Volume 10 Number 9 17 November 2022
The Taxonomic Report
OF THE INTERNATIONAL LEPIDOPTERA SURVEY
ISSN 2643-4776 (print) / ISSN 2643-4806 (online)
Life history and ecology of the San Emigdio blue butterfly
(Lepidoptera: Lycaenidae)
Gregory R. Ballmer
5894 Grand Av., Riverside, CA 92504
eballmer@ gmail.com
ABSTRACT. The San Emigdio blue butterfly, Plebulina emigdionis (Grinnell, 1905), occurs in small,
scattered colonies in and near the southwestern Mojave Desert of California. Colonies depend on a symbiotic
relationship with the ant Formica francoueri (Bolton), and occur only where the ant’s range (primarily in more
mesic cis-montane habitats) narrowly overlaps that of the butterfly’s more widely distributed Atriplex larval hosts in
more xeric habitats. Colonies of P. emigdionis are often localized around a few host plants and, therefore, sensitive
to habitat changes due to anthropocentric causes and environmental stochasticity. The biology, ecology, and status
of known colonies of P emigdionis are presented with intent to offer insights into the species’ conservation. The
status of all P. emigdionis colony sites known from museum records, published accounts, and the personal records
of other lepidopterists is assessed.
KEY WORDS: myrmecophily, extinction.
TAXONOMY
The San Emigdio blue, Plebulina emigdionis (Grinnell, 1905) (Lepidoptera: Lycaenidae:
Polyommatinae), is a small butterfly (Figs. 1, 2) whose restricted range includes scattered
colonies in and around the western Mojave Desert of California (Comstock, 1927, Emmel &
Emmel, 1975). Although some recent authors place this taxon in the Holarctic genus Plebejus
(e.g. Balint & Johnson, 1995; Pelham, 2008), it is nevertheless distinguished by a unique suite of
morphological, ecological, and biochemical traits (e.g. Nabokov, 1945; Ballmer & Pratt, 1991a,
b; Pratt et al, 2006, Talavera et al, 2012) within the subtribe Polyommatina [= informal
Polyommatus Section (sensu Eliot, 1973)]. Talavera et al (ibid), used DNA markers to justify
reinstating the monotypic genus Plebulina Nabokov, 1945; they further indicated its
phylogenetic position as sister to the wholly Nearctic Icaricia Nabokov, 1945, and together with
the latter group of species, as sister to the remainder of Holarctic Polyommatina.
BIOLOGY/ECOLOGY
One remarkable feature of the distribution of P. emigdionis is its occurrence in small,
dense, discrete colonies, where adults and larvae are typically associated with just a few individual
larval host plants, often among or near many other apparently suitable host plants. Fordyce
Grinnell (1905) described the colony at the type locality in San Emigdio Canyon, southern Kern
County, CA, as “... extremely local, being found in only one place, and extending along the
canon for about a hundred yards.” Most, and perhaps all, extant and recently extirpated colonies
of P. emigdionis are/were similarly restricted in geographic extent.
The first reported host for P. emigdionis (as Lycaena Melimona) is that by W. G. Wright
(1905), who asserted that larvae fed on Acmispon americanus (Nutt.) Rybd. (Fabaceae) (as
Hosakia purshianus). Wright (ibid) considered L. Melimona to be a hybrid of Icaricia acmon
(Westwood) and Lycaeides melissa (W. H. Edwards), both of which he reported to use A.
americanus as a larval host. Although Pratt & Ballmer (1991) reported that larvae of P.
emigdionis fed on foliage (but not flowers or fruit) of Acmispon glaber (Vogel) Brouillet (as
Lotus scoparius (Nutt.) Ottley) under laboratory conditions, there are no recent reports of any
fabaceous host being used by P. emigdionis in the field.
Comstock & Dammers (1932) described the life history of P. emigdionis for a colony
near Victorville, CA, and reported the larval host as Atriplex canescens (Pursh.) Nutt
(Amaranthaceae), which is present at nearly all known P. emigdionis colony sites. In May 1989,
the author (with G. F. Pratt), conducted a line transect survey across a colony of P. emigdionis
dominated by two Atriplex species near Victorville, CA. Larvae of P. emigdionis were found on
two of 21 At. canescens and 23 of 97 At. torreyi (S. Watson) S. Watson shrubs. In a subsequent
survey (same month and same observers) of all Atriplex plants within a central 400 sq-ft portion
of the same colony, a total of 26 larvae and pupae were found on six of ten At. canescens and
105 larvae and pupae were found on ten of 11 At. torreyi plants. Adults and larvae of P.
emigdionis have also been associated with At. polycarpa (Torrey) S. Watson at three colony
sites: Helendale, San Bernardino Co. and Alabama Hills and Cartago Creek, Inyo Co., CA. In
captivity, larvae of P. emigdionis have also been reared on foliage of Atriplex lentiformis
(Torrey) S. Watson, a close relative of At. torreyi.
Larvae of P. emigdionis are primarily nocturnally active, although they have been
observed to forage diurnally on cloudy days in the field (G. F. Pratt, personal communication)
and while in captivity under subdued lighting conditions. During daylight hours, larvae of P.
emigdionis have been found resting on the crowns and undersides of prostrate branches of host
plants (figs. 5, 6), beneath leaf litter, and always in association with colonies of Formica
francoueri (Bolton) (Hymenoptera: Formicidae). This ant association was reported earlier (as F.
pilicornis) for the Victorville colony (Ballmer & Pratt, 1991b); now it is confirmed for all extant
colonies recently visited.
It is notable that Atriplex shrubs on which P. emigdionis larvae occur often also harbor
homopterous insects (various aphid and scale species), which produce fluid secretions on which
F. francoueri workers may forage. Scale insects found in association with shrubs utilized by P.
emigdionis larvae include Ceroplastes irregularis Cockerell (Cerococcidae), Lecanodiaspis
rufescens (Cockerell) (Lecanodiaspididae), and Orthezia annae Cockerell (Ortheziidae). Various
aphids have been found on plants occupied by P. emigdionis, primarily during the spring season
on young foliage, but only Pemphigus cf betae Doane has been found in close association with P.
emigdionis larvae and F. francoueri on the roots and crowns of host plants, notably near its
alternate host Populus fremontii S. Watson (Salicaceae).
The occurrence of P. emigdionis with F. francoueri colonies may indicate an obligate
relationship, although the mechanism for this symbiosis remains speculative. Ant presence is not
necessary for P. emigdionis larvae to grow and complete their development in captivity.
However, ant presence may be a necessary cue for oviposition, and may be important for
inhibiting larval predation and parasitization.
Ant symbioses with lycaenid larvae are frequently attributed to the ants providing
protection from insect predators and parasitoids in return for nutritive secretions from the larva’s
dorsal nectary organ (DNO) (see Malicky, 1970; Fiedler, 1995). Although the imbibing by ants
of DNO secretions has not been observed with captive P. emigdionis larvae, the possibility exists
that it may occur under field conditions.
Other ant organs associated with myrmecophily in P. emigdionis and other polyommatine
lycaenids include lenticles, eversible tentacular organs (TOs), and dendritic setae. Lenticles are
small, low cuticular structures (often resembling setal chalazae) with perforations which may
emit substances which affect ant behavior (Malicky, 1970). Paired TOs on the seventh
abdominal segment of most polyommatine species have been associated, when everted, with
excited ant behavior, presumably by releasing a semiochemical mimicking an ant alarm
pheromone. Ants in proximity to larvae of P. emigdionis which have everted their TOs usually
become agitated and/or are attracted to the larvae.
The function of dendritic setae in myrmecophily is not as well recognized as other so-
called ant organs, but seems to be a major contributor to myrmecophily in some species,
especially copper larvae (eg. Lycaena xanthoides (Boisduval)) which lack most other “ant
organs” (Ballmer and Pratt, 1991). Dendritic setae are not prominent on P. emigdionis larvae and
may not contribute to myrmecophily in this species (Ballmer and Pratt, ibid).
Plebulina emigdionis larvae often display their tentacle organs (TOs) on the eighth
abdominal segment when disturbed or while foraging. The TOs on foraging larvae appear to
attract nearby F. francoueri workers, which may serve to deter predators or parasites. Quiescent
larvae on the host plant crown also display their TOs when disturbed worker ants swarm out of
their subterranean nests, presumably in response to an alarm pheromone, to defend against the
invader. Such “alarmed” ants run around rapidly and often come in contact with any P.
emigdionis larvae which are present. Some such larvae often evert their TOs, either when
contacted by ants, or perhaps in response to the same disturbance which excited the ants. In
addition to chemical response, larvae may also evert their TOs in response to substrate-borne
vibrations, by which ants and many lycaenid larvae communicate (DeVries, 1991). Although
such agitated worker ants frequently contact P. emigdionis larvae, they do not usually remain in
protracted contact.
The display of TOs by P. emigdionis larvae varies according to the context. Foraging P.
emigdionis larvae, accompanied by F. francoueri workers, exhibit prolonged display of their
TOs, while quiescent P. emigdionis larvae on host plant crowns usually evert their TOs more
briefly, but repeatedly when disturbed. This could be interpreted as a prolonged release of an
aggregation pheromone in foraging larvae and a briefer, but more concentrated, release of an
alarm pheromone in quiescent larvae. This comports with observations that the same pheromone
may function as an ant alarm pheromone at high concentration and an aggregation pheromone at
lower concentrations (Henning, 1983).
Female P. emigdionis often deposit eggs on peripheral foliage of host plants (Figs. 2, 3),
although eggs may also be found on mature branches near the host plant crown (Fig. 4). In all
cases where eggs have been found, F. francoueri workers were also present on the plants.
Comstock & Dammers (1932) reported that the P. emigdionis egg stage (from oviposition in
May) lasted 8-10 days; the incubation temperature was not reported. In recent laboratory
observations, a captive female from Victorville, produced 28 ova within four days of capture; 10
eggs eclosed within four days, while the remainder eclosed within six days, of her capture. Thus,
the egg stage lasted a minimum four days and maximum six days; all ova were maintained at 80°
F (26.6° C). The differences in egg eclosion times reported here and by Comstock & Dammers
could be due to different incubation temperatures.
Several efforts to induce oviposition by captive females have been only modestly
successful. While occasional captive females produced up to a few dozen ova, more often they
produced only 1-2 or none. Anecdotal observations suggest that oviposition by captive females
may be increased when offered sprigs of host plants on which F. francoueri were present or had
recently been present; suggesting that, ant trail pheromones may serve to stimulate oviposition.
This also comports with field observations that host plants, with which adult female P.
emigdionis are closely associated, invariably also have foraging F. francoueri. Nevertheless,
ovipositing female P. emigdionis are not immune from attack by foraging F. francoueri and take
flight if approached closely by the ants.
Larvae of P. emigdionis (Fig. 7) are relatively glabrous; the ground color is matte pale
green-to-grayish green, or beige, which matches the foliage and/or woody branches and fallen
leaves of the host plants. Young larvae typically feed by scraping the leaf surface tissues but, as
they mature, may also feed at leaf margins (Fig. 8). Larvae may also mine the interior of leaves
(especially At. canescens), leaving the dry papery epidermis to resemble an empty paper bag.
Although the larva may be too large to entirely enter the leaf interior, its extensible neck allows
the head to do so while feeding.
During daylight hours, larvae remain concealed on the underside of prostrate branches
and below ground or leaf litter on the crown within a few inches of the soil surface. Such larvae
often occur in clusters (Figs. 5, 6) and are always associated with ant galleries around the plant
crown and roots.
Larvae have a few anatomical and behavioral traits which distinguish them from other
polyommatine lycaenids and which may serve to facilitate escape from predation. Specifically,
while many foliage-feeding lepidopterous larvae spin a silken lattice on the surface of foliage,
which enables firm attachment by their proleg crochets (similar to Velcro), P. emigdionis larvae
do not. Also, unlike typical lycaenid larvae, P. emigdionis larvae lack a fleshy median lobe on
the planta of the prolegs. These factors may increase larval survival by facilitating rapid
dropping to the ground at the approach of would-be predators and parasites. Such fallen larvae
initially remain motionless and are cryptic among leaf litter where they may gain protection from
foraging F. francoueri.
Pupae of P. emigdionis are pale tan and macroscopically glabrous (Figs. 9, 10). Pupation
occurs on the underside of prostrate branches and on root crowns in proximity to the soil surface
or beneath leaf litter. Pupae are loosely attached, as they lack a silken girdle and silken pad for
attachment, which are commonplace features among other polyommatine lycaenids.
Comstock & Dammers (ibid) reported that two larvae completed growth, pupated, and
eclosed in late July, while others from the same cohort of ova fed intermittently until entering
diapause in October. Similar results were obtained by the author from larvae reared in captivity
representing multiple wild colony sources. This comports with observations of a major spring
flight and partial subsequent flights, as indicated by occasional adults observed in the field
throughout the summer and early fall seasons. Larvae prefer to feed on fresh foliage, which is
most reliably available following late-fall and winter precipitation events, coincident with the
end of larval diapause. However, minimal vegetative growth may also occur on otherwise
dormant host plants during the normally dry summer and early fall seasons, especially following
sporadic summer monsoon rains.
Ballmer & Pratt (1991a) reported that P. emigdionis larvae pass through five instars prior
to pupation, based on larvae which matured and pupated without undergoing a semi-dormant
summer estivation. Those larvae were reared from ova from a female captured in April 1988 at
the Cuyama River Wash, San Luis Obispo Co., CA. The capture date suggests a female from an
overwintering larva; all larvae pupated and all but one pupa eclosed without entering diapause
(the single non-eclosing pupa died). A second cohort of larvae, the progeny of a female captured
in May 1989, at Victorville, CA, was reared under controlled conditions (12 hrs light: 12 hrs
dark) at a constant 80° F (26.6° C), until the temperature regulator failed and caused all larvae to
perish from hyperthermia in the early 5th instar. The mean duration in days was 5.0, 4.9, 5.1, and
6.8, respectively, for 1“, 2nd 3" and 4" instar larvae; as noted, all larvae died before completing
the 5" instar.
Recent observations with captive larvae reinforce the observed phenology of P.
emigdionis which entails a primary flight of adults in late winter to spring from overwintering
larvae. Eggs from the spring flight of adults produce larvae which largely (not entirely) enter a
semidormant quiescent stage in the penultimate larval instar. Those larvae remain more-or-less
dormant during the summer and fall seasons, but molt to the final larval instar and resume active
feeding and maturation in the subsequent winter season. The likely trigger for termination of
larval diapause may be seasonal changes in day length, daily temperature fluctuations, or both.
A cohort of captive P. emigdionis larvae, with F. francoueri workers, monitored daily
from May until the following January remained stationary (dormant) for several-to-many days at
a time. Occasionally, 1-2 larvae broke dormancy by molting and feeding for a brief period before
becoming quiescent again. All larval exuviae recovered between May and January had mean
width of associated cranial exuviae of 0.9 mm, which corresponds to that of A‘h instar larvae.
This suggests that observed sporadic feeding activity of such otherwise dormant larvae is not
accompanied by growth and maturation to the final larval instar until January. Post-diapause
larvae did not molt again prior to pupation.
Larvae of P. emigdionis are resistant to the annual summer drought, typical of their
environment, by virtue of a facultative semi-dormant state prior to the final larval instar.
However, they may not be able to endure prolonged drought resulting in extended absence
(beyond one season) of sufficient suitable host plant foliage. Thus, of 22 captive larvae which
broke diapause, molted to the final (prepupal) instar, and resumed active feeding in early winter,
only four pupated when fresh host plant foliage was withheld; and these pupae produced
abnormally small adults. The 18 less well-fed post-diapause larvae failed to pupate and died
without re-entering a dormant physiological state. This is unlike some other polyommatine
lycaenids, such as Euphilotes spp., which are capable of multiyear pupal diapause (Ballmer,
personal observations), or of multiyear larval diapause, as in the nymphalid Euphydryas editha
quino (Wright) (Pratt & Emmel, 2010). It remains unknown whether larvae may remain in
extended diapause under field conditions.
Natural enemies
As with other myrmecophilous lycaenids, ant attendance or proximity of ants to P.
emigdionis larvae may help to prevent predation or parasitization. While various hymenopterous
parasitoids, chiefly Braconidae, Chalcididae, Eulophidae, and Ichneumonidae (Hymenoptera),
have been reared from the immature stages of many other lycaenid species which have variable
degrees of ant associations, such parasitism of P. emigdionis larvae has not been observed. This
may be due to the larva’s normally nocturnal foraging behavior and/or close association with F.
francoueri. However, parasitoid tachinid flies (Diptera: Tachinidae) have been reared from P.
emigdionis pupae, collected as larvae in the field.
At least two eryciine tachinid species have been reared from P. emigdionis pupae, one of
which appears to be in or near the genus Siphosturmia, while the other has yet to be determined
(Fig.11). Based on voucher specimens in the Entomology Research Museum at the University of
California at Riverside, the tachinid species most frequently reared from various lycaenid
species, other than P. emigdionis, in Southern California is Aplomya theclarum (Scudder)
(Tachinidae, Eryciinae; see also Arnaud 1978). The tachinid species reared from P. emigdionis
may be specialized to parasitize P. emigdionis larvae.
Fig. 1. Adult P. emigdionis. Fig. 2. P. emigdionis ovipositing. Fig. 3. P. emigdionis egg on A.
canescens leaf. Fig. 4. P. emigdionis egg on A. canescens crown. Fig. 5. P. emigdionis larval
cluster on underside of prostrate branch. Fig. 6. F. francoueri with 2™ instar P. emigdionis
larvae. All figures are on A. canescens.
Fig. 7. P. emigdionis mature larva on At. canescens. Fig. 8. P. emigdionis larval feeding
damage on At. canescens. Fig. 9. P. emigdionis pupa, lateral view. Fig. 10. P. emigdionis pupa,
lateral view. Fig. 11. Tachinid fly ex P. emigdionis pupa. Fig. 12. P. emigdionis colony site near
Gorman, Los Angeles Co.
al © extant colonies
®@ presumed extirpated colonies
, ® status unknown epeaalse
Fig. 13. P. emigdionis colony site near Dome Spring, Ventura Co. Fig. 14. P. emigdionis
colony site at Conejo Gate, Coso Mts., Inyo Co. Fig. 15. burned colony site of P. emigdionis
near Cartago, Inyo Co. Fig. 16. P. emigdionis colony site in Mojave River Wash near
Victorville, San Bernardino Co. Fig. 17. Distribution and status of P. emigdionis colonies.
DISTRIBUTION
Colonies of P. emigdionis are known from a relatively narrow zone of overlap of the
primary ranges of its Atriplex hosts and ant associates, chiefly at the western and southwestern
margins of the Mojave Desert (Fig. 17). The elevation of known P. emigdionis colonies ranges
from 490 m at the Cuyama River Bridge (San Luis Obispo County) to 1780 m at El Conejo Gate
in the Coso Mts. (Inyo County). The Atriplex larval hosts are generally confined to xeric and/or
saline/alkaline soil environments and are collectively widespread in Southern California from the
deserts to the sea (Hickman, ed., 1993). Colonies of F. francoueri typically occur in relatively
mesic habitats of cis-montane southern California, including riparian, chaparral, and coniferous
forest communities. Although F. francoueri is rarely found in xeric habitats dominated by
Atriplex, and is uncommon in the Mojave Desert Biome, it is nonetheless invariably present in
association with P. emigdionis colonies.
Where F. francoueri colonies are associated with P. emigdionis, the ants may access soil
moisture at some depth below the dry soil surface (e.g., dry creek washes), or rely on fluids from
homopterous insects (various aphids and scales) associated with Atriplex. Limitations on the
distribution of P. emigdionis, particularly its absence from cis-montane southern California, are
speculative and may be related to intolerance of prolonged exposure to relatively cool, moist
atmospheric conditions, such as often occur within the cis-montane fog belt below 1000 meters
elevation. Coastal fog seldom penetrates inland to the Mojave Desert.
There are no recent records of P.emigdionis from the type locality, San Emigdio Canyon,
Kern Co., CA, although patches of suitable host plants occur there. San Emigdio Canyon is a
central feature of the Wind Wolves Preserve (WWP) bordering the south end of the San Joaquin
Valley, The WWP is managed by the Wildlands Conservancy to conserve and rehabilitate habitat
for native species. The canyon was subjected to intense cattle-grazing for many years, resulting
in major impacts to the local ecology. In visits to this location (March and April 2015), a few
clusters of At. canescens were found, primarily along the entrance road below the mouth of the
canyon, at an ornamental garden near the administrative headquarters, and a short distance
upstream from the mouth of the canyon and upslope from the creek wash. Neither P. emigdionis,
nor any F. francoueri was found in association with the Atriplex.
WWP management staff indicated that At. canescens was heavily grazed and trampled by
cattle, especially when other green foliage was scarce, and was likely more prevalent in the past.
This suggests the possibility that the colony noted by Grinnell (1905) may have been extirpated
by cattle grazing or other ranching activities. WWP management plans currently include re-
establishing At. canescens at additional sites within the preserve, and potential future re-
introduction of P. emigdionis. Another potential host plant for P. emigdionis at San Emigdio
Canyon is At. polycarpa, which is prevalent on the canyon’s lower alluvial fan (beyond the
WWP boundary) and widespread on adjacent portions of the San Joaquin Valley. This plant is
also abundant over much of the western Mojave Desert (Moe & Twisselman, 1995) where a few
P. emigdionis colonies are widely scattered. The nearest (to San Emigdio Canyon) known extant
colony of P. emigdionis is near Frazier Park, in Kern County (ca 26 air km SE of the mouth of
San Emigdio Canyon). This colony site is currently proposed for conversion to recreational uses.
10
Elsewhere in Kern County, a few colonies of P. emigdionis have been observed in the
Cache Creek drainage between the communities of Tehachapi and Mojave, and in the Kern River
drainage between Onyx and Weldon. Two colonies along the middle reach of Cache Creek
(vicinity of Cameron Road) apparently have been extirpated by a recent major debris flow
(Ballmer and K. Davenport, personal observations). One extant colony of P. emigdionis is
known (as of 2019) several km further upstream in Cache Creek Wash, along Sand Canyon
Road. Other colonies of P. emigdionis in the watershed of the middle reach of the Kern River, as
at Weldon and Onyx, appear to be suffering from drought, as few or no adult butterflies were
observed during spring 2021 (K. Davenport, personal communication).
The northernmost known population of P. emigdionis occurs in the Alabama Hills, west
of Lone Pine, Inyo Co. Two additional colonies of P. emigdionis in Inyo County are adjacent to
US Highway 395 near the communities of Cartago and Olancha (Fig. 14), several miles south of
Lone Pine. Further south, a previously observed colony of P. emigdionis near Sage Flat, west of
Haiwee Reservoir, seems to be extirpated, as no P. emigdionis adults have been found there in
recent years. The cause of this colony’s demise is not known, and no overt physical change to the
colony site is apparent. Another colony at El Conejo Gate in the Coso Mts. (Fig. 14), within the
China Lake Air Naval Weapons Test Center, was still present in July 2014, but no adult P.
emigdionis were observed when the site was last visited later in July and August 2014 (G.F.
Pratt, personal communication). This colony seems to have undergone a dramatic population
reduction compared to earlier years, although no obvious changes to the habitat have been noted
(G.F. Pratt, personal communication). The status of another colony site adjacent to the Owens
River, near Haiwee Reservoir, on land managed by the Los Angeles Department of Water and
Power, is not open to the public and has not been assessed in recent years.
A few colonies of P. emigdionis are historically known from the hilly north-central
region of Los Angeles County bordering the Mojave Desert. Some such colonies have been
displaced by Los Angeles regional urban sprawl. The occurrence of P. emigdionis in the
“Canyon Country” vicinity of Santa Clarita was noted in the course of a biological survey for a
proposed large housing tract during the 1990s (G. Bruyea, personal communication). A colony
formerly known from Solemint has been displaced by urban growth. Currently, one small colony
(occupying ca one acre) is known to persist in Los Angeles County adjacent to Interstate
Highway 5 in the general vicinity of Gorman (Fig. 13). This colony is embedded in a modified
coastal sage scrub community as described below.
The first reported colony of P. emigdionis (as Lycaena Melimona), was found in the
1880s on the south slope of the San Bernardino Mts. Wright (1906) described the location thus:
“The locality of Melimona is a little open mesa on the southern slope of the mountain, at an
altitude of 3,500 feet, and there I find it every year in June, but at no other time.” The location of
that site remains obscure, although a site resembling Wright’s description exists near Waterman
Canyon, is accessible by an old road, and is not far from Wright’s former residence in San
Bernardino. No colony of P. emigdionis is currently known from the San Bernardino Mts., but a
colony was formerly reported to occur at Baldy Mesa, just to the north of the Cajon Pass. The
Baldy Mesa area is currently subject to suburban sprawl, with extensive habitat alteration.
iL
All known extant colonies of P. emigdionis in San Bernardino County occur at least
several miles north-west of the San Bernardino Mountains, along the Mojave River. Small
colonies are scattered along the river’s dry wash margins over several miles between the
communities of Victorville and Helendale (Fig. 16). These colonies occur in close proximity to
elements of both riparian and Mojave Desert communities, as described below.
The westernmost known colony of P. emigdionis formerly occurred in the Cuyama River
Wash, adjacent to the Highway CA 166 bridge, near Cuyama. Although technically in San Luis
Obispo County, this site is very near the boundary of Santa Barbara County. The author (with G.
F. Pratt) visited this colony to obtain life history material in 1988; it was apparently extirpated
following subsequent bridge and highway reconstruction activities. When the site was re-visited
in 2015 and 2016, a few At. canescens shrubs were observed adjacent to the road margin at the
site of the former colony; but they were largely buried by debris from road construction activities
and no P. emigdionis was found there. Nor was any P. emigdionis found in searches of numerous
At. canescens shrubs up to | km both up- and down-stream from the former colony site.
Nevertheless, several miles upstream (and about 300 m higher elevation) along the Cuyama
River Wash in Ventura County, colonies of P. emigdionis occur along Lockwood Valley Road in
the vicinity of Dome Spring (Fig. 13).
HABITAT STRUCTURE
Colonies of P. emigdionis are associated with various plant communities, mostly within
or adjacent to the Mojave Desert Biome. Among the most diverse floral associations are those
along the Mojave River where colonies of P. emigdionis frequently occur in the ecotone between
desert and riparian communities. These colonies are associated with Atriplex canescens, At.
polycarpa, and At. torreyi in proximity to riparian species, such as cottonwood (Populus
fremontii S. Watson), various willows (Salix spp.), saltgrass (Distichlis spicata (L.)) and tamarisk
(Tamarix spp.), as well as typical desert species including creosote bush (Larrea tridentata
(D.C.) Coville), mesquite (Prosopis juliflora (SW.) D.C.), and rabbit brush (Ericameria
nauseosa (Pall. ex Pursh) G. L. Nesom & G. I. Baird).
Plant communities which support colonies of P. emigdionis are often dominated by
diverse trees and shrubs. Pinyon pine (Pinus monophylla Torr. & Frem.) and antelope bush
(Purshia glandulosa Curran) are prominent at Dome Spring, Joshua tree (Yucca brevifolia
Engelm) at El Conejo Gate, Coso Mts., canyon live oak (Quercus chrysolepis Liebm.), grape
soda bush lupine (Lupinus excubitus M. E. Jones), and Great Basin sagebrush (Artemisia
tridentata Nutt.) at Sage Flat. The colony near Gorman is in a modified Coastal Sage Scrub
community with scrub oak (Quercus sp.), California juniper (Juniperus californica Catrieri),
California buckwheat (Eriogonum fasciculatum (Benth.) Torr. & A. Gray), rabbit brush,
bladderpod (Peritoma arborea (Nutt.) Iltis), basketbush (Rhus aromatica Aiton), and giant wild
rye grass (Elymus condensatus J. Presl).
Adult P. emigdionis have been observed to imbibe nectar opportunistically from diverse
floral resources. These include ephemeral annual spring flowers during the early flight period
and various perennial floral resources later in the year. For example, at Helendale, San
Bernardino Co., annual plant nectar sources used by P. emigdionis, such as Amsinckia,
12
Cryptantha and Pectocarya (Boraginaceae), are most apparent during March and April, while
perennial nectar sources, such as Heliotropium curassavicum L. (Boraginaceae) become
available later in the flight season. Nectar resources, when present, may extend the life span and
reproductive capacity, of females.
HABITAT THREATS
The potential longevity, as well as fragility, of P. emigdionis colonies is exemplified by
the Victorville colony site which provided material for the original life history description
(Comstock & Dammers, 1932), as well as recent observations reported here. This colony is well
known to lepidopterists due to its convenient location adjacent to Interstate Highway 15 where it
crosses the Mojave River. But as of spring 2016, a major portion of this colony was buried
during construction of expanded bridge and highway infrastructure.
Other colonies of P. emigdionis are (or were) equally vulnerable to extirpation. The
colony adjacent to the CA State Route 166 bridge over the Cuyama River (San Luis Obispo
County) was apparently destroyed by highway construction activities, as noted above. The Dome
Spring population (Ventura County) is also concentrated near the paved highway and therefore at
risk should road widening activities occur. This site is further at risk from damage by cattle
which frequently graze the surrounding National Forest land and have been observed to both eat
and trample the Atriplex canescens host plants colonized by P. emigdionis (GRB, personal
observations). The Cartago colony, adjacent to highway US 395 in Inyo County (Fig. 15), was
largely damaged by a local brush fire caused by a motor vehicle crash in 2017. As noted above,
colonies of P. emigdionis in the so-called “Canyon Country” of Los Angeles County have been
displaced by urban growth, while the single known remaining colony in Los Angeles County
(vicinity of Gorman) is also vulnerable to potential expanded highway construction and related
activities. The site of the P. emigdionis colony near Frazier Park is currently proposed for
conversion to commercial recreational use.
Competitive displacement of Atriplex host plants by invasive exotic plants has been
observed, as by tamarisk (Tamarix spp.) at the Mojave River and Cache Creek localities. Non-
native giant reed (Arundo donax L.) also has potential to displace Atriplex spp. along the Mojave
River, while both that species and tamarisk may present wildfire threats, as well. It is also
notable that exotic grasses (chiefly Bromus diandrus Roth. and B. madritensis L.) have invaded
virtually all localities where P. emigdionis occurs and can contribute to ignition and spread of
wildfires.
Natural environmental stochasticity can also contribute to the demise of P. emigdionis
colonies. The loss of two colonies along Cache Creek (Kern County) following a debris flow
from a locally intense storm was discussed above. Recent extreme drought conditions may have
been a factor in the apparent extirpation of the Sage Flat colony, as well as the apparently
diminished, if not extirpated, El Conejo Gate colony, as noted above. Additionally, ongoing
climate change can be expected to affect all colonies of P. emigdionis to a still unknown degree.
The distribution of existing known P. emigdionis colonies may be interpreted as relicts of a
Pleistocene distribution when climate was probably cooler and generally moister in the Mojave
13
Desert region. Thus, a predicted warmer and drier climate in response to “global warming” may
cause greater pressure on survival of remaining colonies.
The potential threat from insect collectors cannot be ignored, though it should be noted
that the P. emigdionis colony where Highway I-15 crosses the Mojave River has been well
known among lepidopterists at least since the 1930s and continues to persist. The major threat to
that colony has been habitat destruction, as noted above, when recent expansion of the bridge
and related highway infrastructure obliterated a substantial portion of the colony. Similarly,
wanton and/or inadvertent destruction of that and other colonies of P. emigdionis along the
Mojave River (and elsewhere) are possible through ignorance or indifference of the relevant
public and private land owners and land use agencies. It is notable that extensive portions of the
Mojave River Wash’s natural communities have been supplanted by agricultural and pastoral
uses, aS well as by residential conversion. Thus, insect collecting appears to be a far less
important cause of past and likely future P. emigdionis colony extirpation than habitat loss and
degradation from other anthropogenic causes.
COLONY MANAGEMENT
While the preferred strategy for species conservation ordinarily entails preservation of
intact habitat, the current specter of rapid and dramatic climate change indicates that active
management may be required in the future. Although P. emigdionis has apparently survived
multiple periods of climatic extremes, particularly ice-age cold, likely prospects of future
extreme heat and extended drought, as well as changes in land use and wild lands management,
and spread of invasive species, pose additional challenges to P. emigdionis survival. Active
habitat management and/or transplantation of P. emigdionis colonies to more favorable locations
may be advisable. And, while captive breeding (if it were to become practical) may be useful in
augmenting colonies, it cannot substitute for self-sustaining colonies.
Critical factors in need of further study to inform colony management include the value
of nectar sources to adult P. emigdionis longevity and fecundity, food resources (including
homopterous insects) and soil moisture needs of F. francoueri, as well as soil moisture, land
disturbance, and plant competition affecting host plants. The status of managed P. emigdionis
colonies should be monitored annually by butterfly census and perhaps other factors relating to
host plant viability and ant presence, such as pesticide contamination, changes in soil chemistry,
etc. Where augmentation of host plants is deemed necessary, propagules should be obtained from
sources onsite or as near as possible. Adaptive management should consider all aspects of colony
health and be prepared to implement changes, as needed, with respect to unique features of each
colony site.
ACKNOWLEDGEMENTS
The observations reported here are the culmination of the work of many workers, but
three are especially notable. The author owes a great debt of gratitude to Gordon F. Pratt who
accompanied the author on numerous field trips and contributed many independent observations.
John F. Emmel collated records from museums and private collectors, thereby helping to
establish an early basis for describing the distribution of P. emigdionis. Ken Davenport also
14
accompanied the author in assessing the status of some colonies of P. emigdionis and offered his
recent assessment of others. Phillip Ward (University of California at Davis), James Deslauriers
(Chaffey College) and the late Roy Snelling (Los Angeles County Natural History Museum)
identified ants. Doug Yanega, Curator of the Entomology Research Museum at the University of
California at Riverside, identified parasitoid tachinid flies.
LITERATURE
Arnaud. P. H. J., 1978. A Host-Parasite Catalog of North American Tachinidae (Diptera). U.S.
Department of Agriculture, Science and Education Administration. Misc. Pub. No. 1319.
860 pp.
Ballmer, G.R. and G. F. Pratt, 1989 (1991a). A survey of the last instar larvae of the Lycaenidae
(Lepidoptera) of California. Journal of Research on the Lepidoptera 27: 1-80.
Ballmer, G.R. and G. F. Pratt, 1991b. Quantification of ant attendance (Myrmecophily) of lycaenid
larvae. Journal of Research on the Lepidoptera 30 (1-2): 95-112.
Comstock, J. A., 1927. Butterflies of California. Privately published, Los Angeles, CA. 334 pp,
63 plates.
Comstock, J. A. and C. M. Dammers, 1932. The metamorphoses of six California Lepidoptera.
Bulletin of the Southern California Academy of Sciences 31: 88-100.
DeVries, P.J., 1991. Call production by myrmecophilous riodinid and lycaenid butterfly
caterpillars (Lepidoptera): morphological, acoustical, functional, and evolutionary
patterns. American Museum Novitates No. 3025. 23 pp.
Eliot, J.N., 1973. The higher classification of the Lycaenidae (Lepidoptera): a tentative
arrangement. Bulletin of the British Museum of Natural History 28: 371-505.
Emmel, T. C. and J. F. Emmel, 1973. The Butterflies of Southern California. Natural History
Museum of Los Angeles County. 148 pp.
Fiedler, K., 1995. Ants benefit from attending facultatively myrmecophilous Lycaenidae
caterpillars: evidence from a survival study. Oecologia (1995) 104:316-322.
Grinnell, F. Jr., 1905. Lycaena emigdionis n. sp. Entomological News 16 (4): 115-116.
Henning, S. F., 1983. Chemical communication between lycaenid larvae (Lepidoptera:
Lycaenidae) and ants (Hymenoptera: Formicidae). Journal of the Entomological Society
of South Africa 46: 341-366.
Hickman, J. C., Ed., 1993. The Jepson Manual Higher Plants of California. University of
California Press, Berkeley, Los Angeles, London. 1400 pp.
Malicky, H., 1970. New Aspects on the association between lycaenid larvae (Lycaenidae) and
ants (Formicidae, Hymenoptera). Journal of the Lepidopterists Society 24:191-202.
1S
Moe, L. M. and E. C. Twisselman, 1995. A key to vascular plant species of Kern County
California by L. Maynard Moe & A flora of Kern County, California by Ernest C.
Twisselman. California Native Plant Society, Sacramento. 225 + 395 pp.
Nabokov, V., 1945. Notes on Neotropical Plebejinae (Lycaenidae, Lepidoptera). Psyche 51: 1-
61.
Pelham, J. P., 2008. A catalogue of the butterflies of the United States and Canada, with a
complete bibliography of the descriptive and systematic literature. The Lepidoptera
Research Foundation, Beverly Hills, CA. 658 pp.
Pratt, G.F. and G.R. Ballmer. 1991. Acceptance of Lotus scoparius (Fabaceae) by larvae of
Lycaenidae. Journal of the Lepidopterists Society. 45(3): 188-196.
Pratt, G. F., D. M. Wright, and G. R. Ballmer. 2006. Allozyme Phylogeny of North American
Blues (Polyommatini: Lycaenidae). Pan-Pacific Entomologist 82: 283-295.
Pratt, G.F. and J.F. Emmel. 2010. Sites chosen by diapausing or quiescent stage quino
checkerspot butterfly, Euphydryas editha quino (Lepidoptera: Nymphalidae) larvae.
Journal of Insect Conservation. 14: 107-114.
Talavera, G., V. A. Lukhtanov, N. E. Pierce, and R. Vila, 2012. Establishing criteria for higher-
level classification using molecular data: the systematics of Polyommatus blue butterflies
(Lepidoptera, Lycaenidae). Cladistics (2012) 1-27.
Wright, W. G., 1905. Butterflies of the West Coast of the United States. San Francisco. Whitaker
& Ray Co., Inc., vii +257 pp.
16
The Taxonomic Report
is a platinum open access peer-reviewed publication of
The International Lepidoptera Survey (TILS)
The International Lepidoptera Survey is registered as a non-profit Limited Liability Company (LLC) in the state of
Virginia, U.S.A. The Taxonomic Report (TTR), ISSN 2643-4776 (print) / ISSN 2643-4806 (online) is published for
the purpose of providing a public and permanent scientific record. Articles are peer-reviewed but not necessarily
through the anonymous editor-mediated review process. Typically, the authors are encouraged to solicit reviews of
their manuscripts from knowledgeable lepidopterists before submission. TTR appears in digital, open-access form,
is disseminated as a hardcopy to select institutional repositories, and is available as printed copy upon request at the
discretion of authors and/or the editor. Printing and postage charges may apply. An initial run of 25 copies is
printed on paper to meet ICZN Code recommendation 8B. All published TTR articles are freely available at the
archival TTR website (http://lepsurvey.carolinanature.com/report.html) and via the following digital repositories:
Internet Archive (https://archive.org/)
Biodiversity Heritage Library (https://www.biodiversitylibrary.org)
Zobodat (https://www.zobodat.at/)
Zenodo (https://zenodo.org)
TILS Purpose
TILS is devoted to the worldwide collection of Lepidoptera for the purpose of scientific discovery, determination,
and documentation, without which there can be no preservation.
TILS Motto
‘As a world community, we cannot protect that which we do not know”
Articles for publication are sought
Manuscripts may deal with any area of research on Lepidoptera, including faunal surveys, conservation topics, life
histories and foodplant records, matters of nomenclature, descriptions of new taxa, methods, etc. Taxonomic papers
are particularly welcome. There are no publication charges for authors. Before submitting a manuscript, email TTR
editor, Harry Pavulaan, 606 Hunton Place NE, Leesburg, VA, 20176, USA at intlepsurvey@gmail.com (cc: to
harrypav @hotmail.com) if you do not receive a reply within one week) to initiate discussion on how to best handle
your material for publication, and to discuss peer review options.
Visit The International Lepidoptera Survey on the World Wide Web at:
http:/Mepsurvey.carolinanature.com
&
Join the discussion at our list serve on Groups.io at:
https://groups.io/g/TILS
You can subscribe by sending an email to: TILS+subscribe @ groups.io
&
Join The International Lepidoptera Survey on Facebook at: https://www.facebook.com/groups/1072292259768446
LF