The Monarch invasion of Great Britain, 1995


Last autumn a large number of Monarch butterflies arrived in western Europe and were reported from Great Britain, France, Belgium and Holland. Simon Coombes and Vic Tucker (with contributions from Roger Lane and Bart Vanholder) write here about this event and discuss the origin of these magnificent butterflies.

Butterfly Conservation members hoping to see Monarchs this year should keep in touch with Butterfly Line 0891 884505.

The Monarch or Milkweed, Danaus plexippus, has long been suspected to cross the Atlantic Ocean unassisted from America to Europe, mainly during September and October, but perhaps only once or twice a decade.

This article deals with four theories normally encountered in explanation: unassisted transatlantic crossing; entirely ship assisted passage; derivation via the Canaries and/or Iberian Peninsula; released or escaped captive-bred insects. We hope to show that the initially most improbable is actually the most likely. The mass sightings of 1995 raised the debate again and provided some further evidence.


This charismatic yet enigmatic butterfly is a native North American (and elsewhere) and annually it spectacularly traverses a north/south migratory route, from Florida and Mexico to Canada and the USA in spring and summer, returning southwards as another generation in autumn. They overwinter in awe-inspiring millions per roost site, often in fir trees in warm or even cold, dry air (e.g. the Mexican highlands).

An English immigrant Monarch.

Historical perspective

Over many years it has been popularly assumed that Monarchs in Europe arrive from America, usually assisted by powerful westerly winds. However, from 1876, when first sighted in the UK, until 1930s it was thought that all such passages were ‘ship-assisted’, but the recording of almost 100 individuals between 1928 and 1938 made their arrival by ship "no longer plausible as a main explanation" (Emmet & Heath, 1989). Thus an argument for a ‘natural’ arrival was first advanced 50 years ago.

Most notable arrivals to Britain in more recent years, were the influxes of 1968 with about 65 individuals recorded, and in 1981 involving some135 butterflies. However, the exciting 1995 invasion involved at least 170 occurrences, most of which were witnessed in Britain’s south coast counties.

Absence of UK sightings prior to 1876 is not suprising, given a far smaller populace bearing fewer field naturalists who needed to be present at arrival times. These highly unfavourable odds, combined with possibly extended periods of blank years, severely limited sighting opportunities.

Ship assistance

An assisted passage via sailing ships of this general period would be unlikely. Most ships took at least 20 days to cross the Atlantic and, with the severe turbulence caused by stormy conditions, this would prove a difficult journey for Monarchs to survive. The rumour of a ship once carrying over 200 Monarchs is discounted and it is likely anyway, that for thousands of years many invasions have occurred into the UK during conducive atmospheric conditions.

Another consideration is the ships’ destinations. Almost no Atlantic vessels dock in the far south-west ports of UK. Most are bound for major British and European ports like Rotterdam. Since far more Monarchs are seen in south west Britain than in major international European ports, reliance on total ship assistance must be discounted as a viable theory.

Nowadays many birdwatchers go to South Coast headlands searching for migrant birds - the most likely places for Monarchs to make their initial landfall. Many birdwatchers are also knowledgeable lepidopterists and as a result over 45 Monarchs were recorded on Sunday 8 October 1995. Many sightings were made simultaneously throughout coastal southern England.

Despite original disbelief in the Monarch’s ability to fly so far ‘unaided’, the theory for an unassisted transatlantic crossing was firmly strengthened by two well defined ‘migrations’ in 1968 and 1981 when these butterflies arrived together with many American landbirds. It is now an accepted fact that small American birds, such as warblers cross the Atlantic ‘under their own steam’ using prevailing winds. In fact the parallels between the 1968 migration and last year’s are remarkable: very similar bird species arrived (e.g. Blackpoll Warbler, Dendroica striata), weather conditions were near identical and even the arrival dates match perfectly.

Breeder release

The accidental escape of around 30 Monarchs in July 1995 gave rise to a single sighting nearby, extrapolation of these numbers would require thousands of escapees to produce 170 sightings over such a wide area. None of those few breeders of this species lost insects at this time and losses on such a scale are unknown. Deliberate release in Britain in the absence of the right foodplants or conditions is pointless and would constitute a significant financial loss.

Weather conditions over the western Atlantic at the time of the 1995 immigration led to actual prediction of their arrival, e.g. Nick Bowles (Butterfly Line, 2 October) the day before the first two arrivals in Britain occurred (one in Cornwall, one in Dorset), although these had been preceded by one possible 'natural migrant’, in Liverpool.

The verdict that releases account for the presence of all the 1995 butterflies in north west Europe must be without foundation.

Weather in autumn 1995

In America tens of millions of Monarchs undertake two massive migrations of over 2,000 miles each calendar from Mexico and Canada and back. Often they overfly very large tracts of extremely inhospitable terrain, while also encountering relatively low temperatures. These and other severe obstacles must be overcome if they are to reach their intended destinations. Their routes do not cross flowery meadows all the way.

In 1995, a series of three hurricanes, up to the last week of September, off the eastern seaboard of the United States resulted in strong westerly airflow extending to north west Europe, including south west Britain, culminating in a west south west gale on 30 September.

Extensive research incorporating data from the Internet found Monarch numbers attained a peak, a record high, on 28 September on the American East coast - at Cape May, New Jersey, for example, famous as a staging post for thousands of migrant birds. In the wake of the last hurricane (Marilyn, tracking offshore 21 - 26 September) an unusually strong, cloudy, non-turbulent air-flow ('zonal flow') across the Atlantic was established between 28 September and 1 October. It was relatively warm, and travelled at 30-35 knots. This indicates a four day (100 hour) travel time for which the Monarchs’ normal Autumn fat reserve could easily sustain energy levels.

To assist butterflies to continue flying at night the initial effects of coastal, uplifting, (vortexing) thermal air mass stratification was replaced by the relatively low altitude (600m), warm (10-15 0C), quick zonal flow air stream when physical effort by the butterflies would have been minimal: they would have literally just gone with the flow. Monarch’s regularly migrate at a height of 1,000 metres in America and have been seen at 4,000 metres by pilots flying light aircraft.

Daily weather notes taken at this time, to assist a Red Admiral study (RL pers. comm.), including wind force and direction, suggesting many Monarchs might have arrived under cloud-cover on south west winds during 2-5 October, their arrival having been immediately preceded by these strong Atlantic lows. This further strengthened beliefs long held by many that Monarchs utilise westerly airflow (in 1995 between 28 September and 1 October) to traverse the Atlantic. We feel this is the only plausible explanation for their welcome arrival upon our shores.

Possibilities of a jet-stream vector were eliminated as they do not fit in with arrival times and have a mean temperature of minus 41 0C. However, the jet stream (a fast moving high-altitude air current) pulls low pressure systems across the Atlantic from America, therefore its influence cannot be completely discounted.

These south west winds from 2-5 October deposited the first few butterflies upon England’s coastline, but we believe mostly unseen due to inclement weather. Those first sightings between 3 - 5 October, it is thought, numbered less than ten, mainly Cornwall, Devon and one in Dorset. South west winds may have also served to disperse them all along coastal southern England over this period. Total cloud cover on 6 and 7 October prevented many further sightings, but a change of wind direction to southerly on 8 October, a Sunday with 100% sunshine, resulted in many sightings by people outdoors, including large numbers of birders on south coast headlands.

Map 1

The following week saw southern England bathed in over 50% sunshine with daytime temperatures reaching 20 0C (up to 5 0C above the seasonal average for mid-October), due mainly to the continuous warm south and south easterly winds from the Mediterranean. Many Monarchs remained faithful to their favoured nectaring sites during this period. These included the Lands End valleys and the Lizard in Cornwall, Prawle Point and Slapton Ley in South Devon, and Portland Bill and Lodmoor in Dorset. Although described as restless and flying strongly, several remained at least until the 14 October. Cloud, and particularly, coastal fog set in on the afternoon of 14 October. Many Monarch seekers were successful during the morning (e.g. Lizard Point), but were unsuccessful after midday, with no afternoon sightings there on 14 October.

Map 2

Only 24 hours later did cloud and fog clear, but despite strong sunshine all Sunday afternoon (15 October), Monarchs had disappeared, at least from Cornwall. Probably significant, is that this 'dispersal' coincided with a wind change from south east to south west over the night of 14/15 October and this may have assisted a move eastwards,as insects were now reported from Dorset, Hampshire, Sussex and further north (Walney, Lancs, 20 October).

A 75% reduction in sightings occurred abruptly between 14 and15 October. This begs the question whether or not most attempted to resume southward migration (as they would do in America). Did they cross the Channel to France and if so did they simply perish in the damper, cooler European climate? Unfortunately, we have no answer to their sudden disappearance after the 14 October. We await further evidence or theories (attempted hibernation). It does appear they remained at their favoured nectaring sites (including gardens) along the south coast, only as long as the south and south easterly winds persisted. This pattern of sudden disappearance is apparently typical of previous invasions, especially 1968 and 1981 when "very few were seen after about 10 days" (Emmet & Heath, 1989).

Map 3

In mainland Europe Monarchs were seen in France in Brittany and as far south as Les Sables d’Olonne on the Atlantic coast of the Vendee between 9 and 29 October; in Belgium on 15 October and in Rotterdam and Haringvliet, in The Netherlands (where the butterfly has previously been rarely recorded), on 14 October and 16 November resepectively. The last record was of a butterfly flying over water in very wet and windy conditions when the temperature was only 12C (C. van Swaay).

Iberian and Atlantic island origins

Other than a transatlantic crossing or release/escape of captive bred specimens, the only other source of origin of European Monarchs would appear to be either the Iberian peninsula or Atlantic islands such as the Canaries where the species is resident. Such a migration would involve a north east movement of 1,000 miles or more, the precise route, possibly including negotiating mountain ranges (e.g. the Picos of north west Spain), only to be guessed at.

This theory cannot be entirely eliminated, owing to lack of unassailable evidence for a North American origin. With assistance from prevailing south and south easterly winds from 5-14 October 1995 it seems possible. However, considering their accurately predicted arrival from America it appears this must be the source, as enlightened experience was employed in this forecast. Since about 170 arrived safely and in immaculate condition, maybe several thousands would have originally been heading towards Europe and it is only to be expected that the majority would have perished en route. Displacement on this scale from the Iberian Peninsula or the Canaries would so seriously deplete these small populations as to virtually annihilate them (although admittedly a journey largely overland from Spain could result in fewer losses) and there has been no evidence that this is the case.

Support for an entirely American source is now overwhelming, thanks largely to Belgian entomologist Bart Vanholder's extensive research, and it is proposed that the vast majority if not all European Monarchs (other than escapes) originate from America. The eastern North American race of these butterflies is also strongly migratory over long distances and present in millions whereas the small Iberian and Atlantic island populations are virtually sedentary.

All factors combined indicate that far from being an ‘impossible’ oceanic crossing’ their chances of safely reaching our shores are relatively high, involving an easier journey than that which they are expected to encounter on their normal internal American migrations.

Until chemical and/or genetic analysis of north west European-caught arrivals, by cardenolide fingerprinting for example, has been undertaken the theory for an origin from ‘Iberian/Atlantic Island regions’ cannot be disproved, but is weak upon close examination. It also has to be remembered that the Canary Islands and the Iberian Peninsula were colonised by the butterfly (presumably from eastern North America) only relatively recently.


An extensive early draft of part of this article was written by Roger Lane, prior to poor health. Much invaluable contemporary evidence and research was collated via the Internet by Bart Vanholder (Belgian migration). Simon Coombes produced graphs, maps and co-authored the paper with Vic Tucker.

Grateful thanks go to the County Recorder of Cornwall (Frank Smith); CBRU (Adrian Spalding); Devon (Tom Sleep); Dorset (Bill Shreeves) and Hants (John Taverner), and to all those individuals whose vital observations are the raison d’être. It is also a pleasure to thank Nick Bowles for volunteering extensive constructive criticism and suggestions much improving an earlier draft of this article, and Mike Campbell senior meteorologist with the atmospheric environment branch of Environment Canada for a large part of the weather information.



Selected references

Burton J.F. & French R.A. (1969). Monarch butterflies coinciding with American Passerines in 1968 British Birds 62: 493-6

Emmet, A.M. & Heath. J. (1989). Moths and Butterflies of Great Britain and Ireland 7(1) Butterflies. Harley Books, Colchester.

Elkins, N. (1983). Weather and Bird behaviour. T&AD Poyser, Calton.


Interet 1995. Various contributions from a number of people, via the Leps-L listserver.



© All pictures in these pages copyright to Simon Coombes. Permission must be sought and obtained for any use.