Content warning: This blog post contains references to the hunting, agriculture, and research practices of killing birds. If you choose not to read on, I respect and admire your choice.

This multi-part series of blog posts was inspired by a Mallard hen that nested in our neighbor’s yard this spring. After her eggs hatched, the next-yard hen brought her nine ducklings to the dragonfly pond, where we all spent two lovely mornings basking in the relative safety of our tame little yard.

As I watched the hen rest and bathe and forage with her ducklings, I developed an obsession with Mallard physiology. This hen had incubated her eggs for almost the entire prior month, weeks and weeks of inactivity broken only intermittently to forage in nearby yards.

What happened in her body during that month? Especially in her flight muscles? After all, if I spent a month in bed, my muscles would deteriorate. And with the hen facing another dangerous stretch of weeks and months shepherding her flightless brood (ducklings don’t fly until they’re about two months old), what else was going to happen in her flight muscles? After three months mostly grounded, how could she fly at all?

Given my penchant for literature searches, I started looking for answers. A half-hour later my keyword nets were empty. Either I was choosing the wrong keywords, or the search engine ocean was empty, as well. But the search engine ocean is not the only source of information out there. Some answers are older than the internet. (Literature search side-quest unlocked!)

Giving up on keyword nets, I defaulted to my personal version of a bootstrapped search. I read through related papers, cherry-picked references that seemed pertinent, found the non-paywalled references, read more, picked more, and continued on repeat. Uncounted iterations later, I’ve devoted more hours to the search than can be explained by interest, alone. My OCD has clearly joined the hunt. (Obsession upgrade unlocked!)

Setting aside mysteries of my own neural wiring and firing, I’ve learned a lot about waterfowl. And about waterfowl physiology. So much so that I’m tempted to call myself a physiology hobbyist. And, like any good hobbyist, my current passions manifest in my blog.

Recap of Parts I and II

In Part I of this post, I described the anatomy of a bird’s flight muscles and shared a bit of personal history that helps explain why I am so fascinated.

In Part II, I defined what I mean by “knowledge”, reflected on the capitalism behind the curtain, sketched out some practicalities about Mallards and other waterfowl, introduced the literature’s euphemisms for “kill” (and also explained my choice to use the word “slaughter”), dipped into background about research numbers and repurposed data, and presented some findings from the literature about variations in the relative masses/sizes of flight muscles.

Here, I’ve harvested from the literature a few articles about annual changes in flight muscle mass in captive Barnacle Geese, wild Great Crested Grebes, wild Red Knots, and wild Mallards.

Flight muscles and annual cycles

All research is a tangled path, but wildlife research is a centuries long, thicket-strewn snarl of overzealous collection work, Larmarkian and Darwinian scuffles, rogue variables, and funding biases. Most of the research I’m citing here looked at flight muscle changes associated with molt cycles.

In all of the geese, grebes, and knots that embodied the data reported in these articles, the masses of their flight muscles decreased as their flight feathers molted and increased again as the birds regained feathers and flight. For most of the researchers who interpreted this data, these cycles of atrophy (muscle loss) and hypertrophy (muscle gain) were evidence supporting or refuting (for the species in question) a pair of proposed hypotheses.

The use/disuse hypothesis

One simple and obvious (hypothesized) mechanism for muscle gain and loss is use/disuse. When birds fly, they exercise their flight muscles and gain (or maintain) flight muscle through the known benefits of exercise. When waterbirds molt and replace all their flight feathers in a single weeks-long event, they lose muscle during molt because they quit flying. When they begin flying again, muscle returns.

Because simple and obvious tends to prove out (if you’re waiting for the obligatory Occam’s razor reference, here it is), I quickly became a fan of this hypothesis.

The “endogenous trigger” hypothesis

The more complex and less obvious (hypothesized) mechanism is an “endogenous trigger”. Perhaps somewhere in a bird’s body, some tissue or organ follows time (or seasons). Perhaps when the time/season is right, this tissue or organ sends a molt signal to the flight muscles, and the flight muscles begin breaking down. Maybe all that protein is needed for feather production (feathers are, for the most part, protein). Maybe birds with less muscle and therefore lighter body weights will regain flight sooner. Maybe some complex combination of diet, exercise, and behavior before and during molt causes muscle change as a side-effect, not as a benefit.

As complex and less obvious tends to make good storytelling (especially the kind of serendipitous discovery and cautionary tale stories that science loves), I quickly became a fan of this hypothesis, too.

Barnacle geese in molt

In the 2000s, researchers from the United Kingdom followed a flock of captive geese before, during, and after molt (or moult, because the UK).

Portugal et al. (2009) started with 40 adult Barnacle geese that had been bred and raised in captivity. These birds never flew. Their flight feathers were trimmed to keep them grounded in their aviary. Starting in July and continuing through November, the researchers slaughtered (anesthetized, euthanized, and later dissected) four birds from the flock every two weeks, with more frequent slaughter of birds during peak molt.

In this flock of flightless geese, flight muscle mass decreased by more than 35% in the weeks before molt and during the first stages of molt. After the mid-molt minimum, flight muscles started recovering, increasing back to the pre-molt maximum as the geese shed their old flight feathers and grew new feathers.

The researchers achieved this “35%” measurement by comparing the combined and averaged flight muscle masses of the slaughtered birds, four birds at a time. So the first four birds (the earliest data) had a combined and averaged flight muscle mass that was 35% heavier than the dissected and weighed, combined and averaged flight muscles of the four birds slaughtered mid-molt.¹

Setting aside (for the moment) the steady depopulation of this flock, there was a timing mismatch between the muscle and feather changes. Instead of flight muscle loss following feather loss (as a “use it or lose it” consequence of flightlessness), the flight muscle loss preceded feather loss.

But why was there muscle loss at all? These geese didn’t fly, so their grounded condition during molt was their default lifetime condition. How could use/disuse factor in, when there had never been use in the first place?

Behavior changed, too

These same Barnacle geese had been observed through the previous year’s molt. “Despite having constant access to food, the captive barnacle geese lost approximately 25% of their body mass during the wing moult in both years of the study” (Portugal et al., 2007, “Discussion”, para. 1). This is a substantial change in body weight for geese with unlimited access to food and water.

“Anticipatory, rather than responsive.”

So these 40 captive Barnacle geese experienced flight muscle loss prior to onset of feather loss, before their behaviors and metabolisms changed. And their flight muscles began recovering prior to feather regrowth, before the geese resumed normal behavior.

“Therefore, these muscle changes give potential for increased or decreased performance but do so in an anticipatory, rather than a responsive fashion” (Portugal et al., 2009, p. 2409).

That’s an unexpected conclusion. These geese were not experiencing a simple and obvious use/disuse effect. This was a complex and less obvious process. A science story waiting to be told. And uncovering evidence of the complexity cost (only?) 40 captive geese.²

What is the value of a few flocks of captive geese?

Returning to my Part II theme of capital, what is the most valuable capital in the paragraphs above? Portugal et al.’s data, which has racked up some 200 citations? The 2007 and 2009 publications, which have been viewed online some 5000 times?

Is the story I’ve borrowed for this post more valuable than the geese? After all, these were fully realized, fully alive adult geese. Portugal et al. noted that 31% of the flock were paired or attempting to breed (2009, p. 2407). They did not note if they slaughtered the pairs together. Would such a consideration soothe my empathy?

And, speaking of empathy…

At what point, if ever, did the behavior and stress-metabolism of the flock—so accustomed to safety, steady population density, and shared companionship—change in response to their sudden prey status and declining numbers?

Other waterfowl in molt

Between 1978 and 1986, a researcher in the Netherlands (Theunis Piersma) collected the carcasses of 112 adult Great Crested Grebes that drowned in gill nets during the birds’ August–October molt (or moult, because the UK version of English) on Lake IJsselmeer in the Netherlands. Pairing data with observations of the birds’ activity levels before, during, and after molt, Piersma interpreted his findings as use/disuse. As a cycle in Great Crested Grebes in which forced flightlessness triggered disuse atrophy during molt, and return of wing function triggered muscle hypertrophy after molt. (Piersma, 1988) ³

In separate work involving captive Red Knots, reported in 1999 (Dietz et al.) and more in keeping with Portugal et al.’s geese, Piersma (as a co-author with Dietz et al.) concluded a different mechanism was at work. In this instance, the authors concluded that an “endogenous circannual process” (p. 2836) regulated flight muscle changes in Red Knots during molt.⁴

All of this is good and useful information for researchers interested in captive geese, wild grebes, and captive and wild knots. It is even good and useful information for someone like me, who is dabbling through waterfowl research in search of a simple answer to a complex question about Mallards. It shows different physiological processes at work in different species.

In other words, my Mallard answers can’t be intuited from goose, grebe, and knot research.

So…the next-yard Mallard hen’s flight muscles?

Simple answers to complex questions are vanishingly rare in any field, but perhaps a complex answer can by synthesized? Have the simple and complex threads of other, related questions about Mallards crossed often enough to create a pattern? (Unnecessary spoiler alert: No such pattern is discerned here. Only more questions.)

And, is molting at all the same as nesting? (No. Obvs.)

I found a significant body of literature regarding flight muscle changes in Mallards during molt, but only a single flight muscle dataset for nesting Mallards in the wild. I expect ethical concerns explain much of the data imbalance. At least, I hope ethical concerns are a factor.

I prefer a world in which nesting and post-nesting hens, along with their eggs and ducklings and awkward teen-ducks, are safe from the traps and slaughter and scales of researchers. Their world is already dangerous enough.

Besides, even if everyone agreed on a single, simple mechanism for molt-related muscle loss and gain in Mallards, it’s unlikely that the consensus mechanism would also regulate muscle physiology during nesting. After all, molt and nesting share few behavioral, metabolic, or seasonal similarities. For the birds themselves, flightlessness may be the only common factor. And nesting flightlessness is, at least during the onset, choice—nesting hens can fly, they simply don’t fly often. Molt flightlessness is forced.

Mallards in molt

Venturing first into Poland, a 1990 article (Panek & Majewski) looked at Mallards in molt on the floodplain where the River Warta meets the Odra River in western Poland. During the time of the study, some 25,000 male Mallards gathered for molt, a population “many times greater than the number of local breeders” (p. 255). Molting Mallards (3,788 males; 341 females) were herded into net enclosures, weighed, examined, and banded, and then released. After periods of 3–9 days, more herding resulted in recapture of 337 male and 13 female birds, which were again weighed, examined, and released.

These efforts allowed the researchers to determine that both the male and female Mallards lost 12% of their body weights during molt. What’s more, whenever new feathers (even just a few new feathers) were damaged enough to require immediate re-replacement, the prolonged flightless period resulted in continued weight loss. In those cases, the continued weight loss couldn’t be blamed on the metabolic demands of massive feather regrowth because only a few feathers were being replaced. (Panek & Majewski, 1990, p. 258)

“In our opinion, limited foraging and the use of body reserves during flightlessness are responses to high predation on dabbling ducks that forage in shallow waters. Secretive behavior and short forays out of shelter minimize exposure to predation” (Panek & Majewski, 1990, p. 258).

But a hemisphere away in Klamath Basin, California, avian botulism has sometimes been a larger hazard for molting Mallards than predation. In some of the basin wetlands between 2001 and 2006, avian botulism claimed as many as 64% of radio-monitored Mallards during molt (Fleskes et al., 2010, p. 214).⁵ However, after molt, “Hunting was the main cause of mortality for post-molt Mallards both within (16 of 37 deaths) and outside Klamath Basin (six of nine deaths)” (p. 214).

“Increased daily mortality rates of light-weight birds that were captured late in the season during this study suggest some aspect of the molting marsh (e.g., food, water quality, sites safe from predators, predator density) deteriorated as the season progressed causing female Mallards in poor condition to be more susceptible to predation and disease” (Fleskes et al., 2010, p. 217).

Finally, in the Mingo Basin of Missouri, a researcher from the University of Missouri slaughtered a total of 267 female Mallards over the course of three successive winter seasons (1981–1983) (Heitmeyer, 1988). He found that molt timing varied according to age and weather. Adult females molted earlier than immature females, and all of the birds molted earlier in wet winters.

After processing the carcasses, Heitmeyer noted that the birds he slaughtered either before or after molt were heavy, with lipid reserves making up a high percentage of their body weights (1988, p. 673). In other words, the birds Heitmeyer slaughtered prior to molt were healthy and fit and well-prepared for the metabolic demands of molt, and the birds he slaughtered after molt were also healthy and fit and well-prepared for the metabolic demands of migrations to their nesting grounds.

But mid-molt? The Mallards he slaughtered mid-molt were 6% lighter in weight than pre- and post-molt birds. Most of this overall weight loss was due to a 35% decrease in lipid mass (compared to pre-molt birds). So used-up lipid reserves explained 83% of the weight difference between Mallard hens slaughtered prior to molt and Mallard hens slaughtered during the middle of their molt. The hens were losing mostly lipid reserves, not muscle. (Heitmeyer, 1988, p. 673 & “Table 3”, p. 672)

Do these three researches tell a common story?

Not really.

It’s tempting to weave these three researches into an intuitive story about Mallards that reads something along these lines: Mallards lose weight during molt because they hide from predators more and forage less, and their used-up lipid reserves (not atrophied flight muscles) represent most of the lost weight.

But science doesn’t work that way. Nothing does, really. I can’t take the 12% weight loss (Poland), explain it as 83% due to used-up lipid reserves (Missouri), and superimpose an estimate of up to 64% of molting Mallards dying due to disease (California). None of these numbers, variables, or Mallard populations are connected in any rigorous or meaningful way. The only commonalities are the English-language phrase “Mallards in molt” and this sprawling series of blog posts.

Even so, there are tempting threads. Perhaps Mallards in Poland, Missouri, California, and everywhere else actually do lose weight during molt because they hide from predators more and forage less, and perhaps their used-up lipid reserves (not atrophied flight muscles) actually do represent most of the lost weight.

Perhaps Mallards need a third hypothesis, something neither use/disuse atrophy nor annual endogenous trigger. Perhaps if I keep pulling this molting Mallards thread long enough, keep following it deeper into the rabbit hole that I already know doesn’t hold the answer I’m seeking, I’ll find other researchers pulling the same thread. Perhaps we’ll all agree that Mallards need a purely behavioral “hide and fast” hypothesis.

Except it’s time to follow this particular thread back out of the rabbit hole. Whatever mechanisms are at work in a molting Mallard’s physiology, they are (probably) irrelevant to a nesting Mallard’s physiology. (But, as I leave, I’m rolling up the thread metaphor and carrying it with me to the next rabbit hole.)

I like the thread metaphor because I like the idea of reality as a giant tangle of skeins. Step up to the skein, find a loose end, and start pulling. This is how some hypotheses unravel and how some hypotheses knot tighter.

But don’t forget that each thread has a price tag. Like a county fair booth where you pay 40 captive geese to pull the first thread. Or a boat full of drowned grebes to pull the next thread. Or, as in the next research, 51 Mallard hens to pull the specific thread I’ve been searching for all along.

Flight muscle changes in nesting Mallards

“By late incubation, females are highly emaciated; 11 live-trapped females weighed during the last 5 days of incubation averaged 900.3 g ± 30.1 g (mean ± SD)⁶, or 25% less than during prelaying” (Krapu, 1981, p. 31). (For readers accustomed to weights in pounds and ounces, 900.3 grams = 1.98 pounds and 30.1 grams = 0.066 pounds.)

While 11 ducks is a very small sample size, the data suggest that female Mallards lose up to a quarter of their body weight over the course of nesting. But do they lose flight muscles or lipid reserves? Or both?

Along with these 11 hens weighed alive, this study involved capturing and slaughtering 51 other Mallard hens who were at various stages of their nesting cycles: 19 pre-laying hens, 11 laying hens, and 3 hens that had completed the laying process and begun incubating their eggs. Plus 11 hens that were making their first nest and 7 hens that had lost their first nests and begun laying a new clutch.⁷

Here’s those numbers again, with a bit more context

If you read the previous two paragraphs and experienced a brief or extended period of dissonance, I’m with you. That’s a lot of numbers in just a few sentences. The important numbers, for my purposes, are the 25% weight loss, the 11 weighed hens, and the 51 slaughtered hens. Here’s a list of hens, broken down by how their data were sorted:

1. Eleven hens were trapped, weighed, and (hopefully) released back to their nests. These 11 hens were nesting within the study area, and each had a nest with eggs due to hatch within five days. All of these hens were in poor body condition (“emaciated”) compared to hens that had been weighed prior to laying.
2. Thirty-three hens were slaughtered after migrating into the study/nesting area. As the researchers dissected the 33 carcasses, the slaughtered hens were divided into groups based on their ovarian cycles:
   • Nineteen hens had not yet ovulated. These 19 hens were labelled as “pre-laying”.
   • Eleven hens had ovulated and begun laying eggs, but had not yet laid their last egg. These 11 hens were labelled as “laying”.
   • Three hens had laid all of their eggs (had no more eggs developing in their ovaries or oviducts) and had begun incubating their nests, but they were no more than 6 days into their incubation phase. These 3 hens were labelled as “incubating.”
3. Eighteen hens were slaughtered as their ovaries and oviducts began preparing for egg production, but before their first egg ovulated. (It’s unclear if these hens were counted among the hens sorted by ovarian cycle, above, so I’m counting them separately.)
   • Eleven of these hens were making their first nest. These 11 hens were labelled as “initial nest attempts”.
   • Seven hens of these hens had a “brood patch” (a bald/featherless patch on their chest or abdomen), which was considered to be evidence that they had already completed one nest and begun incubating (brooding). But something had gone wrong with the first nest, prompting the hens to restart their ovarian cycle and attempt a second (or third?) nest. These 7 hens were labelled as “renesting”.

(Did you spot the moment(s) when my OCD winced? There ended up being three different groups that numbered 11 hens. This kind of number coincidence is not exactly common in science, but also not exactly uncommon. My OCD does not like coincidences. It’s safe to say that, in general, science doesn’t either.)

Back to North Dakota in springtime

When Krapu compared the weights of various tissues and organs in his slaughtered hens, the laying hens (layers) had actually gained weight, compared to the pre-laying hens (pre-layers), while the incubating hens (incubators) had lost a significant amount of weight compared to both the pre-layers and the layers.

The idea that Mallard hens might gain weight in the early stage of egg laying makes intuitive sense. Think about birds and eggs and ovaries and oviducts. All of those eggs started as follicles in an ovary. Think about eggs in a nest. They’re certainly bigger than ovarian follicles. After all, each egg has to be fortified with enough proteins and lipids and sugars to build an entire duckling from scratch. So producing a nest full of eggs, ovary to nest, means a female Mallard’s reproductive tract gets huge.

As Mallard hens lay (on average) an egg a day during nesting, their ovaries and oviducts during this time often contain several eggs in various stages of growth from follicle to in-the-shell. Krapu’s data support this intuitive explanation. The layers had massively higher ovarian and oviduct weights than the pre-layers and the incubators. (I’m going to call this their pregnancy weight.)

In comparing hens slaughtered at these three stages—pre-layers, layers, and incubators—three trends of interest (to me) emerged:

1. The incubators had lost their pregnancy weight, and then some. Their ovaries and oviducts were not only lighter than the ovaries and oviducts of the pregnancy-heavy layers, but were also significantly lighter than the ovaries and oviducts of the pre-laying hens. (Have mercy. Statistical significance is its own hefty topic.)
2. The incubators’ lipid reserves were nearing depletion. The pre-layers’ total lipids made up some 10% of their overall body weight. For the layers, their total lipids made up about 6% of their body weight. But the incubators, by day 6 of incubation (and with some 3 weeks left to go), were whittled down to the point that their total lipids constituted only about 2% of their body weight.
3. The incubators’ flight muscles were lighter than the pre-layers’ and layers’ flight muscles, though the difference was not statistically significant.

And what about later in incubation? What about weeks 2–4? With lipid reserves already nearing depletion, muscle would be next on the menu. Thankfully, this research didn’t persist in slaughtering nesting Mallards. There are no numbers for weeks 2–4. But there are numbers for those seven hens that lost their first nests and tried to start over.

The seven renesting hens had already gained and lost their pregnancy weights once, and their body weights reflected the toll. They were about 12% lighter than hens at the same stage of laying a first nest (though they were slightly heavier than the incubator hens). Their lipid masses were only about 3% of their body weights, as they had used up much of their lipid reserves during their first nesting attempts.

And their flight muscles? In all the hens, no matter their nesting count or stage, their flight muscles made up 5–6% of their body weights.

	Pre-layers	Layers	Incubators	Initial Nesters	Renesters
Body weight	1199.8 ± 78	1300.6 ± 114.6	967.3 ± 44.5	1217 ± 79.4	1065 ± 54.6
Flight muscle weight	65.1 ± 5.6	65.1 ± 5.5	58.3 ± 1.9	65.3 ± 5.6	60.0 ± 1.9
Flight muscle as rough % of body weight	5.4%	5%	6%	5.3%	5.6%
Lipid mass	109.6 ± 33.7	79.6 ± 37.2	17.1 ± 14.7	116.4 ± 18.9	29.9 ± 17.4
Lipid mass as rough % of body weight	9.1%	6%	1.8%	9.6%	2.8%

Question answered? (No.)

Maybe nesting female Mallards don’t lose significant flight muscle. Maybe used-up lipid reserves and back-to-normal reproductive tracts explain all of that lost body weight, up to 1/4 of their pre-nesting weight. Maybe a nesting Mallard’s flight muscles only atrophy a little? (Unlike Western Grebes in Manitoba, Canada, which lose up to 41% of their flight muscle during nesting—males and females alike (Piersma, 1998, pp. 101–102 & Table 4).)

Maybe. But not likely. After all, the incubators had only been on their nests for up to 6 days. The renesters were still preparing to lay new clutches of eggs, still carrying new rounds of pregnancy weight, and hadn’t started incubating at all. The incubators averaged a weight of 967.3 grams (about 2.1 pounds) and the renesters 1065 grams (about 2.3 pounds).

Somewhere between renesting or early incubation and about 5 days prior to their eggs hatching (3 weeks or so), both incubators and renesters would have been expected to lose more weight. Perhaps even down to the weights recorded for those 11 captured-and-weighed (and hopefully released back to their nests) hens—about 900.3 grams (1.98 pounds).

If statistics mean anything, and if the 11 hens captured and weighed alive were at all representative of North Dakota’s nesting Mallard hens in the spring of 1981, all of the slaughtered hens’ weights would have fallen to about 900.3 grams (1.98 pounds) before their eggs hatched. Another expected weight loss equalling roughly another 7% of the incubators’ and renesters’ body weights. With lipid reserves already diminished, some notable proportion of that 7% would have been muscle.

But, which muscles?

Only the Mallards know.

And I’m content with that answer.

Happily, others are content with that answer, too:

“Wild Mallards breeding under natural conditions are poor subjects on which to accumulate statistically sound population parameters. The species is particularly sensitive to human interference, especially during the brood period. Statements such as ‘unstudied Mallard populations easily maintain themselves’ might be viewed as a general truism. Field workers concerned with duck population dynamics should periodically remind themselves of the Heisenberg Uncertainty Principle (TIME, Canadian Edition 04/15/63, p. 51), ‘the very act of observing or probing a phenomenon changes the phenomenon'” (Dzubin & Gollop, 1971, p. 49).

So, is this the end of these Mallard posts?

Of course not. I have OCD, and I’m perseverating on Mallards. But this is the end of my riff about flight muscle physiology.

The rest of my Mallard series will pull some Mallard hunting threads and some Mallard farming threads, which intersect at ongoing policy controversies surrounding releases of farmed Mallards into the wild.

---

Notes

1. I spy an uncontrolled variable! Because each two-week data set involved slaughtering four birds, in order to dissect and weigh their flight muscles, each two-week data set is an end point. Those four individual birds couldn’t be followed any further. So comparisons of the data sets, comparing the data recorded for the first four birds against the data recorded for any other four birds, requires an assumption that these birds had no significant individual differences. While this is a well-accepted research method, and while individual differences are unlikely to perturb or confound the conclusions, I’m putting a pin in this “individual variation” variable. (Click here to return to your regularly scheduled paragraph.)

2. Okay, yes, I agree. The geese weren’t the only cost. There is a lot of human labor behind this (and all) research. For the researches reported here, and because I’ve brought it up, it’s worth quoting the authors’ acknowledgments (BBSRC=Biotechnology and Biological Sciences Research Council, UK):

• “We are grateful to Alan Gardner, Phil Archer, Ben Heanue and Pete Jones, for looking after the geese. We are very grateful to Craig White for practical help with the birds and logistics, and Jon Codd and Peter Tickle for supplying us with anatomy guides. Thanks also to Graham Martin, Theunis Piersma, Caroline Chadwick, Robert Ker and McNeil Alexander for useful discussions, and two anonymous referees for their comments. S.J.P. and J.P.M. were funded by the BBSRC” (Portugal et al., 2009, p. 2409).
• “We would like to thank Craig White for his assistance with the respirometry equipment and set-up, and for statistical advice. We are also grateful to Alan Gardner, Phil Archer, Ben Heanue and Pete Jones for looking after the geese and helping with the weighing sessions. Thanks also to Peter Frappell for help with software, and Michael Romero, Graham Martin, Jim Reynolds and Lewis Halsey for useful discussions. This work was supported by the BBSRC” (Portugal et al., 2007, p. 1396).
• Click here to return to your regularly scheduled paragraph.

3. Halfway through wing molt, fewer grebes drowned in gill nets. The author notes that the half-molted grebes must either dive less often to forage or dive less deeply (p. 99). Grebes that drowned during wing molt weighed some 9–15% less than grebes that drowned midwinter, but this decrease in body mass involved mostly a loss of fat mass, which was 53–60% decreased during molt as compared to midwinter fat reserves. Despite the fact that most of the weight loss could be explained by loss of fat reserves, flight muscle masses were 28–30% lower in grebes that drowned during molt. So somewhere in the grebes bodies, some organ or tissue increased during molt, offsetting the muscle loss. The author suggested possible liver enlargement, as the liver processes proteins and feather replacement requires a significant investment of protein. (Piersma, 1988, p. 97) Click here to return to your regularly scheduled paragraph.

4. The authors looked at two subspecies of Red Knot that gather on the Dutch Wadden Sea in August. One subspecies, Calidris canutus islandica, undergoes wing molt in August and overwinters in western Europe and the Mediterranean before migrating to arctic regions in Canada, Greenland, and Svalbard for breeding and nesting (Baker et al., 2020, “Subspecies” para. 3, Dietz et al., 1999). The other subspecies, Calidris canutus canutus, stops on the Wadden Sea in August to build reserves in preparation for a 4500km (about 3000 mile) migration to western and southern Africa, where the birds overwinter (or oversummer, for the birds that cross into the Southern hemisphere) and finish their wing molt in March or April before flying back to Russia for breeding and nesting (Baker et al., 2020, “Subspecies” para. 2; Dietz et al., 1999).

In a rare (in my reading for this post) work that did not rely entirely on dissection to measure flight muscles, the researchers captured four individuals of each subspecies of Red Knot and transferred them into a climate-controlled aviary. Over the next eight months, all eight birds stayed in sync with their wild and free-living counterparts despite their controlled living conditions and forced flightlessness. The four C. c. islandica molted and lost muscle mass in August, in sync with their free-living counterparts (Dietz et al., 1999, Figure 1b,f). The four C. c. canutus gained weight and flight muscle mass in August, in preparation for an extraordinary migration they wouldn’t undergo, then lost weight and flight muscle mass as they molted in January–April, in sync with their own free-living, migrating counterparts (Dietz et al., 1999, Figure 1a,e). (Click here to return to your regularly scheduled paragraph.)

5. Note that “radio-monitored birds” always implies a small study set. Radio and GPS monitoring is expensive, labor intensive, and introduces a rogue variable in that many birds change their behaviors after being harnessed or otherwise burdened with devices. Fleskes et al. started with 181 radio-tagged female Mallards (p. 208). (Click here to return to your regularly scheduled paragraph, already in progress.)

6. Just an aside about those body weights: 900.3 g ± 30.1 g (mean ± SD). It’s math. Read aloud, the notation says “…a mean weight of 900.3 grams plus or minus a standard deviation of 30.1 grams…”. It essentially means that:

1. The average weight of these 11 ducks was 900.3 g (1.98 pounds).
2. But the “…plus or minus a standard deviation of…” part of the notation indicates that anyone who wants to use this number to predict how much any other 11 Mallard hens (captured in the same location, at the same time of year, and in the same nesting stage) might weigh, on average, should expect the prediction to be off by as much as 30.1 g (0.066 pounds).
3. So, if I want to open a county fair booth and guess the average weight of 11 Mallard hens (in North Dakota, in spring, who are incubating a nest of eggs that should hatch within 5 days), I should note in my fine print that as long as I am within 0.066 pounds of the correct number, I win. Then, as long as I always guess 1.98 pounds, I should win more often than I lose.
4. BUT, given that this number was derived from only one group of 11 hens in 1981, and given that hundreds and thousands of Mallard hens might simultaneously be incubating a nest of eggs that are within 5 days of hatching, in spring in North Dakota in 2025…? I think I’ll keep plugging away trying to earn a living as a writer and editor, because my I Can Guess the Weight of Your Mallards county fair booth is on shaky statistical ground.
5. If I jump ahead to Table 1 (p. 31), which reports an average ovarian weight for prelayers of 6.3 ± 8.7, I’m in a different statistical bind. That’s a worrying standard deviation number, because if I take the “plus or minus 8.7” at face value, my Guess the Pre-laying Mallard Hen’s Ovarian Weight county fair booth is going to be a hot mess. In this case, my fine print is going to state that I win if I guess within 8.7 grams of the actual weight. Every time I win after guessing a negative number, an interrobang will randomly manifest in a doctoral thesis from the 1940s (!?). (I can’t speak to what, exactly, produced a standard deviation so large that county fair booths and negative ovarian weights intersected in this footnote. Should any readers have ideas, please comment.)
6. Click here to return to your regularly scheduled paragraph, already in progress.

7. The Materials and Methods section of this paper is disappointing, as it doesn’t clarify where and how the comparison data were selected. The seven renesting hens were compared to a subset of 10–11 “initial nest” hens (Krapu, 1988, Table 3, p. 35), with no indication of whether these comparison hens were also included in the earlier analyses of hens at various laying stages. A close reading finds the Table 3 initial nest hens defined as hens in “rapid follicular development…pre-ovulating females with ovary weights ≧ 3.0 g” (p. 30), but this status should have applied also to some of the hens labelled as “pre-laying” in the analyses for laying stages. The author notes that 71 female Mallards were slaughtered over the course of this study, but the math doesn’t work. At most, even if I’m counting some of the prelayers twice, I get 19 prelayers + 11 layers + 3 incubators + 11 initial nesters + 7 renesters = 51 hens. Where are the other 20 hens? And why are there three (3!?) data sets of 11 hens here? (11 layers, 11 initial nesters, 11 late-nesting hens weighed and, hopefully, released…). My OCD doth protest. (Click here to return to your regularly scheduled paragraph.)

References

Baker, A., Gonzalez, P., Morrison, R. I. G., & Harrington, B. A. (2020). Red Knot (Calidris canutus). Birds of the World (CornellLab). https://birdsoftheworld.org/bow/species/redkno/cur/systematics

Dietz, M. W., Piersma, T., & Dekinga, A. (1999). Body-building without power training: endogenously regulated pectoral muscle hypertrophy in confined shorebirds. Journal of Experimental Biology 202(20), 2831-2837. doi: 10.1242/jeb.202.20.2831

Dzubin, A. & Gollop, J. B. (1971). Aspects of Mallard breeding ecology in Canadian parkland and grassland. Canadian Wildlife Services. https://publications.gc.ca/collections/collection_2024/eccc/cw66/CW66-1042-1971-eng.pdf

Heitmeyer, M. E. (1988). Body composition of female Mallards in winter in relation to annual cycle events. The Condor 90(3), 669-680. doi: 10.2307/1368357

Fleskes, J. P., Mauser, D. M., Yee, J. L., Blehert, D. S., & Yarris, G. S. (2010). Flightless and post-molt survival and movements of female Mallards molting in Klamath Basin. Waterbirds 33(2), 208-220. doi: 10.1675/063.033.0209

Krapu, G. L. (1981) The role of nutrient reserves in Mallard reproduction. The Auk 98, 29-38. doi: 10.1093/auk/98.1.29

Panek, M. & Majewski, P. (1990). Remex growth and body mass of Mallards during wing moult. The Auk 107, 255-259. doi: 10.2307/4087607

Piersma, T. (1988). Breast muscle atrophy and constraints on foraging during the flightless period of wing moulting Great Crested Grebes. Ardea 76, 96-106.

Portugal, S. J., Green, J. A., & Butler, P. J. (2007). Annual changes in body mass and resting metabolism in captive barnacle geese (Branta leucopsis): the importance of wing moult. Journal of Experimental Biology 210(8), 1391-1397. doi: 10.1241/jeb.004598

Portugal, S. J., Thorpe, S. K. S., Green, J. A., Myatt, J. P., & Butler, P. J. (2009). Testing the use/disuse hypothesis: pectoral and leg muscle changes in captive barnacle geese Branta leucopsis during wing molt. Journal of Experimental Biology 212, 2403-2410. doi: 10.1242/jeb.021774