Influence of Cover Crops on Argentine Ant Populations and Mealybug Infestation in a California Vineyard

Mealybugs are not unknown to grape growers in regions with a Mediterranean-like climate, including California.  They are known to be an problematic pest in the vineyard, and can cause significant economic harm by damaging grape clusters with their metabolic secretions/waste, eggs, honeydew, and various molds.

Mealybugs share a mutualistic relationship with certain species of ant (Argentine), the latter which feeds on the honeydew produced and “improves living conditions” for the bugs.  Where there are mealybugs,

Photo By Crisco 1492 (Own work) [CC-BY-SA-3.0 (], via Wikimedia Commons

Photo By Crisco 1492 (Own work) [CC-BY-SA-3.0 (], via Wikimedia Commons

there are often Argentine ants “tending to their flock”.

The obscure mealybug (Pseudococcus viburni) is a known species of mealybug that lives on the weeds of and causes problems with vineyards in California.  Native to South America, the obscure mealybug has spread all over the globe, and in California is found in cooler central coast regions, including Monterey, San Luis Obispo, Santa Barbara, and Carneros (southern Napa).

A recent study raised the question: how does the vegetation on the ground in the vineyard affect this relationship and the symbiotic interactions between ants and mealybugs?  Do cover crops and other weeds provide a sort of protective cover for the ants, increasing the mealybug population and causing further damage to the grapes?  Or do cover crops decrease habitat quality for these insects, ultimately reducing their populations and avoiding economic damage to the grapes?

As an aside, this study also examined the frequency of Botrytis cinerea infections, since the area of study is known for heavy fog and high dew points, which are linked to increased Botrytis infections.

Quick re-cap of study methods

This study occurred at the Laetitia Vineyards and Winery, which is located in Arroyo Grande, San Luis Obispo County.

A two-factor split-plot was utilized in this study, with weeds versus no weeds as the main plot factor, and ants versus no ants as the split plot factor (each factor was replicated 4 times).   In other words, there were several different plot types studied: 1) weeds with ants; 2) weeds with no ants; 3) no weeds with ants; and 4) no weeds with no ants.

To keep the ants out of specific plots, sticky barricades and insecticides were used.  The authors noted that the insecticides would not affect the mealybug population.

Photo By USDA ARS Photo Unit [CC-BY-3.0 (], via Wikimedia Commons

Photo By USDA ARS Photo Unit [CC-BY-3.0 (], via Wikimedia Commons

To attract ants into certain plots, sugar traps were used, where the ants could come and go and feed on this supplied food source as they wished.

The researchers measured several things, including: 1) volume of sugar solution removed by the ants;  2) the estimated number of ants visiting in a 24 hour period; 3) mealybug infestation; 4) and percent Botrytis infection.

The data from 7 weeks prior to harvest were used for statistical analysis.


  • 22 different weed species were found, and 6 of these were found to house obscure mealybugs.

o   The 6 species observed hosting mealybugs were: 1) annual sow thistle; 2) burr medic; 3) mustard; 4) field bindweed; 5) cheeseweed; and 6) garden yellow rocket.

  • Ant activity was increased in plots containing weeds.
  • Mealybug infestation increased by 53.2% when the ants were not excluded from the plots.
  • Botrytis infections were 3 times higher in plots containing ants.

What do these results mean?

The results of this study pretty much confirmed what many others have already found in other plant systems: mealybug infestations are increased when ants are present.  Argentine ants and mealybugs have a mutualistic relationship with each other, so when there is one, you will likely find the other.

What about the presence of cover crops and other weeds?

This study showed that ant activity was increased in plots containing weeds.  Based on the relationship noted above, one can say that mealybug infestations were also increased in plots containing weeds.  While it was not tested, the researchers theorized that the presence of weeds might have improved soil moisture in the plot and maintained a soil temperature that is enjoyed by the Argentine ant.  Additionally, for those plots where weeds were forcibly removed, the local ant population was likely decreased by repeated cultivations.

Is there a relationship between ant populations, mealybugs, and Botrytis?

According to these results, it appears as though Botrytis infections were three times higher in plots containing ants than in those plots that excluded them.  This, according to the researchers, has never been documented in the primary literature before.

Botrytis is not carried by mealybugs, and it has not been documented that the Argentine ant causes damage to the grapes directly, so what could be causing this increase in infections when the ant and mealybug populations were increased?  According to the researchers, one possible explanation is that the mealybugs cause small cuts/holes in the grapes (which has been documented), creating a possible vector for infection.  Since the ant population and mealybug population is strongly tied, one might expect to see increased Botrytis infection where they also see higher numbers of the Argentine ant.

Potential source of error

It is important to note, and the researchers admitted this: by using sugar traps, they may have artificially increased the ant population due to ants in adjacent plots being attracted to sweet yummy nectar provided.  We might not be seeing a true interaction between the weeds and the ant and mealybug populations due to this potentially inflated ant population cause by a strong attractant like sugar.


This was a pretty basic study, and one that really only scratched the surface of some potentially interesting findings.  First, it pretty much confirmed what many already know about the relationship between ants and mealybugs.

Second, they found that the presence of certain weeds increased the ant and mealybug populations.  What would happen if the researchers were to exclude only those 6 species of plant that were found to

Photo by Lynn Betts / Photo courtesy of USDA Natural Resources Conservation Service. / Public Domain, via Wikimedia Commons

Photo by Lynn Betts / Photo courtesy of USDA Natural Resources Conservation Service. / Public Domain, via Wikimedia Commons

contain mealybugs?  Would it be just as effective in controlling the population as total cover crop (read an example here), would simply avoiding the use of the species of plants known to harbor mealybugs be just as beneficial in reducing the mealybug population than total removal of the entire cover crop?

Finally, I’d like to see a “spin off” study from this one examining the observation that Botrytis infections seemed to be more prevalent in vineyards with increased Argentine ant populations (and thus increased mealybug populations).  What’s really going on here?  Where is the cause and effect?  Correlation does not mean causation, so while the ant population and increased Botrytis infection seem to be correlated based on the results of this study, it does not necessarily mean that the presence of ants are causing this increase.  The results are certainly interesting, but I would like to see a lot more.

What do you all think?  If you need more details about the study, I’m happy to supply them.  Just ask!  Please feel free to share your thoughts, questions, etc.

Source:  Costello, M.J., and Welch, M.D. 2014. Influence of weeds on Argentine Ant (Hymenoptera: Formicidae) and Obscure Mealybug (Hemiptera: Pseudococcidae) in a Central California Vineyard.  Journal of Economic Entomology 107(3): 1194-1200.