UK Butterflies

Building a Community of Responsible Butterfly Enthusiasts in Britain & Ireland

Large Blue

Phengaris arion

fen-GAR-iss a-RY-on

Wingspan
♂ 38 - 48 mm, ♀ 42 - 52 mm

Checklist Number
61.013

Large Blue

Phengaris arion

fen-GAR-iss a-RY-on

Wingspan
♂ 38 - 48 mm, ♀ 42 - 52 mm

Checklist Number
61.013


This butterfly was first recorded as a British species in 1795 and, even then, was considered a rare insect. Due to the loss of suitable habitat, the endemic subspecies of Large Blue became extinct in the British Isles in 1979, the last site being on Dartmoor in Devon.

This magnificent insect has since been "brought back from the dead" through the dedication of several conservation organisations and many individuals. After its extinction in the British Isles in 1979, the Large Blue became the subject of a highly-organised reintroduction programme, using stock from Sweden. The estimated number of adults flying in 2006 was 10,000 on 11 sites, which is the largest number seen in the British Isles for over 60 years. This is a magnificent example of conservation in action.

The successful reintroduction of the Large Blue is made even more remarkable when one considers its elaborate lifecycle. The larva is parasitic in that it feeds on the grubs of a red ant, Myrmica sabuleti, on whom its existence depends. Although the dependence on ants had been known for many years, the dependence on a single species of ant, in order to maintain a viable population, was unknown to conservationists for many years until Jeremy Thomas discovered the association in the late 1970s. Unfortunately, the discovery came too late to save the native population. Today's reintroduction efforts focus as much on the population of ants present, as they do on the Large Blue itself.

Anyone wanting to see this species in the British Isles should visit the open access site at Collard Hill in Somerset. A "Large Blue Hotline" is usually set up each year that provides an up-to-date status of the emergence at this site. Details are available on the Butterfly Conservation website. In addition, Butterfly Conservation members and Somerset Wildlife Trust members have the opportunity to visit a private site, Green Down, each year, although places are limited. The majority of reintroduction sites are in the south-west of England, notable colonies being in the Polden Hills in Somerset, Dartmoor and Gloucestershire.

This species was first defined in Linnaeus (1758) as shown here (type locality: Europe).

The reintroduced British population is represented by the nominate subspecies.

This subspecies was first defined in Fruhstorfer (1915) as shown here (type locality: Cornwall, England).

This subspecies was endemic to the British Isles and is now, unfortunately, extinct. Two of the larger populations were found in the Cotswolds and the north Cornwall coast, extending into Devon at its eastern edge.

Ford (1945) describes the differences between several populations: "On the Cotswolds it appears in June or even in late May, and it is of a rather dark iron-blue shade. The species emerges a little later in Cornwall, where the specimens are of a clearer and brighter blue, while the spots on the upper-side are rather larger, though they may be less numerous that in the Gloucestershire examples. The extinct race from Barnwell Wold, Northants ... more nearly resembled the Cotswold insects. Ancient specimens from the Langport district of Somerset, where the species is also presumably extinct, are similarly of the darker blue shade, but they are remarkable for the large size of their spots, either on the upper or the under-side of their wings, or both. Those from the Salcombe district of South Devon on the whole resemble the Cornish form, though some of the specimens are of rather a duller tint ... It is clear that in England there is a tendency for the Large Blue to be of a brighter colour in the peninsula of Devon and Cornwall than elsewhere, though the discontinuity is not a large one. Somerset specimens tend to bridge the gap between the south-western habitats of this butterfly and those of the Cotswolds".

Phengaris arion ssp. eutyphron (Fruhstorfer, 1915)

Original (German)

♂♂ in der Regel relativ klein, bleich, Vorderflügel meist nur mit vier unbedeutenden Punktflecken. Weiblich nur wenig kräftiger schwarz umrahmt als der männlich. Patria: England, Cornwall, eine Serie in Kollektion Ch. Blachier. Nach Oberthür ist damit die Rasse der Bretagne identisch.

Translation

Males as a rule relatively small and pale, the forewing with at most four weak spots. The female only slightly darker-bordered than the male. (Type?) locality: Cornwall, England, a series in Ch. Blachier's collection. According to Oberthür the race from Brittany is identical with it.

Description to be completed.

The distribution data (2000-2009) has been made available through the generosity of Butterfly Conservation and any subspecies distribution is taken from Riley (2007). Click here to see the distribution of this species together with site information overlaid.

The Large Blue requires fairly closely-grazed grassland, where Wild Thyme is found in abundance, and where the host ant is able to flourish in good numbers. Overgrown conditions result in a cooler ground temperature, and a smaller population of ants as a result. Typical sites are on a south-facing slope that receives the full benefit of the sun.

It is generally agreed that the demise of the native population of this species was not the result of over-enthusiastic collectors, but large-scale habitat loss. The presence of this butterfly in the British Isles today is the result of a coordinated and ongoing conservation effort.

The table below shows the occurrence (distribution) and abundance (population) trends, using information from The State of the UK's Butterflies 2015 (Fox, 2015). Any UK BAP status is taken from the UK Biodiversity Action Plan (UK BAP) (2007 review).

UK BAP Status
Occurrence Change
1976-2014 (%)
Abundance Change
1976-2014 (%)
Occurrence Change
2005-2014 (%)
Abundance Change
2005-2014 (%)
Priority Species
Click here to access the Biodiversity Action Plan (BAP).
Insufficient DataLarge Increase +1440Insufficient DataDecrease -20

This butterfly has one generation each year and a relatively short flight period. Depending on the year, the adult butterfly can be seen from late May to early July, with a peak in mid-June.

Phengaris arion ssp. arion

This is a warmth and sun-loving butterfly. In bright sunlight the adults rarely bask with their wings open, and this is one of the few butterflies where photographers welcome intermittent sunshine or overcast conditions, when the adults will bask with their wings held open, revealing the characteristic pattern on the forewings.

After emerging, females typically fly to the bottom of the slope, where they are intercepted by males in search of a mate. The couple mate without any discernable courtship and remain together for an hour or so, after which the female rests and takes nectar. After another hour or so, the female will commence her search for plants on which to lay. Females are often seen probing the unopened flower heads of Wild Thyme with their abdomen, only to find that no egg has been laid, presumably because the flower head is deemed unsuitable. However, if a suitable plant is found, then the female typically lays a single egg, although 2 or 3 eggs may be found on the same flower head on occasion. This is presumably from different females since the larvae are cannibalistic while in the first instar.

Adults feed primarily on Selfheal (Prunella vulgaris), Wild Marjoram (Origanum vulgare) and Wild Thyme (Thymus polytrichus).

The white eggs are laid singly in the flower heads of Wild Thyme. Eggs are laid only in those flower heads that have not yet fully-opened, and are still relatively compact. The eggs hatch in 5 to 10 days.

"From July 5th to 17th inclusive, 1902, numerous females were observed depositing their eggs on the blossoms of wild thyme in North Cornwall, four being seen doing so on the last of the dates (17th) on the summit of a hill. The thyme was here growing in patches among the short turf and on ant-hills, nests of the common yellow ant (Donisthorpea flava). This turf was composed of the usual small plants which clothe the surface of the Cornish downs, a few furze bushes being dotted about. It was noticed that the plants selected by the females on the occasion in question were those growing in the open at some distance from the furze, and consequently fully exposed to all conditions of weather. It was also noticed that ants' nests were present under every patch of thyme visited by the butterflies. Several females were seen depositing on the thyme growing on ant-hills, also upon the thyme covering the turf walls where ants likewise abound. The species also very frequently deposits its eggs on thyme growing in more sheltered positions under the protection of furze, heather or other densely growing plants, both on the southern slopes of the hills and also in the valleys. In such situations the thyme frequently extends its sprigs through stunted and flattened furze clumps, so that the flower-heads are interspersed among those of the furze; on these blossoms of the thyme anion very commonly lays its eggs; usually only one egg is to be found on a single head of bloom, but not infrequently two, and sometimes three, eggs may be so found, these eggs having in all probability been deposited by different females. When laid the egg is generally inserted among, and fixed to, the calyx of one of the buds. It is very small in proportion to the butterfly, measuring only 0.50 mm. wide and 0.27 mm. high. The form is that of a very compressed spheroid, with the crown and the micropyle both sunken, the latter so much so that it appears as a dark spot to the naked eye. The entire surface is covered with white raised reticulations which form an irregular network, usually with a rhomboidal pattern, and are most strongly developed over the side, where they present a deep cellular structure. The ground colour is a delicate greenish-blue, but the whiteness of the reticulations gives the egg a whitish-blue appearance. Egg-laying extends over a period of four or five weeks, i.e., during the last week of June and throughout the greater part of July. The period of incubation of the egg varies from seven to ten days, according to temperature, the normal duration being about eight days." - Frohawk (1924)

The newly-emerged larva leaves the egg without eating the shell, and proceeds to feed on the flowers of Wild Thyme, which is its sole food source for the first 3 instars. The first 2 instars feed inside the flower head, while the 3rd instar is more exposed and may even move to new flower heads. However, with a change into the 4th instar, a distinct change in behaviour occurs. The larva will drop to the ground in the hope of being found by a red ant. An ant taps the larva once located, causing the larva to secrete a droplet from a special gland, called a Newcomer's gland, located on its 7th segment. This process is sometimes referred to "milking" the larva.

Eventually, after a period a 30 minutes to 4 hours, the larva distorts its body, by rearing up on its prolegs, to give the illusion of being an ant grub. On realising this, the ant immediately picks the larva up in its jaws, and carries it back to its nest where it lives alongside the ant grubs which, unknown to the ants themselves, form its future diet. The larva overwinters while in the final (4th) instar.

"During the first three stages the larva readily devour each other, and so strongly developed are cannibal habits that even the smaller larva will seize and feed on the larger ones. Immediately after the completion of the third and last moult the cannibalism disappears." - Frohawk (1924)

The primary larval foodplant is Wild Thyme (Thymus polytrichus). Wild Marjoram (Origanum vulgare) is also used.

1st Instar

"On emergence the larva is exceedingly small, only measuring 0.79 mm. long, but it is rather stout in proportion. The segmental divisions are deeply incised, and there is a longitudinal dorsal furrow. On the dorsal aspect of the first segment there is a large dusky-coloured disc, and there is also a similar but smaller disc on the anal segment. The colour of the body is pale ochreous-yellow tinged with greenish. On the dorsal surface are longitudinal rows of glassy white serrated hairs, placed in pairs on each side of each segment above the spiracles; all these dorsal hairs curve backwards, the anterior hair on each segment being decidedly the longer, and all the dorsal hairs have pedestal-like bases of an olive colour. All the hairs in the sub-dorsal rows are short, the anterior one of each pair curving forwards, the posterior one backwards. In addition to the rows of hairs already mentioned, there are below the spiracle (which is black) on each segment three brownish-coloured serrated hairs arranged in a triangle; all project laterally and have dark bases. Below these on the first lateral lobe of each segment is a single simple white hair, and two other similar hairs are present on the base of each clasper. The head is of a shining olive-black. The whole surface of the body is densely sprinkled with blackish points, giving it a rough appearance and enhancing the depth of the segmental divisions. The legs and claspers are similar in colour to the body." - Frohawk (1924)

2nd Instar

"After the first moult the larva measures 2.12 mm. in length. The first and last segments are flattened, projecting and rounded, and overlap the head (which is withdrawn while at rest) and the anal claspers. The body is much arched, having a prominent medio-dorsal ridge; the sides are flattened and slope down to a lateral ridge; the under surface is also flattened. The ground colour is of a pearly-white, thickly studded with very minute blackish points. Scattered over the whole surface are a number of bristles, varying in length, all having dark shining brown bulbous bases; the longest of these bristles are situated on the dorsal and lateral regions. The spiracles are also dark shining brown. The segments are beautifully marked with purplish-pink in the form of longitudinal stripes — a medio-dorsal stripe and four on each side. The most conspicuous of these stripes are the medio-dorsal and the lateral; the other stripes, two above and one below the spiracles on each side, are made up of oblique markings. The head and prolegs are brown, and the claspers whitish. A remarkable similarity exists between the buds of the thyme bloom and the larvae, both in colouring and pubescence. So great is the resemblance that it requires very close examination to discern the little larvae, especially as they generally conceal themselves within the blossom. Not infrequently a small hole is eaten through the calyx, and the larva may sometimes be seen with only the anal segments protruding through the hole." - Frohawk (1924)

3rd Instar

"After the second moult the larva is 3.18 mm. long. In appearance this stage is similar to the previous stage, excepting that the whole colouring is brighter, the hairs are a good deal longer, and the head is now shining black. In this stage the dorsal gland on the tenth segment becomes plainly visible. The larvae feed first on the downy exterior of the thyme blossoms, and then bore through the calyx so as to enable them to obtain access to the interior of the blossoms. They are especially fond of the substance of the base of the petals." - Frohawk (1924)

4th Instar

"The third and last moult takes place when the larva is about twenty days old. The exact age varies from eighteen to twenty-two days, according to temperature. Twenty hours after the third moult the larva only measures 3.18 mm. long when fully extended. The general colouring is now more uniformly ochreous-pink and duller. The first segment slopes to the front, and is sunk in the middle and rounded, so as to overlap the head; it bears a comparatively large central black dorsal disc. The three posterior segments are also compressed and sunk. The remaining segments (from the second to ninth inclusive) are humped sub-dorsally, a deep longitudinal medio-dorsal furrow being formed thereby. The sides are concave and a lateral ridge projects along the entire length, so as to conceal the legs and claspers. The larva bears four longitudinal rows of long curved hairs, there being a sub-dorsal and a lateral row on each side; each row includes a single hair on each segment from the fourth to the ninth inclusive, the sub-dorsal series terminating on the latter segment. The first three segments have each a set of three sub-dorsal hairs on each side, those on the first segment curving forwards. The hairs of the lateral rows are directed laterally and surround the extremities of the larva. All the hairs have bases of a remarkable structure, resembling glass-like pedestals with fluted sides. The entire surface of the body is densely studded with extremely minute pyriform glassy processes. The ventral aspect of the larva is densely clothed with very short stoutish hairs. On the tenth segment is the dorsal honey gland. The fully grown larva, after hibernation and still in its fourth stage, measures 14.8 mm. long. The head, which was proportionate to the size of the larva immediately after the third moult, when it measured only 3.2 mm. long, is now disproportionately small. It is set on a very flexible retractile neck, which can be readily protruded beyond the first segment while the larva is in motion, or completely hidden and withdrawn into the ventral surface of the segment when the larva is resting. In colour the larva is much duller than before, being of a pale creamy-ochre shade with a pinky-lilac tinge along the lateral ridge and over the first and last segments. The entire skin has a shining distended appearance as if too tight for its bulky contents. Owing to its great increase in size, the fully grown larva appears to differ considerably from examples which have just entered upon their fourth stage before hibernation. The black disc on the first segment now appears as a mere speck. The body is no longer narrowest at the middle segment, but gradually increases in width to the tenth segment. The medio-dorsal furrow is less pronounced owing to the swelling up of the body. The lateral ridge is dilated, prominent and much rounded. The ventral surface is full and of a bulbous contour. The feet appear disproportionately small, but their strongly curved hooks are very prominent. All the long dorsal hairs of the single larva, from which the foregoing description is taken, were broken off sharp, leaving only a series of fractured stumps. The bases of these hairs are very slender where they joined their pedestals. The fully grown larva described was found on June 3rd, 1906, just below the surface of the ground beside some small grass plants growing with thyme on the top of an ants' nest (D. flava), and pupated on the soil free of web on the evening of June 10th, 1906." - Frohawk (1924)

"Until 1915 failure had so far attended all attempts to ascertain the food and habit of the larva of arion after the third and final moult (usually occurring in August), when it leaves the thyme plants. It may here be mentioned that in captivity the larvae readily and greedily bore into and feed on both green peas and runner beans. On the former they have been kept alive for several weeks, during which time they increased in length from 3.2 mm. to 6.35 mm. This fact, coupled with the structure of the head and mouth parts, which is closely analogous to that of nearly related species which feed on succulent vegetable substances, suggested the idea that arion, in its last stage, might have been a vegetable feeder. Also its cannibalistic habits entirely cease after its third moult, which appears all the more curious owing to the fact that it is solely carnivorous during its last stage. Subsequently, in May, 1915, further search for arion larvae was made in one of its native haunts by Dr. T. A. Chapman and the author, when fortunately a larva was found by the Doctor on pulling up plants covering a nest of a common ant (Myrmica scabrinodis); it seemed to be amongst loose earth that the ants had worked over, and if not actually in the ants' nest, was within less than an inch of ground actually occupied by the ants. Its length when found was 11 mm., therefore not quite fully grown. This discovery determined what its food consisted of, viz., the larvae of ants, and that the ants' nest was its home. With this important step forward and keynote to success, Capt. E. B. Purefoy lost no opportunity in following up the problem surrounding this remarkable species; he not only succeeded in the undertaking, but worked out the complete life-history of arion in such a manner that every detail of its existence after the third moult was thoroughly investigated by him, resulting in complete success, and a large number of the butterflies were raised to perfection. This laborious task was undertaken expressly for the author, to enable him to complete the wonderful life-history of this species. Without his indefatigable and elaborate researches our present knowledge would in all probability have remained incomplete in this respect. The author must here express his deep indebtedness to Capt. Purefoy for his unlimited and generous assistance. The final moult (i.e., the third) usually occupies three or four days, and sometimes five days, and after the process it rests for several hours. During this resting period the larva does not arouse interest in passing ants. After about six hours' rest the larva starts wandering about in an aimless fashion, and being somewhat clumsy in its movements it often loses its foothold of a stem or leaf and falls to the ground, and, as Capt. Purefoy aptly remarks, it does not appear to have the faintest idea what it does want. If placed close to an ant-run or even at the entrance to a nest it will probably wander away in the opposite direction. It may and frequently does wander aimlessly about for many hours, and long pauses are made sometimes on the ground, sometimes on a stem. It is waiting for something, but does it know what? Finally when a foraging ant of the genus Myrmica comes across the larva, it at once takes great interest in it. Arion itself does not seem in the least pleased at the meeting. If it is wandering at the time it instantly stops, but if the ant leaves it for a moment it resumes its journey as if nothing had happened. Directly an ant meets the larva it immediately begins caressing it, waving its antennae over and upon it, at the same time closes its jaws and quickly starts "milking" it, i.e., imbibing the beads of liquid exuded from the dorsal honey gland on the tenth segment; the number of such beads which are exuded during the subsequent "courtship" is extraordinary. The ant leaves the larva at intervals and walks round again and again, returning each time to caress and "milk" it. The "courtship" often lasts for an hour or more. Finally by some mystic sense arion prepares itself to be carried off by the ant. It gives the signal to the ant by assuming a most extraordinary attitude, swelling up the thoracic segments, while the rest of its body remains of normal form. The ant, upon receiving the signal, gets well astride the larva, seizes it in its jaws between the third and fourth segments immediately behind the hunch, and at once starts off with its burden at a quick pace.

Highslide JSHighslide JS

The journey may be long or short, but all obstacles in their path are overcome, and the pair finally disappear down one of the entrances of the nest. It frequently happens that when the hunch is made the ant may not see the signal. The author has seen it repeated four times before it was noticed by the attendant ant while it was an inch away, facing in an opposite direction. The individual ant which first finds the larva is always the one to remain in attendance and carry it away. Although during the time several other ants may also find it, and stay by, and even "milk" it, they soon depart and leave it to the original attendant, who apparently instructs them that their services are not required. Having arrived at the entrance of the nest it descends with its burden deep down into the pitch darkness of the centre of the nest - a sombre contrast to its previous abode amid blossoming thyme in glowing sunlight. It there enters into its new existence and partakes of its first meal of the strange new pabulum, i.e., an ant larva of very small size. For the following five or six weeks it feeds and rapidly grows until it has trebled itself in size, and is vastly different in appearance, having become almost white (a fleshy-white) and resembling the fully grown larva in general form and colour, so very different to when carried into the nest. As winter approaches it settles down for hibernation in a cavity or chamber deep down in the nest, where the ant larvae in their last instar are tended by the workers. Probably, in some cases, the larva draws somewhat apart from the ants previous to hibernation, but as a general rule (according to observations made) it just remains in its chosen spot, and in many cases surrounded by its hosts. In the spring arion awakens from its long winter's sleep, and feeds again where it has slept without shifting from the spot. The question was raised by Dr. Chapman whether or not arion voided any faces after hibernation. By good fortune we were able to answer the question. A larva, on being disturbed by us about the middle of May, was seen to void two pellets in rapid succession. They were secured and sent to Dr. Chapman for examination. He reported they consisted of ant remains. Unlike its hosts, who enjoy the warmth of the sun and regularly bring their larger larvae and pupae up close to the surface to benefit by such heat, arion never attempts to do so, but remains in the depth and darkest part of the nest with the smaller ant larvae. It continues to feed on its diet of ant larvae until early in June, when it attains full growth, measuring 14.8 mm. long, without again moulting since the previous August, when it measured only 3.18 mm. in length, a very remarkable fact. When about to prepare for pupation it does not, as a rule, move from the spot where it has been living. It is in some cavity or chamber and fixes itself to the roof, attached by the anal claspers to a pad of silk, and gradually changes colour from an ochreous hue to a dull, dead white. The change to the pupa occupies about a week. At first the pupa is very lightly coloured, and the exertion of freeing itself of the larval skin pitches it slightly forward, and it remains hanging at an angle of 45 degrees. In a few days the cremastral hooks lose their hold, probably caused by the movements of the ants, and the pupa then lies at the bottom of the chamber. After remaining in the pupal state for about twenty-one days the imago emerges and finds its way through the ants' passages to the outer world of warmth and sunlight, when it immediately ascends some stem or other object to cling to until fully developed and ready for flight. Although the two first pupae discovered by the author in situ in Cornwall were away from ants' nests, the larvae must have wandered before pupation. Nevertheless, from observations carried out, in the great majority of cases, in a state of nature the pupae occur in the heart of the nest." - Frohawk (1924)

The pupa is formed in the ant chamber. It is attended by the ants which keep it clean. On emerging from the pupa, the butterfly crawls through the ant chamber to the surface, where it crawls up nearby vegetation before expanding its wings.

"The pupa bears a general affinity to the pupa of L. aegon [argus, Silver-studded Blue]. It is, however, much larger, measuring 12.7 mm. long. Dorsal view: Across the middle of its greatest diameter the pupa measures 5.44 mm. The head is obtuse; the base of the wings is slightly angular and swollen, while the wing itself is slightly concave. The abdomen is swollen at the third and fourth segments, becoming narrower and more rounded posteriorly. Lateral view: The pupa measures 4.75 mm. across the middle; the head is rounded; the thorax is convex and rises dorsally into a slight ridge; the meta-thorax and first abdominal segment are sunken; the abdomen is swollen at the middle and curves towards the posterior segments, which, as already mentioned, are rounded; the anal segment adheres closely to the ventral surface; cremastral hooks are absent; the wing is ample; it is swollen and rounded across the middle and extends to the fifth abdominal segment. The entire surface is minutely granulated and covered with very fine reticulations of a deep amber colour; the spiracles are prominent and blackish; the surface posteriorly adjoining them is covered with a number of shining raised bead-like processes, some bearing minute amber-coloured spines, which branch out on the apical half into extremely small bristles. At first the pupa is a clear pale apricot-yellow, but this tint very gradually deepens to a dark amber colour, excepting over the wings, which remain during this stage light ochreous. A few days before emergence the eyes and the hind margins of the wings gradually darken to a leaden-grey, and the head, thorax and abdomen also become darker. Then the median wing spots appear, and the whole pupa becomes uniformly deeper in colour until it attains a deep leaden-grey all over. It thereafter remains unchanged for over thirty hours immediately preceding emergence. The above description applies to a male pupa found wild on the surface of the ground in a shallow depression under the shelter of a stem of furze on July 12th, 1905; the imago (male) emerged on July 16th at 8.30 a.m. The adult larva previously described, which pupated on the evening of June 10th, produced a female imago on July 1st. The pupal state thus occupied twenty and a half days." - Frohawk (1924)