of leaf curl virus on the new susceptible Bt-hybrids; a high level of susceptibility to sucking pests (straight varieties were resistant); problems with nutrient deficiencies and physiological disorders; and mealybugs, whiteflies and miscellaneous insect problems that are likely to increase.

The tobacco caterpillar, Spodoptera litura, resurfaced as a problem again as predicted, because of the low toxicity of the Cry1Ac toxin on the species. Bollworms started reappearing on Bt cotton. 

In February 2010, Monsanto, India declared that pink bollworm had developed resistance to Cry1Ac and that only Bollgard-II would be effective thereafter. Resistance monitoring studies at CICR showed that the American bollworm Helicoverpa armigera was also showing incipient tolerance in some parts of India.

The leaf hoppers showed very high levels of resistance of up to 5,000-fold to imidacloprid and other neonicotinoid insecticides in central India. The neonicotinoids were introduced barely a decade ago.

Progressive nutrient (macro and micro) depletion due to the source sink relationship because of Bt-cotton after Bt-cotton hybrid cultivation. Bt-cotton hybrids utilize more nutrients and water for higher yields and profits, therefore the soils are getting progressively depleted and need more nutrient recharging.

The cotton crop is showing nutrient deficiency symptoms in many regions, especially in rainfed zones where wilt and leaf-reddening problems are getting more severe over the years. The productivity is maximum in good soils."

The salient points of the GoI’s evaluation of Bt Cotton’s performance during the last decade accordingly are as follows:

 1. Imidacloprid ( used as seed treatment in cotton) that gave a 20-30% yield gain was succumbing to resistance, significantly, coinciding with the observed yield decline of cotton seen during the last 3 years.

2. The Bt gene was touted basically a protection against the bollworm complex, particularly effective against the American bollworm.  A review of literature suggests the single-protein Cry1Ac products continue to more or less control the bollworm complex other than pink bollworm (Pectinophora gossypiella). 

Two possible reasons are attributed to its ineffectiveness against the latter - either the period of expression of the Bt endotoxin does not coincide with the time of the bollworm attack or it's populations harbour three genetic mutations that confer resistance to Bt toxin (Morin et al, 2003), the latter being a more popularly accepted explanation. 

The real setback for Bt is that after proving that it could reduce the incidence of American Bollworm for some years, there is growing evidence of incipient resistance in the Central region accounting for nearly 60% of all cotton cultivation in the country.  Some of the reasons for that are as follows:

-  The toxic expression depending on the parental background of the hybrid varies widely due to the proliferation of brands.

- In some plants, toxin expressions decreases 110 days after sowing to levels that are inadequate to protect the plants from bollworm attacks. (Doyle et al) However maximum and extensive damage potential from bollworms is in their early stages.

- In India, cotton bollworm is more serious than in USA or in China; this means that Indian farmers needed to spray more pesticides against bollworm as resistance built up. 

 The second-generation Bt cotton technology, Bollgard II containing two proteins, Cry1Ac and Cry2Ab introduced in 2006 was supposed to check both these limitations on the assumption that the combination overcomes pests that are resistant to just one toxin. But this too has not fully succeeded.  The assumption that the hybrid offspring produced by mating between susceptible and resistant moths are resistant had been falsified with the appearance of incipient resistance, indicating the opposite happening - the inheritance tends to be dominant.

3. Bt cotton by inserting a gene into hybrid cotton was intended to make manual spraying obsolete. In fact, that was the claim of their hardsell when it was first introduced in this country. The object was to make the entire plant toxic so that the insects feeding on it are killed by increasing toxicity to 1,000 times to prevent resistance build up. But in reality, Bt toxins Cry34Ab and Cry35Ab often does not meet this high-dose standard (Meihls et al) so it does not kill all insects feeding on the Bt crop but permits support to a significant population of resistant insects. 

Further since Bt’s endotoxins do not confer protection in the entire life cycle of the Bt plant, manual sprays though reduced cannot be eliminated pointed out Suman  Sahai of Ngo, Gene Campaign. She explains:

“The expression of the Bt cotton gene worked for only 90 days while Indian cotton took 160 days to mature. In other words, during the crucial period when the crop needed protection from pests, it remained unprotected”  

Besides, since the Bt gene confers protection against only Lepidopteran insects and not others like sucking insects, farmers continued spraying broad spectrum pesticides. This realization made Bt advocates such as Keshav Kranti to clarify that Bt is just one element in the total package of pest management, which is a big come down by Bt lobbyists who had claimed at entry that Bt will make manual spraying obsolete.

4. The impact of (2 and 3) was two-fold. On one hand it catalyzed the rise of hitherto unknown insect pests like mealybugs and on the other hand it also catalysed the resurgence of major pests that previously was brought under control. Besides it elevated other minor pests to major pests. If all these are an India specific development, Bt as a technology could be given a benefit of doubt. But it ties very well with similar happenings in other Bt growing countries, whether the US, China or Pakistan.  

Thus the rise of predation by non-target crops completely neutralized the gains of the Bt gene reflected on productivity rates which began to decline after peaking in 2006-07. Keshva Kranti observed:

A mealybug named Phenacoccus solenopsis, not observed earlier in India, has spread across northern, central and western states after it was first recognised as a cotton pest about five years ago. In desperation, farmers have begun to spray "extremely hazardous" pesticides on the cotton to fight the insect, which has a waxy coating over its surface that makes it hard to kill with less toxic pesticides”.

We neither have a transgenic gene nor a variety, whether hybrid or desi, that is resistant to this mealybug. As a result, farmers need to spray at least 6 times, just to contain its damages, at the cost of Rs 3,000-4,000 per hectare.

5. The combined unintended side effect was these limitations of transgenic technology reinforced each other, spiking pesticide consumption and usage that resulted in indiscriminate killing of both pest and beneficial insects or bacteria including predators of till then minor pests of cotton. It was also found to having adverse effects on natural predators or parasites of bollworms themselves.

Pesticide consumption after moderately declining till 2006-07 started rising again and accordingly input expenditure of farmers for pesticides which after declining till 2006-07 doubled by 20010-11.  A new study by a Washington University in St. Louis anthropologist by Dr Glen Stone (2011) used triangulation methods to confirm this trend. Stone observed that the prices of Lepidopteran insect’s pesticides remaining more or else constant at 2003 prices while those of sucking insecticides commanding a premium.

So while Bt moderately succeeded in keeping in check one set of pest (Lepidopteran insects) damages, it created several sets of new pest damages (e.g. sucking insects) that completely negates its gains. The result had a telling reflection on productivity that peaked in 2006-07 at 554 kgs/link per hectare and then declined for three consequent years to 474 kgs/link per hectare.

6. Perhaps the most loaded statement in Keshva Kranti’s paper is when he notably observed:

“The area under Bt cotton has reached above 90% in many parts of the country but farmers are not following the recommended refugia practices. The intensive Bt cultivation and the noncompliance of refugia is likely to hasten resistance development.”

The concept of a refuge on one hand reflects the inefficiency and on the other hand, the inappropriateness of Bt technology:

a. Inappropriateness: In countries like the US or China, land is less of a constraint than it is India, and so they can more afford to set aside 20% of the cultivation area for refuge as their average land holding sizes are much larger. In India, this is viewed as waste to put arable land for non-productive purpose.

b. Inefficiency: Bt was touted to have been designed to kill pests by increasing toxicity levels to 1,000 times. But this did not happen at all or at all the times as Bt endotoxins lacked consistency and timing of production of high dosage toxins. At other times, despite the high dosages, pests survived by mutation to acquire toxin resistance all the same.

According to regulatory conditions, the size of the refuge belt should be either 5 rows of non-Bt cotton or 20% of total sown area, whichever is more. Due to small and marginal land holdings In the country, these norms are not followed. If in the US the default rate is 20%, whereas in India the compliance rate is less than 10%.

For quicker adoption of Bt seeds, sections of the Bt industry even argued there was in fact no need for a refuge. Manjunath (2004 & 2005) a former director of Monsanto India  argued instead that there existed a large number of alternative hosts like chickpea; pigeon pea; sorghum and tomato which could serve as natural refuge. Bt had then a share in overall cultivation less than 15%. So non-compliance of refuge requirements cannot per se explain the hastening of resistance build up. What can more probably completely explain it is that as Bt reached a share of 90% of area cultivation, it did so at the cost of cannibalizing this ‘natural’ refuge. There are several implications arising from this:

a. The effectiveness of Bt’s endotoxins reduces as the technology expands its share in area of cultivation.  The more it expands its share in area of cultivation, greater the prospects of pests acquiring resistance to its endotoxins. Now that Bt commands nearly 90% of area of cultivation, it signed its own death warrant. The question of ‘mass’ resistance is not of if but when it explodes. This is the real import of Dr Keshav Kranti comment: “Reduction in area of intercrops can hasten the development of bollworm resistance to Bt-cotton” .

b. The concept of refuge was for its touted utility to slow down the onset of resistance. The assumption is that pests surviving on Bt crops will mate with abundant susceptible pests from nearby refuges of host plants without Bt toxins and the inherited resistant gene will be recessive so that pests will be killed by endotoxins of the Bt crop.

Now recent studies show these series of assumptions as not completely correct. One bunch of studies point that the inherited resistant gene instead can be dominant too, particularly if the resistant insects mate among themselves! (Meihls et al & Ranjith et al). 

Further a study (Anilkumar et al) published in the Applied and Environmental Microbiology concluded that in the case of Helicoverpa zea,, thought to have evolved from the American bollworm), the inheritance trait appears to be statistically more dominant than in other cotton pests. Accordingly, this might explain why Helicoverpa zea developed resistance to Bt.  

All these simply mean even if there is full compliance of the refuge system, its efficiency is highly questionable. The refuge is just a red herring explanation why Bt as a technology failed.

c. Further, if resistant insects visit a refuge, then it is not only for mating, it is also for food. Now here lies the danger of treating alternative hosts like chickpea; pigeon pea; sorghum and tomato serving as “natural” refuge. Damage from infestation to these crops increases, causing significant yield drops for them. Consequently, any benefit-cost analysis on Bt cotton should encompass its impact on “natural” refuge. A review of literature suggests that this is never done.

Despite the refuge system being integral in the Bt package, the irony is that cost implications whether land set aside, seeds and other inputs are not considered in the benefit-cost analysis of the Bt crop. Once factored in, it would look less rosy even if economical otherwise.  

Conclusion

So back to the question:  Has the Bt gene reduced pest infestation and decreased expenditure on pesticides?

Bt’s weakness as a technology stems from not incorporating the lessons of our hybrid experience. To prevent resistance build up in insects, pesticide management should reflect its judicious use - finding the right toxin-pest fit; right dosage-degree of infestation fit combined with timely and required frequency of applications. This principle was observed to be violated in common practices of hybrid cultivation resulting in contributing to resistant build up in insects, necessitating application of new toxins and/or higher dosages to control the same degree of pest infestation.  

The endotoxins secreted from Bt besides does not permit effective control over of either their timing or dosage. The very fact that it targets only one set of insects, necessitates manual spraying of other categories of broad spectrum acting pesticides, including ‘cocktails’ that violates judicious norms of sustainable pest management principles. Since Bt does not offer protection for bollworms during the entire life cycle of the plant, it also entails limited manual spraying to complement its endotoxins. Multiple stacking of genes, the new generation of Bt’s are in fact no different from a ‘cocktail’ in effect except that it is much worse in impact.

The refuge system was neither complied with by farmers or if complied with, was ineffective to prevent the build up resistance.  In fact, the goal of refuge was defensive to begin with. Prevention was never its aim but only to delay resistance from developing long enough so that it becomes manageable, so that, perhaps, by the time super-bollworms evolve, there are new versions of super-Bollgard available to farmers to contain the threat so that the cycle begins again, escalating seed costs for farmers and boosting revenue earnings for seed companies.

The refuge system was a poor alternate to the system of crop rotation, one of the oldest and most effective cultural control strategies. It meant that the succeeding crop must be a different family than the previous one, based on the appreciation that most significant pests are crop specific and their populations can naturally decline if deprived of food by rotating crops, dying due to absence of food. Bt however created the illusion that it was a superior alternative by offering the refuge system. 

It is no surprise that the attempt to stall the evolution of insect resistance through transgenic crops producing Bacillus Thuringiensis (Bt) toxins ended up a failure as instead of pests getting killed, they instead began thriving on it. (MT Ranjith et al).  But Glenn Gladstone of Washington University in St. Louis in his study warns the failure goes much beyond this level:

"Looking beyond the field level to the farm level you see the real problem was a set of factors that eroded the normal process of farmer evaluation of technologies — there were too many rapid, undecipherable changes. 

Each new technology — hybrids, then pesticide after pesticide — brought short-term gains but further eroded farm management. Bt cotton has raised yields on average, but already we are seeing erosion of benefits as non-target pest populations are booming. It has also brought a quickening of technological change and undecipherability, which is the real underlying problem."

Here ends Part III Click here for Part I and Part II. This paper is written by Rajan Alexander, Development Consultancy Group, Bangalore. Rajan is a livelihood consultant and had been in the NGO sector for more than 31 years.  He can be contacted at rajan.alexander@yahoo.co.in