How To Become a Test Prep Goon, Part 2

The first part of the application process was the simple application submission, phone interview, and the first checkpoint, the diagnostic exams, which I had outline in my previous post. Up to this point, everything had been conducted remotely by phone or email. However, now that I had passed the diagnostic exams, however, I was to come in for the face-to-face component of the process: the teaching audition, which would take place at the local company branch office.

The audition consists of a 5-10 minute presentation on a non-academic topic of the applicant’s choice, without using Powerpoint or other software, and with whiteboard and markers provided and props, as needed, brought by the applicant. This seems to be a standard prompt for the whole company as well as its competitors. Opinions on the Internet varied on how selective this step is compared to the diagnostic exams and training sessions, proper attire at the audition, and on various aspects of the presentation, such as the props, level of audience participation, and seriousness and complexity of the topic chosen.

I chose to talk about grafting, and specifically budding (the simplest of the myriad grafting techniques), due to my interest in gardening and plant biology. It’s a fairly dry-sounding topic compared to some of the other topics that people apparently have chosen (including falling in love, or crossing the street in Japan), but still original and interesting (compared to topics such as how to set up a chess board or play a sport). I’m also of the opinion that nothing needs to be intentionally gimmicky as long as it is conveyed with enthusiasm and backed up by extensive knowledge, and I felt sufficiently motivated by my own interest in plant biology and my gardening hobby to learn about the topic in depth. During the week I gave myself to prepare for my audition, I read online extension guides on budding, watched YouTube videos of gardeners demonstrating budding, and read relevant chapters from one recent, well-illustrated textbook on grafting. I also made sure to incorporate questions for the audience into my presentation (i.e., what might be some practical applications of grafting?), as well as a little dark humor.

The audition itself went relatively smoothly, since I was confident in my knowledge on the topic and had practiced a number to times to whittle my content down to 5-10 minutes. Some testimonials have said that auditions were done in front of the teaching staff and various other applicants, so I expected many more people and a lot more pressure, but this was not the case. The office was very small, and my audition took place in a tiny classroom in the back with only the very friendly regional coordinator and the course coordinator. I wore dress pants and a dress shirt for the occasion, which was surely still overdressing for this setting. Although I don’t remember the details of what I did during the audition (I tend to get selective amnesia during talks), I was able to exchange friendly banter with them for a few minutes and felt comfortable enough to ask for feedback on my performance, which seemed very positive. Though there were more applicants coming in to audition later on, I was told I had nothing to worry about.

Indeed, I was accepted to complete the next step this morning, sparing me a week of sitting around waiting to hear back, which was very considerate. At this point, after some paperwork, I would be moving on to the next step, being certified as a test prep instructor.


How to Become a Test Prep Goon, Part 1

Although I consider myself good at juggling multiple tasks when I need to, when I’m under no particular pressure I can’t help but fixate on a single goal at a time, especially if that goal is about to come to fruition in a matter of weeks. So during the past week, I’ve been working towards one lofty milestone: part-time employment!

“Let’s check on your undergraduate ambitions… Yeah, your dreams of going to med school right out of college? They didn’t make it.”

Granted, I’ve been very picky about what full- and part-time jobs I’ve wanted to apply for, based mainly on how I can develop myself professionally. I’ve made the most progress in my applications as an MCAT instructor for a major test prep company. After going through one of the company’s MCAT courses and accomplishing the score I wanted to get on that exam, I was motivated, and felt I was qualified, to at least apply to teach two of the MCAT topics as an instructor with this company. Though I did hear that the pay is well above minimum wage for instructors of these classes, I cared more about just having something constructive and meaningful to fill my time for the next two years, and I think most individuals who apply to teach for these test prep companies feel the same way. So I’m going to detail my whole experience applying and interviewing for this position below.

I began my application process a few days after I had finished the GRE Biology subject test, since I now needed something else to occupy my time. The application process itself was simple, if a bit frustrating. The portal for applicants to search for positions and submit applications is very bare-bones and prone to glitches: the interface is ugly and difficult to navigate, the login page requires you to ask where you heard about the company EVERY. SINGLE TIME., and it doesn’t save your information so you have to complete your information in one sitting. I was astounded that the resume text extractor even worked correctly. However, once I submitted my first application (to be an SAT instructor), I completed my next two applications in quick succession with no trouble.

I moved on and forgot about all this until mid-November, when I was invited by email to have a phone interview with the regional coordinator/manager. The interview itself was very cordial and not at all what I expected, as it was more a confirmation of my teaching experiences and an overview of the extended interview, training process, and teaching responsibilities. Though the company had enough local SAT instructors, it had a few openings for MCAT instructors, so only my MCAT applications were really considered. I was told to expect to take two diagnostic exams, one for each MCAT subject for which I had applied to teach. If I passed these, I would next come to the company’s local outpost for a brief teaching audition. Once I passed these two hurdles, I would receive paid training for three days to qualify as an MCAT instructor with the company. It was implied that the interview was more of a notification that I was selected to continue through this extended application process than a weeding-out step in itself. However, it definitely helped to seek information on the Internet beforehand about the position, the pay, and the company’s culture, and I was able to use that information to take the initiative at the start of the interview and ask my questions first.

The two diagnostic exams were sent to me by email, and I was given several days to fill in my answers and submit them. I took this time to review my class notes from the course, then emailed my answers and scratch work back to the regional coordinator for grading. The same day I submitted them, I was notified that I had passed both exams, and I subsequently scheduled my teaching audition for this afternoon at their office.


I is Smart

I checked my GRE Biology scores at 12:08 am this morning, after awaiting them for the past week and being blocked off the score report page all day Sunday by the “routine maintenance” ETS was undertaking on everyone’s accounts. I don’t mean to brag, but I was very pleased with my results:Screenshot 2015-11-23 00.26.13.png

In preparation for the real thing, I had taken the official practice test offered by ETS twice: first about a month before my exam date, before I had started studying, and second the evening before the exam, to gauge what I had learned. For comparison, that second time, I scored a 910 total (98th percentile), with an 85 (91st), 91 (97th), and 92 (99th) in the cell/molecular, organismal, and ecology/evolution sections respectively. These numbers were the benchmarks I’d set for myself going into the test at 8:30 am on October 24th.

Thus, my real exam scores reflect improvements almost across the board from the night before to the day of the test, in all categories except ecology/evolution. I attribute much of this to the last minute studying I did the night before the test, sitting in bed going through my wrong practice test answers and mastering concepts that I was supposed to have finished weeks before. However, I also believe a big part was starting the real test in medias res, by beginning with the data analysis problems (see my previous GRE post for further details).

This would also explain my unfortunate drop in the ecology/evolution section. By the new test-taking scheme, most of the questions I had saved for last were in the ecology/evolution recall section. Since these were my final questions and I was pressed for time, I definitely got sloppy, reading too quickly, not thinking thoroughly, and filling in answers for questions that were best left blank. It also didn’t help that I believed ecology/evolution to be my ‘easy’ section: On my practice test, the same topic saw me score in the 99th percentile both times, so I thought I could get away with it again without reviewing some concepts that were huge on the real test (looking at you, ecosystem ecology and equilibrium theory of island biogeography).

Overall, however, I am extremely satisfied with my scores on this GRE subject test. Although I came up slightly short in my strongest subject, the difference was insubstantial, especially considering my marked improvement in my two weaker subjects. By basically replicating my performance on the practice test, I did what I set out to do, and got the best results I could hope to use to top off my applications.

“So are you going to get over yourself and prepare for the general GRE ex-” “NO STOP LET ME ENJOY THIS MOM”

Generating Genetic Diversity Inside a Single Cancer Tumor

Survey a group of intro bio freshmen on what factors drive evolution, and natural selection (and its close allies, artificial and sexual selection) will assuredly be almost an exclusive answer. This is understandable, because Darwinian selection produces the most spectacular and most predictable changes in populations over time, from anatomy to physiology to behavior to development. But besides Darwinian selection, other forces also drive the evolution of populations: random drift of allele frequencies over generations, migration between neighboring populations, and mutation of genetic material itself. But these forces are so difficult to predict or observe that even biologists have a relatively poor understanding about these compared to Darwinian selection.

Given the delicate gene networks involved in regulating cell growth in the human body, our constant exposure to cancer-causing environmental agents, and just the sheer number of cells that are born and die over the course of a human lifetime, potentially cancerous cell lineages probably arise very often, but are almost always nipped in the bud by our immune system. The few cells that escape detection and establish tumors are also susceptible to medical treatments, including chemotherapy and outright removal. So it should make sense that mutated cells that do lead to cancer must have had some heritable traits that allowed them to grow rapidly and survive persecution and were passed down to their descendants: That is, cancer cells must evolve by Darwinian selection. But a new paper by Ling et al. out in PNAS asserts that this intuitive assumption might be a poor description of cancer evolution in the real world.

As the authors point out, simple math would predict a high amount of genetic diversity generated by mutation within a single tumor, which is essentially a big population of cancer cells. Though the expected mutation rate is only about 1% of the (functional) genome mutated per cancer cell division, an average tumor can grow to a population size of millions or billions of cells, so multiple mutations ought to occur during a tumor’s developmental history. If there is a strong selective force experienced by this population that favors certain cancer cell genotypes over others, one would expect very few of these mutations to survive; instead, the tumor would contain a homogenous genetic profile after the extinction of less beneficial mutations. Indeed, previous studies that sampled tumors found that each contains only tens to hundreds of distinct genotypes, so scientists simply assumed that Darwinian selection was taking place among cancer cells.

To test whether these estimates were accurate, Ling et al. analyzed a single liver tumor, just 3.5 cm in diameter, at high resolution. From 286 samples punched out of a mere 1 mm thick slice from the middle of the tumor, they verified and classified 269 mutations called single-nucleotide variations (SNVs) (other mutation types, based on abnormal numbers of gene copies rather than gene alteration, were ignored for logistical reasons). From these numbers, the authors determined that about 100 million coding SNVs would have occurred during the tumor’s growth into a population of billions of cells.

After sorting out which samples contained which SNVs, the authors were able to construct a full phylogeny, a family tree, to illustrate the history of diversification between all the samples. Interrelated samples were identified by shared SNVs, and major branches of related samples were classified into 20 families called clones, delineated by the SNVs unique to that branch. Interestingly, genetic relationships between the samples were also reflected in the structure of the tumor, in that closely related samples were located close together and nested within the same clone, and younger, more mutated clones were located closer to the edge.

Figure A displays a map of the tumor slice, showing where samples were taken and to which clone groups (denoted by color and Greek letter) they belong. Darker shades of the color indicate daughter subclones that have arisen within a clone group. For example, Figure B shows how the delta clone gave rise to a single daughter, which in turn gave rise to three daughters and a granddaughter of its own as it grew outward. Black stars indicate samples where mutations were found that are unique that sample and found nowhere else in the tumor slice, either because they were new or because they belonged to a subclone that was cut out of the sample.

If Darwinian selection was occurring within the tumor, one would expect certain clones to have grown faster, and thus be more prevalent, than other clones. Mutated genotypes that confer the greatest overall advantage on a cancer cell’s growth or survival would be swept to fixation in the population, crowding out cells unlucky enough not to carry those mutations. This was not the case, as the clone sizes instead corresponded closely to a null model in which selection was entirely absent. Indeed, the population contained many, many SNVs that each occurred within only a few cells in the tumor, as if mutations were generated rapidly but, being neutral in their effects on cell fitness, were never extinguished. Even the shape of the tumor and spatial distribution of the clones suggests that no cell lineages grew and divided outward at a faster or slower rate than others.

I can’t pretend to be an expert on this topic or the techniques and models the authors used, but I found the conclusions of this study to be quite robust, even conservative. Remember that their conclusions rest on characterizing a single type of mutation and searching for them only in the coding portion of the genome (that is, the 2% of the genome that code for proteins and so are directly responsible for a cell’s characteristics). If they were to have complicated their models by including the prevalent gene duplication events, or expanded their mutation screening to the 98% of DNA that doesn’t code for anything, it’s possible that they would have reached the same conclusions. In the future, I’d like more direct evidence for genetic diversification within a tumor as it grows, to complement evidence taken after the fact. Of course, this opens a whole can of worms on whether lineages of HeLa cells grown in vitro for innumerable generations are equivalent to tumors from a human or animal model.

Cancer is a broad category of diseases that develop in different ways and are caused by many unknown factors, and variables such as tumor structure can change how strongly Darwinian selection influences cancer cell evolution. Ultimately, the authors take a page from the rest of evolutionary biology, and suggest that future studies assume a null model of no Darwinian selection in order to evaluate hypotheses about selective forces in cancer. Clearly, the study of evolutionary processes in cancer, and how they shape disease trajectories and patient outcomes, is still in its early stages.

Baby Bees Grow By Eating Their Fungus With Their Pollen

Among biologists who study social insects, the development of sophisticated fungus agriculture in leafcutter ants and some termites is considered one of the landmark events in social insect evolution. In the most advanced species, the insects organize massive foraging trips for vegetation, converting it into a suitable medium on which fungal gardens proliferate. In exchange, the fungi compose the main diet of their gardeners, supporting vast nests of hundreds of thousands of individuals. We continue to make interesting discoveries on the ecology and evolution of these mutualisms, even though we’ve studied these systems to death in the past few decades. As it turns out, other animals as ecologically and phylogenetically diverse as bark beetles, sea snails, and damselfish have evolved ‘farming’ of some kind.

Being mainly subterranean insects that live in dark, humid nests, it would make sense that ant and termite lineages would have coevolved ancient and advanced partnerships with equally specialized fungi over millions of years. But what about other social insects? In The Insect Societies (1971), E.O. Wilson, noting that social bees and wasps seem to lack the accompanying communities of inquiline insects (highly specialized freeloaders) that inhabit ant and termite nests, proposed that this might be because bees and wasps tend to build more tightly sealed and inaccessible nests, sanitize nests more thoroughly of detritus, use more durable materials to build their nests, and construct them in arboreal environments rather than underground. I suspect that Wilson would have extended this hypothesis to microorganisms as well. Aside from my humble opinion that this explanation is excessively reductive and hand-wavy, Wilson’s observation is true. After all, the bees, despite their overall uniform ecology as nectar feeders and pollen hoarders, have spawned eusocial lineages just as advanced and varied as the ants and termites, so we ought to expect that some of these lineages have likewise evolved intimate associations with fungi in a convergent matter. Thus, my excitement at the news that we now have our first fungus-cultivating bee, as described by Menezes et al. in Current Biology.

Scaptotrigona depilis, an already known and rather well-studied tropical stingless bee, has a fairly typical social life compared to its relatives in the Meliponini. The colony’s home life is centered on the nest, a complex structure built out of cerumen, a mixture of wax secreted by young workers and resin brought in from outside. It is this cerumen that is used to build individual cells to store food and house bee larvae, which will develop into the next generation of adult workers. Cerumen is recycled throughout the hive to build and repair various other structures, and when a new queen leaves the mother colony to found a new colony elsewhere, a swarm of workers uses cerumen from the old hive to begin construction on the new nest. Atypically, however, the fungus, an unidentified species in the genus Monascus, also inhabits the cerumen, and as the bees reuse and transport cerumen, they spread the fungus throughout the nest and disperse it to new sites. In fact, the fungus seems to be so dependent on bees for propagating it that the authors found no evidence of the fungus producing conidia, the durable clonal spores generated by normal fungi. Instead, it grows exclusively as a webby mass of active growing cells called mycelium.

large Image
Progress of bee larva development from newly laid egg to mature larva. The mycelium only grows after the egg hatches, and completely disappears with the food store as the larva eats and grows.

Also like other social bees, after a new brood cell is completed, S. depilis workers fill it with a barfed-up soup of pollen and honey, enough to fuel the development of one bee larva to maturity. After the enormous queen deposits a single egg on the mess and shuffles away, the cell is completely sealed off; the baby bee hatches literally swimming in its own exclusive food mass. As Monascus grows on the food–in fact, the brood cells are the only place where the fungal mycelium actively grows–the larva actively consumes the fungus as well, so that nothing remains in the cell by the time the larva matures and begins spinning its cocoon. To test the importance of the fungus to bee larva survival, the researchers grew 300 S. depilis larvae in the lab on either UV-sterilized food or sterilized food that was then retreated with the fungus. While 76% of larvae fed food containing the fungus survived, only a piddling 8% of larvae fed sterilized food did the same.

large Image
In all of the colonies tested, bee larvae survived at much higher rates when raised on fungus-infested food than on sterilized food.

Clearly, the fungus was playing a key role in bee development. But what was it doing? During the lab experiments, they found that sterilized bee food quickly spoiled, but food containing Monascus stayed fresh. Incidentally, other species of Monascus are known to secrete antimicrobial agents, and the fungus has been long used by human cultures to preserve and ferment food. However, when Menezes et al. tested the effectiveness of bee food against E. coli and staph bacteria compared to the antibiotics penicillin and streptomycin, they found it was not effective all. (Newly made bee food was as effective as the antibiotics in inhibiting growth, but this was probably due to properties independent of the fungus, similar to the way that honey in your cupboard doesn’t spoil). Perhaps the fungus is only effective in excluding other microbes more specialized in growing on the bee food or plays another role entirely.

Overall, this paper’s main significance was in providing the first proof of a bee relying on a close association with a fungal ectosymbiont, and in this respect the authors did demonstrate that bee larvae require the fungus to develop. However, the results were preliminary, and I was dissatisfied that no experiments were performed to better confirm the mutualistic nature of this relationship. I would have liked to know how well the fungus grows outside of S. depilis hives and on media besides bee larva food, or if larval mortality without the fungus does result in decreased worker recruitment and lower colony fitness. I was also personally skeptical of the comparison of the bee-fungus mutualism with advanced agriculture in ants and termites, since the bees’ adaptations to inoculate the fungus, cultivate it, and utilize it in their diet were either not proven or not noticeably different from other stingless bees in general. I agree with the authors that the disappointing lack of obvious and elaborated anatomic and behavioral adaptations is more akin to other examples of proto-farming.

Sourdough miche & boule.jpg
To make an analogy, it seems S. depilis’s version of fungus cultivation is more similar to our species’s propagation of sourdough culturesfor bread, rather than wheat or corn agriculture. After all, you don’t see bakers weeding their dough crops very often.

Of course, these problems can be addressed in the many interesting and potentially fruitful future directions that can be taken with this research. I’d especially like to know if the unknown Monascus in question is a distinct species exclusively associated with S. depilis bees, or a unique form of an unknown that grows facultatively lives alongside other bee species or in other environments. I’m also confident that we will uncover more examples of fungal ‘agriculture’ in the large genus Scaptotrigona, among the diverse meliponine bees, and in other social bee lineages in general. The lesson is that when we don’t know about species natural history in ‘well-studied’ groups, we are prone to overgeneralizations and overlooking major discoveries. In this respect, this paper deserves to generate more buzz among scientists.

My thoughts on the GRE Biology subject test

I took my GRE Biology subject test just this morning. For those unfamiliar with the Biology subject test, it is given only a few times every year, once in the spring and twice in the fall, and contains 190 multiple choice questions to be answered in 2 hours and 50 minutes without breaks. The material is as broad as the field of biology itself, with some questions about very specific concepts and others more general, and is roughly equally divided among the subfields of cellular/molecular biology, physiology/organismal biology, and ecology/evolutionary biology.

Questions are multiple choice, and grouped by question type. The first section of the test mostly asks for straight information recall and rapid calculation/figure analysis, the second asks for matching entries in lists or diagrams with respective descriptions, and the third mainly tests critical analysis of experiments, tables, and figures. In calculating raw total and subfield scores, 1 point is given to each correct answer, 1/4 point is deducted for each incorrect answer, and 0 points are deducted for questions left blank. Raw scores are converted into scaled scores, which are assigned a percentile rank.

Here are my personal reflections on the test:

  • This summer, I’ve been focusing all my attention on the MCAT. It was only after that was over that I realized I still had time to register for the October subject test. Since I’m a little unsure about my future plans, I figured this was worth a shot, especially since the next test date wouldn’t be until April 2016. Unfortunately, because I hadn’t bothered to register until 5 weeks before the test, all the seats closest to me were all out; I had to sign up for a test location at a small college an hour’s drive away. It was somewhat unpleasant to have to wake up in the darkness of 6 am for a 3 hour test on a Saturday morning. Obvious advice: Don’t register for tests without planning ahead. 😦
  • I was surprised at the rather casual atmosphere just before the test. Part of it was the fact that a lot of the test takers were students at the university who knew each other, part of it was the very understanding, laid back, and experienced proctor we had, and part of it was the fact that I was stressed out in the minutes before. Even so, having taken the MCAT, I had expected a whole array of security and procedural ceremonies coming into the test, but we all just talked in the lobby and then sat down when it was time for business.
  • The material was more predictable than what I expected. ETS itself has said that the scope of the Biology subject test is so extensive that “no one is expected to be familiar with the content of every question.” Thus, I expected to be tested on very different material than the official practice test. Though there were some topics that I should have studied more intensively, the surprise was actually how well the material corresponded, not only with my old high school Campbell & Reece textbook, but also with the practice test. It was definitely worth going through the practice test again to compare my answers with the key and learn what I got right and what I did wrong.
  • Then again, I probably should have started studying a little earlier. Part of it was because I did quite well during my first pass through the practice test, so I got lazy and turned my attention to other work. I lingered learning plant anatomy and growth, and didn’t end up actually *studying* until maybe 2 days prior to the test. I was still scrambling to get the basics of plant transportation and development this morning before my trip. It also didn’t help that I slept less than 6 hours last night due to my anxiety.
  • Guessing on plant bio like
“I got this, the answer is A!” *aggressively bubbles in C*
  • My approach to taking this test was a little different than I think most people would expect. I started in the middle of the test in the matching section, to settle my brain into exam-taking mode. I went all the way through into the analysis section, where each question takes the most time (over a minute per question at least to parse out figures and numbers). After reaching the end of the test, I went back and only then started at the beginning of the test; I gave myself 15-30 seconds for each recall question. I’ve found that starting the test in medias res allows me to complete the analysis section, the most difficult and time-consuming questions, before I tire out or run short on time at the end of the test. The recall questions themselves are usually straightforward and stand alone, so I find it vastly preferable to be going through these rapid-fire know-or-don’t-know problems during the last few minutes of a test, rather than be stuck on analysis problems I could have gotten right if I only had more time.
  • I’m still unsure whether taking this subject test was worth the time and money. At least a few top graduate programs in EEB ‘highly recommend’ submitting Biology subject test scores, which I take as basically being the same as ‘required’. But it seems many more programs don’t care about the subject test scores in admissions (and for good reason, in my opinion), and after surfing various boards and talking to graduate students, I believe that most applicants don’t bother with the subject test either. Of course, scoring very high will be impressive regardless, and I think I need to have performed well all the more to at least partly compensate for the deficiencies in my potential graduate applications.
  • Overall, I am quite satisfied with my performance. I felt about the same taking the real thing as I did the practice test, so I don’t believe I did too shabby. After the test, I had a snack, looked over what material I might have missed, then had a nice drive back home. In about a month I’ll find out how I did!

Helpful resources:

GRE Biology practice test

GRE subject test percentiles

GRE subject test subscore percentiles

Compsilura concinnata: A Hidden Force Behind The Decline of North American Moths?

A female Compsilura concinnata tachinid fly. By Tony T, from the forums.

National Moth Week is coming up in less than a month, but experienced mothing enthusiasts are already deep into the summer season. The past few months have seen the steady emergence of the most charismatic family of moths in North America: the Saturniidae, the giant silk moths. American naturalists have admired the huge, colorful moths, and their equally large and ornamented larvae, before there was even an American nation to speak of.

But even their large size and fame have not saved them from the environmental assaults of the 20th century. The population declines observed for some saturniid species has been well-documented in regions like New England, where the cecropia moth (Hyalophora cecropia), regal moth (Citheronia regalis), and imperial moth (Eacles imperialis) are now rare and/or extinct. In fact, evidence is suggesting that many other moth species are also experiencing regional declines. Many hypotheses have been put forth to explain their disappearance, including habitat loss, electric street lamps, and the indiscriminate use of pesticides. But could it be a plain-looking fly that holds the key to the loss of giant silk moths in parts of the United States?

The European fly Compsilura concinnata is one of a variety of flies in a diverse family, the Tachinidae, that has evolved a curious lifestyle called parasitoidism in which the larval stage feeds and grows on a single host insect’s tissues, like a regular parasite, but always kills it, like a slow-motion, single-target predator. But unlike many parasitoids, which selectively attack only a few species, C. concinnata has very catholic tastes, with a recorded host range of about 180 species ranging from butterfly and moth caterpillars to even beetle larvae and sawfly larvae. The mother fly, discovering a suitably-sized caterpillar, swiftly dive-bombs the host and uses her sharp, hook-like ovipositor to inject eggs, already containing fully-formed embryos, inside its body. These immediately hatch and burrow into the midgut, where they feed on the caterpillar’s gut contents. During this time, they evade the host’s immune response and hijack its respiratory system, attaching themselves to the caterpillar’s network of tracheoles (breathing tubes) near the midgut. When the caterpillar completes development and stops feeding in preparation for metamorphosis, the maggots inside suddenly feed and grow rapidly, killing it. After scavenging the rest of their dying host, the larvae bore out of its body to complete their own development into adult flies.

spot behind head

A polyphemus moth caterpillar (Antheraea polyphemus) infected by tachinid fly larvae. By LunaGraphics2005, from Flickr (beware, the pictures in this series get nasty)

Knowing that it feeds on caterpillars during its larval stage, entomologists initially released C. concinnata into the US in 1906 in a bid to control the invasive gypsy moth, Lymantria dispar, and browntail moth, Euproctis chrysorrhoea. Sporadic introductions of the fly continued throughout the 20th century. Today, we now know that C. concinnata does play an important role in controlling populations of L. dispar, and may have almost single-handedly eradicated E. chrysorrhoea from most of the United States. Among its compatible hosts are other multiple agricultural pests, though its effectiveness in controlling those species is unclear. But what was the price paid by native butterflies and moths?

Even then, the scientists knew that the fly wasn’t picky in choosing what its larvae would eat. Not only can C. concinnata attack caterpillars besides gypsy moths, it is required to, since its fast generation time is incompatible with the gypsy moth’s longer life cycle. A single field observation noted that the fly accounted for most of the parasitism on the buck moth Hemileuca leucina. However, work attempting to characterize the extent of this damage in the field didn’t begin until 1995, when a team of scientists at the University of Massachusetts-Amherst set out to describe the effects of the fly on wild silk moth mortality. In a series of field experiments in which caterpillars were exposed to the outside environment, C. concinnata infected nearly 70% of Callosamia promethea caterpillars in the course of just 6 days. In another experiment, they showed that Compsilura parasitism caused a vast majority (81%) of mortality in young caterpillars of Hyalophora cecropia; in the end, not one of the 500 H. cecropia caterpillars survived to pupate. Another study done in Virginia showed that C. concinnata flies accounted for 28 of 36 parasitoids recovered from luna moth caterpillars (Actias luna) over the course of the summer of 2001.

Of course, the most obvious problem caused by the introduction of C. concinnata to North America is the threat it poses to butterflies and moth populations, arguably the most cherished component of our native insect biodiversity. And this is certainly an important issue: Not only are they enjoyed by the general public, they play an important role as herbivores, often specializing on a few species each, and also provide food for predators. But it is also possible that C. concinnata, with its generalist appetite and enormous impact on native caterpillars, may be directly competing with native specialist parasitoid species that have evolved to parasitize only a few species. Although the fly inhabits the host midgut, and can thus coexist with parasitoids that develop in the body cavity or other organs, the heavy burden of sustaining multiple larvae outpaces the host’s capacity to nourish all its parasites, resulting in higher mortality and smaller body sizes for all the parasitoids that develop.

There is some hope, however. C. concinnata is itself attacked at high frequencies by its own parasitoids, called hyperparasitoids, which are small wasps that access the fly larvae by laying their eggs onto leaves that are ingested by host caterpillars. The wasp larvae develop as internal parasites of the fly larvae, ultimately killing them (but only after the maggots have already emerged from their host). Some evidence shows that properly managing and protecting habitats can mitigate the parasitism rate of C. concinnata on at least one silk moth species, the buck moth Hemileuca hera. Most of the moths whose declines have been documented are still relatively common throughout the United States, though some species attacked by the fly are considered threatened in certain regions. Overall, however, it is still impossible to draw definite conclusions and set conservation policy without more data on the ecology and population dynamics of the hosts, their parasitoids, and the parasitoids’ parasitoids, all of which are understudied in an environmental context. If anything, the story of C. concinnata has been increasingly cited as an example of biological control going wrong, and hopefully will inform future efforts to utilize natural enemies as a counter to the slew of invasive insects in North American forests.