After my last blog on tern species in Chicago, I decided to take a deep dive into the underwing patterns of terns more generally. A number of resources point to the underwing pattern in terns as a valuable place to look for species identification. When I tried to look at the details of these patterns in photos, however, I found it to be a challenge to discern what exactly I was seeing. There are a number of variables that make these feathers hard to read. So I took a few visits to the Field Museum collection to gain some clarity. Below are illustrations comparing Arctic, Common and Forster’s Terns; Roseate and Sandwich Terns; and Gull-billed and Caspian Terns. In the Sandwich Tern section I touch on the hot topic of American vs. Eurasian Sandwich Terns. At the end there are some important notes about the illustrations as well as a discussion of “pale bloom.”
The overlapping primary feathers in this Arctic Tern photo demonstrate how difficult it can be to understand what the pattern is on each individual feather.
A note on “mature” vs. “juvenile”: as cited below, many tern species may not be definitively aged once they lose their juvenile feathers, including birds that appear to be “adults.” For this reason, “juvenile” here will refer to first generation primaries, and “mature” to any later generation. Those without a designation are species for which I did not see any obvious distinction between these two categories.
Sterna terns with hooked patterns in the primaries: Forster’s, Common and Arctic
These three species from the Sterna genus illustrate some of the basic differences seen in tern primary patterns. The outer primaries show dark along the feather shaft, and then towards the tip the dark may “hook back” along the inner edge of the feather. This hook pattern may be long, as in the P9 on Forster’s, or shorter, like the P9 on common and Arctic (see the Forster’s Tern illustration above for the primary numbering scheme). The hooks may be distinct, as in P9-P6 on Common Tern, or less distinct and more washed out, as in P8 on adult Forster’s Tern. Finally, hooks may be reduced or absent, as in P10 on Common and Arctic Tern. The hook patterns on each feather usually show a limited range of variation within each species. For example, most Common Terns show hooks that are distinct to P6 and get washy by P5, but some show distinct hooks on P5 and less frequently P4. Common and Arctic Terns tend to lack hooks on P10; however I found some with hooks on P10, and one juvenile Common that had a hook on its left wing’s P10 and no hook on the right wing’s P10.
The relative darkness of each species shown above I believe to be a real difference, not just due to wear; for example, even when worn, Arctic Tern primaries appear to have a paler pattern than those in common tern, with the hooks appearing slightly darker than the rest of the pattern. Arctic Terns also have subtly reduced dark areas at the tips of the primaries, measuring from the peak of the “pale tongue” to the feather tip, and usually have distinct hooks to P4, though in some the hooks get washed out by P5 and thus overlap with the normal pattern in Common Tern. The shape of each primary in Arctic Tern becomes narrower towards the tip, and the wings themselves have a different shape, neither of which is illustrated here. All of these differences combine to form the impression of a longer, narrower dark trailing edge on Arctic Tern wingtips noted in field guides.
Common Tern, Wisconsin, July 2023. This individual has distinct dark hooks to P4, a pattern more normally seen in Arctic Tern, and an unusually long hook on P8. Most Common Terns have distinct dark hooks to only P6 or P5, regardless of level of wear. If inner primaries are more recently replaced and fresh (usually P1-P5 in mature Common Tern), any dark pattern there, with or without a hook, will be less visible due to pale bloom. Note how the transparency of the feathers creates darker areas where feathers overlap.
One thing I was on the lookout for were any obvious differences that might associate with age. Accurately ageing terns after they lose their juvenile plumage may be mostly guesswork at present, including in birds with “adult” appearance (R.M. Ward 2002; White and Kehoe 2001). Some species that are known to show obvious differences in juvenile primary pattern are Forster’s Tern and Eurasian Sandwich Tern (shown below); another example I found is White-winged Tern (not shown here, but the juveniles I looked at showed obviously longer hooks on P10 and P9). Other than these species, I did not find any obvious differences, i.e. differences I could eyeball, in the juveniles I looked through at the Field Museum. I also did not find any clues to intermediate immature plumages in the primary patterns. Pyle’s Identification Guide notes some more subtle measurements in primary patterns that may help determine age with birds in hand, but I do not know if these have been verified with known-age birds.
With Forster’s Tern I took out a ruler and did some rough measurements to see what exactly could be said about the difference in juvenile primaries. In juvenile Forster’s, P10 is either lacking a hook or it is greatly reduced, P9-P8 usually have noticeably shorter hooks than adults, and hooks are usually distinct to P7 and get washed out by P6. Some Forster’s juveniles showed drastically shorter dark tips and hooks in their primaries, while others were closer in length to adults, as in the example shown here; I do not have any guesses as to what causes this variation, or whether it is linked to more “mature” hormone levels as has been suggested for similar cases in gulls. The juvenile Forster’s shows a pattern much more similar to Common and Arctic Tern, which may hint at some phylogenetic history (which pattern evolved first, etc.). As a general rule, juveniles of many tern species seem to have more intensely dark patterns not only on their primaries but elsewhere, either due to more dark pigment, less pale bloom, or likely both.
The Sandwich Tern question, with a note about Roseate Tern
As I was working on this project, David Sibley had an in-depth article published in ABA’s North American Birds that discussed the potential species divide of Sandwich Terns living on different sides of the Atlantic Ocean. This prompted me to compare the hook patterns in Sandwich Terns throughout their range. It’s important to note that the patterns in question are largely hidden in the folded wings of sitting birds, while they are fairly discernible in photographs of the underwing in flight. Also, don’t confuse the word “hook” here with the hooked shape of worn primary tips, which is a separate (but related) field mark discussed in Sibley’s article.
Mature and juvenile American Sandwich Terns, but possibly only juvenile Eurasian Sandwich Terns, show a pattern like that on the right: no hook on P10, with hooks starting on P9 and usually getting smaller and less distinct by P6. Most mature Eurasian Sandwich Terns show a pattern somewhere between the center and left illustrations. I did find a couple Sandwich Terns collected in the US and the Caribbean that showed a pattern like the center illustration, and one example from Florida shown below. Note that as patterns move towards the pale end of the spectrum, the hooks become reduced and then lost on P9 and P6 before P8 and P7. I tried to approximate the variation in the thickness of the pale fringing across types, which, as Sibley discussed, appears to be consistently different between American and Eurasian birds.
How much these primary patterns show a gradient vs. two distinct types is an open question as far as I can tell. My little research here suggests that it is difficult to assign birds showing the center pattern in the context of a possible vagrant without the use of other field marks. Even so, due to the clarity with which the hook patterns can be discerned in photographs of flying birds, I believe this to be one of the more useful field marks, especially in birds showing patterns at the extremes.
The Sandwich Tern above in Texas shows distinct hooks on P9-P7, typical of American birds, while P6 is less worn and the pattern is masked by pale bloom.
The Sandwich Tern above photographed in Florida shows a hook pattern that may be more common in Eurasian birds than in American birds, or at least the pattern lies somewhere in between the two extremes. Note the very short hooks on P9, and the lack of hooks by P7. Other field marks should be used to determine if a bird like this is a possible vagrant.
The individual above from Portugal shows no hooks, similar to the left pattern in the illustration. This pattern may be exclusive to Eurasian Sandwich Terns.
The left pattern is interestingly quite similar to the pattern shown by Roseate Terns of all ages. I’m not sure if this resemblance is of any relevance to the topic of a Sandwich Tern split, but it may hint at a more broad mechanism for pigmentation in tern primaries.
Roseate Terns, Massachusetts, July 2022. Mature Roseate Terns actually have quite a bit of dark pigment in their primaries on both inner and outer webs, but very little of it shows in a typical view. This is for a couple reasons. Mature birds usually have only three worn outer primaries during the spring and summer that reveal their pattern due to lack of bloom, while the fresher inner primaries presumably have their pattern concealed by bloom. The pale part of each primary also overlaps much of the dark part of the next more inner primary. In the first photo, the bird on the left has fairly worn P5 and P6, and is also spreading its feathers more, revealing the dark pattern more clearly than the other individuals. In the second photo, you can see the dark pattern showing through the pale part of each overlaying feather; this effects how transparent (or “translucent”) the wingtip appears at a distance.
Terns with less hooked patterns on the primaries
Finally, here are a couple species that do not show a clear “hook” pattern on most primaries. Caspian Tern seems to lack any clear hook pattern, but has reminiscent dark borders on the inner edges of most primaries. Gull-billed Tern is somewhere in between, somewhat similar to mature Forster’s, although the “hook” on P10 is already quite long. It may be that these are examples of what can happen as dark pigmentation increases (or has been lost to a lesser degree?), the other end of the spectrum from Roseate and paler Eurasian Sandwich Tern.
About the illustrations
The underwing pattern illustrations here are intentionally different from real life in a number of ways that deserve mention. The feathers are artificially spread and shown without transparency, so that each feather’s pattern may be seen clearly. One thing I learned by holding these species in hand is that the patterns viewed from the underside can be masked by “pale bloom” just as the uppersides of tern feathers are, though perhaps to a lesser degree. The drawings here show each pattern as if the feather’s bloom has mostly worn off, revealing the underlying dark pattern completely, to an equal degree on each feather, to the best of my estimation. In reality, feathers that are worn enough to show their dark pigment fully may be frayed at the edges and have reduced or absent pale fringes. This is an important consideration: the amount of pale fringing and the clarity of the dark pigment pattern are at odds with each other. This seems to be more of an issue with the middle primaries, where pigment patterns appear to be more masked than outer primaries even with similar levels of wear. In some species, the inner primaries rarely if ever get very worn, as they are frequently molted, creating the “molt contrast” with darker outer primaries seen in many tern species. In these cases, I looked at juvenile plumage, which is either lacking in pale bloom(?), has increased dark pigment, or both, and these informed my estimations for the “adult” pattern. However, as shown, in some species the juveniles have obviously different patterns from adults, so in these cases the inner primaries are just estimations. The pale fringes shown here are rough approximations of what they look like in mostly fresh feathers. Finally, each species was drawn over the same template, so any feather and wing shape differences are not illustrated.
What is “pale bloom”?
For those unfamiliar with this concept, and I’m curious the degree to which anyone is very familiar with it, “pale bloom” is one of a few terms used to describe the silvery-pale appearance of some tern feathers when they are newly grown in mature birds (I do not know to if this occurs in juvenile plumage). It’s important to note that the dark pigment patterns are understood to be present at the time the feather is grown; the patterns are not somehow created by the wear process, the shape of overlying feathers, etc., but rather revealed as the bloom wears off. The only reference I could find to anyone claiming to observe the cause of the “pale bloom” was from Dwight in 1901, where he wrote:
The 'frosting' is due to elongated, curved and frilled barbules on the distal sides of the barbs, and when the barbules are worn off their black basal portion becomes conspicuous.
Here Dwight is referring broadly to “wings” in many species of tern. I’m curious if anyone since 1901 has looked into this, including which feathers have this bloom (for example, do tail feathers? Which wing coverts? Do any body feathers?), and to what degree it might vary with species, age, sex, etc. As noted by Pyle and others, the secondaries and tertials become darker with with wear, and the darkness of these feather groups is an important field mark for identification of Sandwich Tern types and other species. I also wonder if “pale bloom” is unique to terns, or if there are similar traits in other bird groups. It seems to me to be an understudied topic, considering how drastically it alters the appearance of these birds. If you have more info, please let me know.
Citations
Dwight Jr, Jonathan. "The sequence of moults and plumages of the Laridae (gulls and terns)." The Auk (1901): 49-63.
Pyle, Peter. Identification guide to North American birds: a compendium of information on identifying, ageing, and sexing" near-passerines" and passerines in the hand. Slate Creek Press, 1997.
Sibley, David. "‘European’ Sandwich Tern in North America." North American Birds 74.1 (2023): 16-25.
Ward, R. M. "Ageing and moult in Common Terns." British Birds 95.6 (2002): 314-316.
White, S. J., and C. V. Kehoe. "Difficulties in determining the age of Common Terns in the field." British Birds 94.6 (2001): 268-277.
Special thanks to Ben Marks and John Bates for granting me access to the Field Museum collection, and Judy Pollock for help in editing.