Myth-busting: Birds, buckthorn, and diarrhea

[This post is in response to local interest in the effects of buckthorn fruit on the birds that eat it.]

For the last 5 or 6 years, my research has focused on the use of fruit by birds in the fall, especially non-native fruit and migratory birds. One of the most abundant fruiting shrubs in my southeast Michigan study area is Common Buckthorn (Rhamnus cathartica), an invasive fruiting tree not native to North America.

Before I continue, let me issue this statement: I strongly believe in the use and preservation of native plants. I am not "pro-buckthorn." I believe it has many ecological liabilities. One prevailing notion is that buckthorn causes diarrhea in birds and is harmful to them. Based on a peer-reviewed literature search, I cannot find evidence that this is true.

Much of the background information here is from an excellent overview paper by Izhaki (2002). Other research is also cited.

Background: Chemical properties of buckthorns

Many plants have what are known as "secondary compounds," chemicals that play important roles in plant fitness and survival. These compounds or metabolites are generally believed to deter the consumption of unripe fruit, since the seeds in unripe fruit are not yet viable. See Cipollini and Levey (1997) for a discussion of other very interesting hypothetical functions of secondary compounds in ripe fruit.

The relevant secondary compound in buckthorns is emodin, a free-form (aglycone) anthraquinone found in 17 families, 28 genera, and 94 species of plants, including 23 species in the genus Rhamnus*. It can be found in many plant parts, including the leaves and fruit. Emodin has many properties. In regards to fruit, emodin can act as an antimicrobial, inhibiting fruit damage. Levels of emodin in fruit pulp typically decrease as the fruit ripens. As explained above, this may help to protect the seeds from being dispersed before they are mature, or eaten and destroyed by seed predators.

The biological activity of emodin (as well as many other compounds) is different in mammals than in birds. Emodin is well-known as a laxative in humans as well as other mammals. What about birds?

Buckthorn and birds

Wherever it occurs, buckthorn tends to be distributed by birds. There are two commonly held and frequently repeated notions about buckthorn fruit: that it causes diarrhea in birds, and that birds in North America don't "know" not to eat unripe fruit and can become very sick or even die from eating it. (Actually, there is a third notion, that buckthorn fruit is "junk food." That isn't entirely true either, but it a subject of a future post.)

The diarrhea myth, I think, starts with the fact that when people see birds eating fruit, they also see the birds poop a lot. This is, in fact, true of most birds and most types of fruit, especially wet, pulpy fruit and fruit with bulky seeds. Fruit passes quickly through bird guts -- many fruits are composed of mainly water and seeds, and nutrients (especially sugars) that are quickly assimilated. Birds only have so much room in the digestive system and it makes sense for them to process fruit as quickly as possible, including jettisoning the seeds. Birds are also able to consume more fruit if they choose species that have short retention times (Cipollini and Levey 1997). Hence, the appearance of diarrhea. I suppose if you define diarrhea as loose, wet, and frequent defecation, then many fruits give birds "diarrhea." But the implication in public discussions is that the frequent, conspicuous defectations of buckthorn pulp is abnormal or harmful. I've seen similar messes near large stands of Pokeweed (Phytolacca americana) and Red Mulberry (Morus rubra), both native plants, and do not hear similar concern for birds' health.

Studies of secondary compounds in fruits (e.g., Wahaj et al. 1998) have indicated that plants exert "control" over the gut retention time of their seeds in birds through these chemicals; this includes both speeding up and/or slowing down of the passage through the digestive system.  This is presumably to influence the number of seeds expelled per defecation, the dispersal distance, and/or reduce or enhance the effects of gut passage on the seed coat. This control is advantageous to the plant if it increases its reproductive success.

Regarding emodin, Tsahar et al. (2003), working with Yellow-vented [White-spectacled] Bulbuls (Pycnonotus xanthopygos), found that emodin increased digestive efficiency in birds, aiding in the processing of dry matter, nitrogen, and other nutrients. They determined that this effect was the result of emodin slowing transit time of food through the gut. This study, then, found that emodin has a costive (constipating) effect, not a laxative one, at least in bulbuls.

In their review, Levey et al. (2007) noted, 

Despite the widespread detrimental effect of emodin on fruit consumption by vertebrates, ripe fruits that contain emodin are frequently consumed by a wide variety of seed dispersers... suggesting that the net benefits of fruit consumption somehow outweigh the negative effects of emodin. In particular, the nutritional reward of carbohydrates, lipids, and proteins in fruit pulp can be viewed as more important to consumers than the presence of co-occurring secondary compounds.

Do North American birds unwittingly eat unripe non-native buckthorn fruit, which contain higher concentrations of emodin than ripe fruit? Could they consume enough to harm them?

There are at least 14 native species of Rhamnus and Frangula in North America. Several species are widespread, and have similar fruit characteristics (phenology, size, color) as the non-native species. Given the widespread occurrence of emodin in this family, presumably it also occurs in the fruit of the native species. This indicates that North American birds are not naïve to buckthorn or emodin.

Even if they had no experience with fruits containing emodin, studies have shown that many bird species are sensitive to it and adjust their consumption accordingly. Tsahar et al. (2002) found the highly frugivorous Yellow-vented Bulbul could distinguish among artificial diets containing a range of emodin concentrations similar to what is found in ripe fruits, while House Sparrows could only discern emodin amounts similar to what is found in ripe versus unripe fruits.

Both of those bird species are Old World in origin. One of the only studies to take a long look at R. cathartica and North American birds was a doctoral dissertation by James Sherburne (1972). He found American Robins avoided feeding on fruits of other plants coated with emodin, and that even if starved for 12 hours, would only voluntarily eat a few unripe R. cathartica fruits. The unripe fruit contained between 0.6 to 1 microgram of emodin.

Sherburne went on to force American Robins and Gray Catbirds eat unripe Rhamnus fruit or capsules containing emodin. The birds showed signs of diarrhea after consuming 50 micrograms to 5 mg of emodin. At the lowest concentration showing effect (50 micrograms) and the highest concentration given for the unripe fruit (1 microgram), it would require ingestion of at least 50 unripe fruits to cause diarrhea.

Schafer et al. (1983) tested the toxicity of a wide range of compounds on birds, including emodin. The dose of emodin needed to kill half (LD50) of Red-winged Blackbirds and European Starlings was greater than 100 mg per kg of body mass. Under this scenario, an average-sized 65-gram male blackbird would need to consume at least 6500 unripe fruits of R. cathartica (based on the high estimate of 1 microgram of emodin in an unripe fruit from Sherburne) to reach the "LD50" dose. Little wonder the authors calculated a "hazard index" indicating little or no potential for emodin to cause acute poisoning in these species.

Given the ability of birds to detect emodin, their reluctance to eat unripe fruit (three other studies cited in Izhaki [2002] noted that in North America, "most bird species do not consume the unripe fruits"), and the high levels needed to create a negative response, we can probably infer that wild birds do not consume large enough quantities of unripe fruit to cause diarrhea or serious toxicity.

As for ripe fruit, Sherburne found no emodin in ripe fruits of R. cathartica, although other studies have found it in the ripe fruit of other Rhamnus species not yet established in North America (Tsahar et al. 2002). As it is found in far smaller amounts in ripe fruit, the number of fruits that would need to be consumed to cause harm would be many times greater than the examples listed above. Sherburne reported on 11 species of birds that readily ate ripe R. cathartica fruit, but did not mention any negative effects (in Knight et al. 2007).

In the wild, birds tend to mix their diets, choosing fruits and other foods within short feeding bouts, and this may be an attempt to balance nutrient intake or to avoid ingesting too much of a particular secondary compound (Cipollini and Levey 1997) . Further, the varied dietary inputs may contain their own qualities that could mediate any adverse physiological effects of emodin.

There are many ecological crimes we can pin on buckthorn. It doesn't appear that causing harmful diarrhea is one of them. If anybody has seen any other studies or peer-reviewed publications that do show that eating buckthorn fruit is harmful to birds, please post them in the comments. I am fascinated by this topic and interested in learning of new research.

-----

*Glossy Buckthorn, Frangula alnus, has often been included in the genus Rhamnus as R. frangula. It also contains emodin. When I refer to "buckthorn" here, I am referring to both Common and Glossy Buckthorn in general, unless otherwise noted.

Cipollini, M. L., and D. J. Levey. 1997. Secondary metabolites of fleshy vertebrate-dispersed fruits: adaptive hypotheses and implications for seed dispersal. American Naturalist 150:346-372.

Knight, K. S., J. S. Kurylo, A. G. Endress, and J. R. Stewart. 2007. Ecology and ecosystem impacts of common buckthorn (Rhamnus cathartica): a review. Biol. Invasions 9: 925-937.

Izhaki, I. 2002. Emodin -- a secondary metabolite with multiple ecological functions in higher plants. New Phytologist 155:205-217.

Levey, D. J., J. J. Tewksbury, I. Izhaki, E. Tsahar, and D. C. Haak. 2007. Evolutionary ecology of secondary compounds in ripe fruit: case studies with capsaicin and emodin. Pages 37-58 in A. J. Dennis, E. W. Schupp, R. J. Green, and D. A. Westcott, eds. Seed Dispersal: Theory and its Application in a Changing World. CAB International, Cambridge, MA.

Schafer, E.W. Jr, Bowles, W.A. Jr, and J. Hurlbut. 1983. The acute oral toxicity, repellency, and hazard potential of 998 chemicals to one or more species of wild and domestic birds. Arch. Environ. Contam. Toxicol. 12:355-382.

Sherburne, J. A. 1972. Effects of seasonal changes in the abundance and chemistry of the fleshy fruits of northeastern woody shrubs on patterns of exploitation by frugivorous birds. Ph.D. dissertation, Cornell University, Ithaca, New York.

Tsahar, E., J. Friedman, and I. Izhaki. 2002. Impact of fruit removal and seed predation of a secondary metabolite, emodin, in Rhamnus alaternus fruit pulp. Oikos 99:290–299.

Tsahar, E. 2001. The impact of the secondary metabolite emodin in Rhamnus alaternus fruits on fruit removal and seed predation. M.S. thesis, Tel-Aviv University, Israel.

Tsahar, E., J. Friedman, and I. Izhaki. 2003. Secondary metabolite emodin increases food assimilation efficiency of Yellow-vented bulbuls (Pycnonotus xanthopygos). Auk 120: 411-417.

Wahaj, S. A., D. J. Levey, A. K. Sanders, and M. L. Cipollini. 1998. Control of gut retention time by secondary metabolites in ripe Solanum fruits. Ecology 79:2309-2319.

Fall banding: Weeks 8-10

Good riddance to a wet September. October has been occasionally wet or windy, but overall the weather has been much better. Unlike some Octobers, we have had only two mornings with frost, which can delay us from opening in the morning if the nets get stuck shut. On the downside, deer are mating and they've managed to get in the banding area regularly. One night a buck was presumably chasing a doe and they ran into the gate of the chain link fence so hard they bent open the latch! We are now operating with a reduced number of nets, and all but two of the ones remaining are damaged from their activities.

The number of birds we have banded so far in October represents over 40% of the total for the season. Most have come after mid-month, when American Robin numbers increased, and the first big push of Yellow-rumped Warblers, Hermit Thrushes, and sparrows began. It is heartening to see respectable Hermit Thrush numbers. The number of Swainson's Thrushes banded this fall was our second lowest. We have already hit our average for Hermit Thrushes, and are getting plenty of seed samples from them, which is excellent. We're still below average for White-throats, and they don't seem as numerous this year as usual. We have only banded four White-crowned Sparrows, which is ridiculously low. Our average is 22.

White-throated Sparrow.

Warblers continue to be banded. In addition to Yellow-rumps, late species such as Common Yellowthroat, Orange-crowned, and Palm still are around, as is Tennessee. Every day we still get a Nashville Warbler or two, and we are far over our fall high of 59 Nashvilles banded. Notable was a Blackpoll Warbler banded on 22 October that tied the late fall date for Dearborn -- but one seen the next day furnished the new late date. An American Redstart was also banded  on 22 October, and that was a new late date for that species.

Orange-crowned Warbler.

Two species were banded this month that we don't get too often: Eastern Towhee and Yellow-bellied Sapsucker.

Young male Eastern Towhee.

Young female Yellow-bellied Sapsucker.

We are now around 1000 new birds banded of 72 species, approaching average but with a capture rate (based on number of nets and hours opened) still well below average. Typically we band into the first week of November, but we may go longer than that this year if the weather holds. The next banding update will be a summary of the season.

Poison ivy: Breakfast of champions

Our study of the fruit composition of the diets of fall migratory birds focuses on Catharus thrushes and the super-abundant non-native fruits found in urban areas. However, we do look for seeds in the "samples" (droppings) provided by all the birds we band. This helps us examine which other bird species are eating fruit, what types of fruit they consume, and whether the diets of different bird species favor different fruit species.

One of the native fruiting plants that is common on our site is poison-ivy (Toxicodendron radicans). The flowers are rather insignificant and can be overlooked.

They do result in fruit, and large mature vines produce good crops of small, off-white drupes. This often occurs overhead, where the vine has climbed up a tree and gets good sun exposure.

This portion of poison-ivy is growing on a fence where it gets direct sun, and each fall it has a lot of fruit.

Poison-ivy fruits themselves are about half the size of a pea, quite dry, and papery. They contain very little pulp. Each fruit has one to three, odd-looking, globular seeds that are very distinctive.

We find the seeds in samples from only a few species of birds. Most often it's woodpeckers, and we've gotten poison-ivy seeds from Downy and Hairy Woodpeckers, and Yellow-shafted Flickers. The sample below is from one dropping from a flicker.

The three seeds in the upper right are from wild grapes (probably Vitis riparia). The other 13 are poison-ivy. Over the past several years, the only other bird species contributing poison-ivy seeds has been Yellow-rumped Warbler.

When we find seeds in a Yellow-rump sample, it's always poison-ivy. The other day I was holding a Yellow-rump while writing down some data. Suddenly, a single poison-ivy seed shot out of the warbler! Usually, we just collect the seeds that are deposited in the holding bags we use to transport the birds from the nets to the banding lab.

This week, we got two poison-ivy seeds from a Ruby-crowned Kinglet, the first seed samples of any kind we've had from a kinglet. This particular bird was first banded on 7 October, when it weighed 6.3 grams. We recaptured it several more times. On 12 October, it weighed 6.7 grams, on 16 October it was 7.2 grams, and on 18 October it was 7.4 grams. It doesn't sound like much, but that's a 17.5% increase in the bird's original weight.

Poison-ivy fruit are very high in fats (over 40%) and low in sugars. Not all birds are able to easily or efficiently digest and absorb fats. Differences in digestive physiology, as well as the varied nutritional needs among residents,  long-distance, and short-distance migrants has a great influence on fruit choice.

In the special case of poison-ivy, the fats in the fruits are also of a waxy (high melting point) nature, and only a few bird species are able to process plant waxes. The eastern form of Yellow-rumped Warbler, found here, is called "Myrtle Warbler" because of its ability to eat the fruit of wax myrtle (Myrica sp.). Poison-ivy fruits are very similar in composition to myrtle fruit.

You may wonder if the compounds (urushiol) in poison-ivy that cause a skin rash in most people are present in the fruit. I'm not entirely sure; some sources say yes, others indicate they are only present in the sap, and therefore only the stems and leaves. In other plants with urushiol, such as mangoes, the fruit does not contain the poison. I'm not very sensitive to poison-ivy, and I have not had a problem handling the seeds once pooped from a bird.

Barred Owl in Dearborn

[2017 update at bottom of post]

I get a fair number of calls to help rescue birds from various situations, often inside buildings. Usually they are very common species (European Starlings, in particular, seem to have a knack for falling down chimneys). Often there isn't much I can do but offer advice, which is often sufficient to solve the problem.

Late yesterday afternoon I got a message from Jim Barber at the First Presbyterian Church of Dearborn, directly across the Rouge River from campus. They had a large owl (he thought Great Horned) trapped in the basement boiler room of the church. He was at wits end -- the bird had been there for at least several days and nobody he called was taking any action. The notion of a Great Horned Owl getting down a ventilation shaft seemed pretty unlikely to me, and he told me the room was around 20 by 20 feet with high ceilings. This seemed like a difficult situation, but my curiosity, the plight of the bird, and Jim's frustration compelled me to immediately head up to the church and figure out how I could help.

Within 15 minutes I was looking in the boiler room. Not only was it big, it was cluttered with all sorts of equipment and supplies. This was the part of the room I saw when I walked in:

To the right was the large boiler, and the ceiling went up another 10 feet, criss-crossed with pipes and supports. A large owl did indeed take off from a ceiling beam and fly over to a pipe above the boiler. To my great astonishment it was not a Great Horned Owl (or a Cooper's Hawk, which I sort of expected), but a Barred Owl (Strix varia), a species with only a single record from Dearborn, in 1976 (see history, below).

I decided there were two possibilities for catching the owl. One, with a long-handled net, although with so many obstacles I thought this might result in a lot of chasing around, which would stress the bird. The other was to use a bal-chatri trap, a common way of catching raptors. Bal-chatris are wire traps topped with small nooses made of fishing line. A mouse is placed in the trap, and when the raptor pounces on the trap it gets its talons caught in the nooses. The mouse isn't harmed, and the bird can be quickly released from the nooses. I figured the owl would be very hungry after three days, and this should work pretty well.

After outlining my plan, I called my husband Darrin O'Brien to have him bring a bal-chatri from home as well as a net for back up. I also called my good friend Jim Fowler. Jim has a great deal of experience handling raptors, and also spent 30 years at Greenfield Village as the head of the grounds department, rescuing plenty of birds from buildings. He would bring a long-handled net and a second bal-chatri. I went up to the RRBO banding lab on campus to fetch my banding gear.

When we all reconvened at the church, a member of the Michigan Humane Society wildlife department had shown up. He and the church employees had ambushed the owl and had it wrapped in a shirt, saving us a great deal of trouble. We took the owl from him, and after one loud screech, it settled down and stared up at Jim, who held it while we checked the bird's condition and I measured the wing.

Considering its ordeal, the owl was in great shape. It had some fat, and had no injuries or even broken feathers. Jim Barber and his co-worker Jamie told us there were certainly mice in the boiler room (and they also put out water for the owl), so we can only assume it probably ate while holed up.

Darrin took the bird so we could band it.

I let Jim Fowler do the honors. While it looks like Jim's nose might be in danger here, it's really the talons you have to worry about. Sure enough, it did grab Jim's thumb at one point, drawing blood.

Outside the breeding season, Barred Owls cannot be sexed, so we don't know if the owl is a male or female. But looking at the color, shape, and wear of the wing and tail feathers can help determine age.

All the wing feathers and their coverts (the row of feathers covering the bases of wing feathers) are quite uniform in color and wear, which indicates that this is a hatching-year (HY) bird. The outer six wing feathers (primaries) are slightly more worn that the inner four, but that might be due to flying around a room full of close objects and brushing against them. Further, at no age would we really expect all the inner primaries to be replaced and the outer primaries retained. The shape of the primary coverts is rounded, with the pale areas more round than block-shaped, also an indication of HY.

In the photo above, we can count the number of pale bars on primary number 9, the second feather from the outside. Five bars are visible, not counting the pale tip. In the first set of wing feathers on a Barred Owl, p9 will have 4 to 5 bars; there will be only 3 to 4 bars on it after it is molted. This feather might be molted in the owl's second fall, or not until the third fall. This owl was not actively molting any wing or tail feathers. If it was older than HY, it should have been molting some feathers, or there should have been some new wing feathers that looked different -- less worn with narrower bars spaced further apart.

The photo above not only proves I was actually present at this event, but also shows the pointed, whitish tips of the tail feathers. These are the first tail feathers, which will probably not be replaced until the bird's third fall molt, after which they will have darker, blunter tips and narrower bars. All of this leads to my conclusion that this is an HY bird. By October 10, we would expect that any flight feather molt would be nearly or fully complete, so if it were an older bird, we would see the contrast in old and new feathers discussed above. It's always tricky determining the age of a bird with which you have no experience, so I'm happy to take any corroboration or contradiction in the comments!

We walked to the edge of the parking lot to release the owl. It took flight and landed in a tree, shook itself, gave us one last look, and took off toward the Rouge River and UM-Dearborn campus. 

Barred Owls in Wayne County

According to historical accounts, Barred Owls nested in Wayne County until about 1910, after which they were apparently extirpated, being hunted and pushed out by agricultural and urban development. The next record I found was a report from the UM-Dearborn campus on 12 May 1971. I was unable to confirm the veracity of this record, but it was by a reliable observer. The only reliable report I have had from Dearborn is a bird seen on 7 April 1976 by Gary Hutman on land that is now the TPC Golf Course. Over a decade passed before a report came from Grosse Ile on 27 December 1987. In 1994, one was seen at Lake Erie Metropark by my husband Darrin on 17 September.

In recent years, Wayne County reports have become more frequent. Daryl Asprey found one at Crosswinds Marsh (phase 1) on 6 December 2003. When RRBO was coordinating the breeding bird atlas work for the county, one or two were reported calling in a wooded area in Canton Township from 2004-2006; we were never able to confirm nesting. Several years ago, a pair took up residence at Oakwoods Metropark, first heard calling by Daryl Asprey on 23 February 2007. Feel free to add any further records in the comments.

Thanks to everybody who participated in this rescue. Good luck to the owl, and to the guys who I understand will be on the roof of the church today, closing off a certain ventilation shaft...

Update: I had a report of a Barred Owl being heard at the far north end of campus on 23 April 2017.   

Fall banding 2011: Weeks 6 and 7

The month of September drew to a close on a windy, rainy note -- the 9th full day of banding missed this month due to weather. Notable mid-way through the season is the near total absence of sparrow migration, with only two White-throated Sparrows banded so far, and no noticeable movement of Swamp, Song, or Lincoln's Sparrows.

Thrush migration has also been dismal. Gray-cheeked Thrushes are never numerous, and Hermit Thrushes have yet to arrive, but Swainson's Thrushes (left) are one of our bread-and-butter birds. Our average for Swainson's Thrush is 70, and a number of years we have topped 150. Our lowest year we caught 24, and we are tied for that number right now. It sounds like other banding stations are also below average. South of us, Black Swamp Bird Observatory was at 131 Swainson's Thrushes as of mid-month, and their mean is 480. If you have any thoughts or observations on thrush migration in the eastern U.S., please feel free to leave a comment! Looks like we'll have to put in at least one more year to increase our sample sizes for our dietary study.

Speaking of which, many trees and shrubs are still laden with late-ripening fruit. I don't ever recall having any substantial crops of arrowood (Viburnum dentatum) or elderberry (Sambucus canadensis) into September, yet these trees have had fruit most of the month, some still coming ripe. The Amur honeysuckle (Lonicera maackii) crop, a non-native, is still only about half-ripe. On the final day of September, I found multiple plants of this species, as well as gray dogwood (Cornus racemosa), woodbine (Parthenocissus inserta), and Japanese honeysuckle (L. japonica) still blooming! This year we have large crops of wild grapes (Vitis spp.), poison ivy (Toxicodendron radicans), and non-native climbing nightshade (Solanum dulcamara). Does anybody else think fruit crops are a little tardy this year?

Good numbers of warblers, relatively speaking, are still moving through, although we have yet to see Yellow-rumped Warblers in any numbers. This past week we exceeded our high number of Nashville Warblers (above), with 60 banded so far (our fall mean is 28). Two more uncommon species were also banded this period.

 This Connecticut Warbler was banded on 27 September...

...and so was this female Blue-winged Warbler. The Blue-winged is actually much rarer here in fall than the Connecticut. This is only the fourth Blue-winged Warbler we've banded in fall, whereas we've banded over two dozen Connecticut Warblers.

A reader asked for an update on the condition of American Robins, since I mentioned in our week 1 summary that the young robins we were catching were quite emaciated. For the month of August, the average weight of young robins was a full 5 grams below the average for previous Augusts. For August and September combined, it was about 1 gram below average. However, our sample size for this fall is very low. The ratio of young robins to adults is very out of whack. Normally for August-September, 75% of the robins we band are young. This year, it's only 36%. This is really surprising, since productivity (as judged by the multitude of young robins we saw this summer) seemed very high. What happened to them? Was there high mortality among other underweight robins? Unfortunately, we will probably never know.

Let's hope October brings us better weather, and more robins, catbirds, and thrushes!