Most ruby-throated hummingbirds begin migrating south in late July and August, with adult males leaving first and the last stragglers clearing the Gulf Coast by early November. Peak southbound movement through the eastern United States falls between mid-August and mid-September, driven by declining day length rather than cold or food scarcity.
That date range holds for the eastern ruby-throated population, which makes up the bulk of what most U.S. and Canadian observers see. Western species track different calendars, and the reasons behind any individual bird's departure are more interesting than the calendar itself.
The proximate trigger: photoperiod, not temperature
Hummingbirds do not leave because nights get cold. They leave because the ratio of daylight to darkness crosses a physiological threshold, and that threshold is set by an internal clock rather than by the thermometer.
As summer wanes, the shrinking photoperiod reaches the birds' pineal gland and hypothalamus, shifting circadian and circannual rhythms. That shift drives two measurable changes. It increases migratory restlessness, a captive-observable behavior called zugunruhe in which birds orient and beat their wings at night. It also reorganizes fat metabolism, so the liver and adipose tissue deposit fuel reserves the bird will burn in sustained flight.
A hummingbird departing the eastern United States must carry roughly 2 to 2.5 grams of fat on a body that weighs about 3 to 4 grams. That is a fuel load approaching its own lean body mass. The bird cannot grow that reserve quickly; depositing it takes days of hyperphagia, during which intake may climb well above baseline maintenance. This is why departure is not a sudden decision. The bird has been preparing for weeks, and photoperiod set the schedule before the first feeder went quiet.
Proximate cues and ultimate causes deserve separation. Temperature, storm fronts, and local food collapse can delay or accelerate a departure already scheduled. They rarely cause the schedule itself.
Quantified timing: who leaves when
A widely visible pattern holds across the eastern breeding range. Adult males are first. They finished breeding duties early and have the least reason to linger. Adult females follow after their last brood. Hatch-year birds, especially late-nesting northern young, depart last, and they leave without any prior experience of the route.
For ruby-throated hummingbirds, the sequence generally looks like this:
Adult males: southbound movement begins late July and is largely complete by late August. Adult females: main departure mid-August to early September. Hatch-year birds: main passage late August through mid-September, with lingering individuals into October. Gulf Coast passage: most crossing of the Gulf of Mexico occurs late August to mid-October. Late stragglers: a few birds remain at feeders into November in the South, but these are outliers rather than the norm.
The southbound window fits a coherent biological logic. Leaving in late summer lets birds arrive at overwintering grounds in Central America and southern Mexico while flowers, insects, and nectar corridors are still productive. Staying until leaf-fall would mean crossing food deserts with depleted reserves. Departure timing therefore maximizes the chance of reaching a feeding territory before the bird must compete for one.
The Gulf crossing and the fuel-loading arithmetic
The Gulf of Mexico is the defining constraint on eastern migration. A ruby-throated hummingbird leaving the U.S. Gulf Coast must cross roughly 500 to 600 miles of open water, a nonstop flight of 18 to 22 hours in still air. There is no rest, no refueling, and no margin for error.
That flight is why the pre-departure fat load matters so much. A bird with a full reserve can survive the crossing; a bird a few tenths of a gram short may not. A counterintuitive detail: many hummingbirds do not fly straight across. They take advantage of tailwinds and weather windows, and some eastern birds move along the coast and cross at the narrowest points. Others island-hop through the Caribbean.
For western populations the geography changes the problem. Species migrating through the western interior and along the Pacific slope face mountain barriers and more complex weather, but not the same single open-water barrier. Their routes trace descents into Mexico and Central America along corridors where nectar resources are more continuous.
Edge cases and the "it depends" reality
Species matter at least as much as location. The rufous hummingbird migrates earlier and travels farther than most, with some adults beginning southbound movement in July from as far north as Alaska and British Columbia. Rufous migration is famously early and famously long-distance. Anna's hummingbird, in contrast, is largely nonmigratory on the Pacific coast. Where feeder provisioning is reliable and winters are mild, many Annas stay put. This proves migration is a population-level tendency, not an absolute species rule.
Weather is a second edge case. A strong cold front or sustained north winds can concentrate or delay southbound movement. A warm autumn can keep nectar and insects available later, and feeders stay occupied longer. But weather modulates timing; it rarely drives it. Reports of hummingbirds remaining into December are almost always isolated birds at well-maintained feeders in mild coastal climates, not a population shift.
Age and experience matter too. Hatch-year birds must find the route on their own. Their departure timing is more variable, and their first migration is where many die. This is a limit on what any calendar can tell you. It describes the population, not the individual bird at your feeder.
Misconceptions worth clearing up
One common belief holds that feeders cause birds to delay migration or make them dependent. Evidence does not support the idea that a feeder prevents departure. The trigger is photoperiod, and a bird that is physiologically ready to leave will leave even with a full feeder nearby. Long-persisting individuals are birds that have not yet entered the departure window, not birds trapped by handouts.
A second misconception treats all hummingbirds as one migrating species. Region and species change everything. A feeder in Arizona in September supports a different mix, on a different calendar, than one in Ohio.
A third sees autumn migration as a single synchronized wave. It is not. It is a staggered exodus by sex and age class over three months.
Red flag: a hummingbird that remains through a hard freeze with no insects or flowers is likely in trouble and not simply "staying late." Red flag: treating a late-November bird as evidence the whole species is shifting southbound timing. Red flag: assuming one tagged or attested individual's behavior represents the population.
Migration does not happen overnight, and it does not happen all at once. A single bird's schedule depends on when it hatched, when it last nested, and what its body was able to store before departure.
What to do with this information
If you manage feeders and want to support southbound birds without misreading them:
Keep feeders filled through late October in most of the eastern United States, and through mid-November in the South. Fresh food costs more maintenance but removes a variable. Track local first and last observation dates across years. Farms and birding groups maintain useful records. Watch for the earliest departures by sex: males going quiet in late July is normal, not a sign of a problem. Resist the urge to "help" stragglers by keeping food out indefinitely. A feeder on a hard freeze does not replace the insects a hummingbird needs, and a bird that has not migrated is not a rescue candidate.
A cleaner heuristic: match feeder availability to the local migration window, not to your wish that a particular bird stay.
What this explainer cannot tell you
Migration timing is genuinely variable across species, regions, and individual years, and the data come overwhelmingly from tagging studies, banding recoveries, and observer records rather than from continuous monitoring of every bird. Averages hide the variance. What is well established is the sequence and the mechanism. Male-first departure, the photoperiod trigger, and the Gulf crossing as a hard constraint are all supported. The exact dates for a particular county in a particular year are not something this article can promise.
Treatment of a sick or stranded hummingbird is also outside what general information can address. If you find a bird in distress, contact a licensed wildlife rehabilitator or your state wildlife agency.
This article is general educational information about hummingbird migration biology, not a diagnosis or care plan, and it does not promise outcomes for any individual bird. For species-specific or wildlife-rescue decisions, consult a qualified local wildlife professional or licensed rehabilitator.
Comparison table: southbound timing by population
| Population | Approximate southbound start | Peak passage | Typical overwintering region |
|---|---|---|---|
| Ruby-throated, eastern adult males | Late July | Late July to late August | Southern Mexico and Central America |
| Ruby-throated, eastern adult females | Mid to late August | Late August to mid-September | Same, with broader timing |
| Ruby-throated, hatch-year birds | Late August to September | September | Same, timing more variable |
| Rufous, western | July from far-north breeding grounds | July to September | Mexico, with route complexity |
| Anna's, Pacific coast | Often none | Minimal movement | Resident in many coastal areas |
| Other western species | August to September | August to October | Mexico and Central America |
Timeline table: a typical eastern migration cycle
| Period | Event |
|---|---|
| Late July | Adult males enter zugunruhe and begin southbound departure |
| August | Females follow after their last broods |
| Late August | Hatch-year birds and late-breeding adults enter passage |
| September | Peak southbound movement through the eastern U.S. |
| September to October | Main Gulf crossing window |
| October | Late passage, primarily hatch-year birds and southern breeders |
| November | Last stragglers descend to Gulf Coast; few birds remain in South |
| December onward | Rare nonmigratory stragglers persist in mild coastal areas |