Phally polymorphism and reproductive biology in Ariolimax (Ariolimax) buttoni (Pilsbry and Vanatta, 1896) (Stylommatophora: Arionidae)*

2007·9.5 MB·English

by  Janet L Leonard

Amer. Maine. Bull. 23: 121-135
Phally polymorphism and reproductive biology in Ariolimax (Ariolimax) buttoni
(Pilsbry and Vanatta, 1896) (Stylommatophora: Arionidae)"^
Janet L. Leonard', Jane A. WestfalP, and John S. Pearse'
' Joseph M. Long Marine Laboratory, University ofCalifornia-Santa Cruz, SantaCruz,California,95060, U.S.A., [email protected]
^ Department ofAnatomy and Physiology, Kansas State University, Manhattan, Kansas66506-5802, U.S.A.
Abstract: Phally polymorphism, whereby some individuals in a hermaphroditic species have a complete functional penis (euphally),
whereas others lack a penis (aphally) or have a reduced, non-functional penis (hemiphally), has evolved many times in pulmonate
gastropods.Sincethediscoveryofapophallation (penisamputation) inArioliiiiaxMdrch, 1860,aphallatesinthisgenushavebeenattributed
to apophallation. In laboratory studies in Ariolimax (Ariolimax) Imttom (Pilsbry and Vanatta, 1896), we found aphally in juveniles as well
as in individuals reared in isolation to adulthood, demonstrating that the aphaliate condition is not always due to apophallation. Some
aphallateindividualsreared in isolationfromhatchingproducedeggsandviablehatchlings,providingthefirstdemonstrationofuniparental
reproduction in this species. Egg-to-egggeneration time in laboratory-reared individuals ranged from 10 monthsto more than 24 months.
Anatomical data also elucidatethe reproductivecycleofthisspecies. Four reproductivestateshavebeen identifiedbytheappearance ofthe
reproductivesystem. Springandearlysummerpopulationsconsistofindividualsintheimmatureand intermediatereproductivestates.The
hypertrophied state was found from autumn until early spring. Egg-laying occurred in the laboratory in the fall and winter. Copulation
consistsofunilateralorsimultaneouslyreciprocal intromissionsandoccurredin thelaboratorybetween FebruaryandSeptember.Verylong
copulations (more than 7 h) are more frequent than in other species ofAriolimax. Phally polymorphism, uniparental reproduction, and
the variation in generation time should play important roles in determining the variance ofmating success and the potential for sexual
selection in this hermaphroditic species.
Key words: gastropod, genitalia, aphally, life history, self-fertilization
Banana slugs, giant stylommatophoran slugs of the ge- One ofthemost mysteriousaspectsofAriolimaxbiology
nus Ariolimax Morch, 1860, are common and conspicuous is the existence of aphallate individuals and the extent to
members oftemperate rain forests and other mesic habitats which these can be explained by apophallation. Apophalla-
of the northwestern coast of North America. Although ba- tion is a behavior first observed in Ariolimax (Meadarioii)
nana slugs are well-known and popular with the general californicus (Cooper, 1872) (Heath 1916, Leonard et al.
public, remarkably little is known about their biology (but 2002) and subsequently in Ariolimax (Meadarioii) dolicho-
see Harper 1988, Leonard etal. 2002, Cody 2006, Pearson et phalhts Mead, 1943 (Mead 1942, 1943, Harper 1988, Leo-
al. 2006). InAriolimax, as in stylommatophorans in general, nard etal. 2002) whereby the penis is sometimes chewed oft
taxonomy has been based on genital characters (Pilsbry at the end of copulation in these simultaneous hermaphro-
1948). Eberhard (1985) suggested that where genital char- dites. Since Heath (1916) most authors have been content to
acters have evolved rapidly enough to distinguish species explain aphallate individuals ofAriolimax as the product of
(and even subspecies), sexual selection has played an impor- apophallation. However, Heath himself expressed doubt,
tant role in the evolution ofthe group. The genusAriolimax saying, “years ago I visited Hog Island in Tomales Bay, and
offers a particularly good opportunityto test this hypothesis found over 400 specimens ... every one ofthe specimens was
since it consists ofa small number oftaxa (Rotb and Sade- totally lacking a penis or any sign ofone.” (quoted in Mead,
ghian 2003, Pearse et al. 2007), most ofwhich are found in 1943, p. 685). Mead (1943) later collected in the same area,
coastal Central California in c]uite similar habitats, but findingno slugs at Hog Island and an entirelyphallate popu-
which have very divergent genitalia and sexual behavior lation nearby. Pauli (1951) examined the gross anatomy of
(Leonard etal. 2002, 2005). Accurate information about the Ariolimax buttoni (Pilsbry and Vanatta, 1896) (= A. colum-
reproductive biology ofAriolimax spp. is necessary to per- bianus (Gould in Binny, 1951 ), see below) at Mills College in
form such a test. Oakland, California and found that of the 27 adult speci-
From the symposium “Gastropod Mating Systems” presented at the joint meeting ofthe American Malacological Society and Western
Society ofMalacologists, held 26-30 June 2005 at Asilomar, Pacific Grove, California.
121
122 AMERICAN MALACOLOGICAL BULLETIN 23 • 1/2 • 2007
mens she dissected, none had a penis. More recently, Roth Pilsbry 1948) are evolutionarily distinct from the more
(2004) reviewed anatomical descriptions ofArioUmax in de- southern populations (Leonard et al. 2005, Pearse et al.
tail, and suggestedthat thedescriptions are inconsistentwith 2005). The name ArioUmax buttoni (Pilsbry & Vanatta,
what would be expected if a penis were amputated during 1896) has been revived to designate the southern clades for-
copulation. He argued that apophallation should sever both merly included in A. columbiamis (Pearse et al. 2007). Ario-
the penis andvas deferens, leaving unconnected remnants of Umax buttoui and A. columbiamis do not even represent
both and an intact penis retractor muscle, whereas Pilsbry sister clades (Pearse et al. 2007). Like A. columbiamis, A.
and Vanatta’s 1896) drawing ofAphallariou Imttoni showed buttoui may be either maculate or immaculate. ArioUmax
(
an intact vas deferens connected to the genital atrium by a (Meadariou) brachyphalliis Mead, 1943 is sympatric with A.
shortbulbwith nopenisand no penis retractormuscle. Roth buttoui in San Francisco but all individuals used in thisstudy
(2004) concluded that the descriptionswere more consistent from San Francisco have been identified bymolecular mark-
with a phally polymorphism. Such phally polymorphisms ers as belonging to A. buttoui. There are no reports ofsym-
are well-known and widespread in pulmonates, including patric ariolimacines in Alameda, Sacramento, Mendocino,
stylommatophorans (Tompa 1984, Pokryszko 1987, Baur or Marin counties.
and Chen 1993, Doums and Jarne 1996, Viard et al. 1997,
Animals
Doums et nl. 1998, Gomez 2001), and there are even in-
stances ofa phally polymorphism having lead to an errone- Mills College animals
ous splitting ot a single species into two genera (e.g., Lace Specimens were collected from March 1951 to February
1992). 1952 in two locations; the Mills College campus (37° 46’
Here, we present previously unpublished observations 41”N, 122° 10’ 49”W) and the nearby Leona Heights Park
on aphallate and euphallate individuals in two populations (37° 47’ 31”N, 122° 10’ 41”W) in Oakland,Alameda County,
ofArioUmax (ArioUmax) buttoniand data showingaphallyin California. Approximately 24 animals from each location
individuals which cannot have been involved in apophalla- were housed in the laboratory in terraria filled with damp
tion because they were reared in isolation to adulthood. earth and fed with lettuce. Other animals were dissected
Theseobservations demonstrate thatthisspecieshasa phally shortly after collection.
polymorphism, with both aphallate and euphallate individu-
j
als in some populations. We also present laboratory obser- UCSC animals '
vations on the sexual behavior and life cycle ofA. buttoui Specimens were collected from five locations in Central
and document uniparental reproduction in this species. California: (i) a wooded area of Mount Parnassus on the [j
campus of the University of California, San Francisco, San
!j
MATERIALS AND METHODS Francisco County (37°45’ 38”N; 122° 27’ 28”W); (ii) a levee I
ofStaten Island in the Cosumnes River, Sacramento County
Taxonomy (38° 15’ 56”N, 121° 26’ 31”W); (hi) near Comptche, Men- |
The animals used in this study are ArioUmax (ArioU- docino County (39° 15’ 54” N, 122° 35’ 24” W); and two |
max) buttoni. This species was first described by Pilsbry and locations in Marin County, namely (iv) near MuirWoods in fi
Vanatta (1896) from a large series ofaphallic ariolimacines central Marin County (37° 53’ 07”N, 122° 32’ 26”W) and (v)
from Oakland, California as a new genus. Subsequently, this the west side ofTomales Bay (38° 10’ 25”N, 122° 55’ 26”W) |j
taxon was synonymized to ArioUmax (ArioUmax) coUimbi- in western Marin County. Animals were maintained in the j
anus (Gould in Binny, 1951 (Waste 1940, Mead 1943, Pils- laboratory as groups or individuals in plastic boxes as de-
)
bry 1948) in the expectation that the aphally could be at- scribed elsewhere (Leonard et al. 2002). I
tributed to apophallation during copulation. This species
Anatomical studies
was considered to be the onlyariolimacine to include macu-
late individuals and was stated to have a range extending Mills College animals
from Tuolomne County, Monterey County, the eastern A total of 67 individuals ofArioUmax buttoui, ranging
shore ofSan Francisco Bay, and the city ofSan Francisco in from 16 to 55 grams in weight, were dissected. Three slugs
|
California, north to southeastern Alaska (Pilsbry 1948, Roth were dissected at the beginning and middle ofeach month
and Sadeghian 2003). Recent molecular evidence suggests from March to May 1951. Five slugs were dissected every
that A. columbiauus, as defined by Mead (1943) and Pilsbry two weeks from October 1951 to February 1952. Two ofthe
|
(1948), is not monophyletic but rather that populations of slugs were taken from the animals kept in terraria whereas '
ArioUmax north ofMendocino County, California (the true the other three were collected from the field. The daybefore
A. coUimbiamis, since the species was described from speci- dissection, the slugs were weighed and the next day they
mens collected near the Columbia River, see discussion in were drowned, dried to remove excess mucus, pinned to a j
|
PHALLY POLYMORPHISM IN ARIOLIMAX BUTTONI 123
dissecting tray and dissected in water. A longitudinal inci- group boxes. Eggs laid in tall and early winter 2003-2004
sion was made on the right side from the caudal pore to the were maintained as described previously (Leonard et al.
mouth, the dorsal skin retracted, and the genitalia freed by 2002) and hatchlings weighed when found and transferred
severing connective tissue and the protractor muscles offhe to individual plastic containers. All containers were cleaned
tentacles. After observation ofthe intact condition, the skin and lettuce added weekly. After the juveniles reached 5 g in
was cut around the atrium and the reproductive system was weight, drycat food was also given. Slugs were transferred to
laid out separately for measurement. The data recorded larger containers as necessary as they grew. They were
were: a) the color and degree of development of the repro- weighed weekly until 6 mo and biweekly thereafter.
ductive system; b) the position of the vas deferens and the All hatchlings were reared in isolation until lime 2004
presence or absence ofa penis and/or terminal bulb on the when, as part of a mating study (Leonard, et al. unpubl.
vas deferens; c) the approx, width ofthe vagina and oviduct; obs.), 40 individuals from the Staten Island group and 40
d) the approx, length and wicith ofthe spermatheca; e) the from the Tomales Bay group were randomly assigned to a
approx, length and width ofthe ovotestis; and f) the approx, matingtreatment. Eight individuals, each from Staten Island
length ofthe albumen gland. and Tomales Bay parents, were assigned to the single animal
(continued isolation) treatment. Equal numbers ofindividu-
UCSCAnimals als from Tomales Bay and Staten Island parents were also
Anatomical studies were conducted withAriolimax but- assigned to each of four other rearing conditions: a) paired
toni derived from two populations (Staten Island and To- with an individual from the same source population; b)
males Bay) and reared as described by Leonard etal. (2002). paired with an individual from the other source population;
Three of the individuals were collected from the field and c) in a group box with three others from the same source
then maintained in the laboratory until their deaths, at population; or d) in a group box with one individual from
which time they were frozen. The other individuals were the same source population and two from the other source
hatched in the laboratory and frozen when slightly over 30 population. All of the individuals derived from the Staten
months of age. They were thawed just before dissection, Island population were spotted whereas all of the Tomales
pinned to a dissecting tray, and dissected under water. An Bay descendants were immaculate. Redwood bark mulch
incision was made through the bodywall with scissors along was added to each box to encourage egg-laying. Slugs were
the left side, just above the foot, and the dorsal body wall weighed biweeklyand cleaned and fedweekly (as in Leonard
peeled back to expose the digestive system and genital or- et al. 2002), so that all boxes were checked for eggs at least
gans, andthealbumen gland, ovotestes, and maleandfemale weekly. Clutches found were treated as described above.
parts ofthe genital organs teased apart. The condition ofthe Hatchlings were weighed when found and preserved in 95%
spermatotheca and the presence or absence ofthe penis was ethanol for future genetic analysis.
noted. Ifabsent, the presence or absence ofa penis stub was
noted aswell as the course ofthevas deferens. The male and Behavioral observations
female portions ofthe genital organs were sketched for most All observations on sexual behavior reported here for
specimens; in a few cases they were measured and photo- Ariolimax biittotii (see Table 3) resulted from casual obser-
graphed through the dissecting microscope, using an vations of UCSC animals held in group boxes. Where ani-
Olympus Camedia C-3040 digital camera. Upon comple- mals were observed to be copulating when the box was
tion, the dissected animals were preserved and archived in opened for cleaning, notes were made and/or the behavior
10% formalin. was videotaped as time allowed. Where possible, observa-
tions were continued for 30 min after the end ofthe copu-
Rearing studies lation. The copulation of Mendocino individuals on April
Two sets of rearing studies have been conducted with 10, 2001 was noted but not followed for any length oftime.
Ariolimax buttoni. The first series was conducted at Mills
College (Mills College animals) from 1951 to 1952 (Westfall
1960). Eggs and juveniles were maintained in terraria filled RESULTS
with damp earth along with the parents. They were fed let-
tuceleaves. In total 17juveniles ranging in age from 1 dayto Phally status
19 weeks were taken for anatomical study.
The second rearing study was conducted at the Long Mills College animals
Marine Laboratory of the University of California-Santa Only 7 of the 67 Mills College slugs (4 of the 55 slugs
Cruz (UCSC animals) from 2003 to 2006. Slugs were col- collected on the Mills College campus, and 3 ofthe 12 slugs
lected from Staten Island and Tomales Bay and held in collected at Leona Park), were found to have a penis and, of
)
124 AMERICAN MALACOLOGICAL BULLETIN 23 • 1/2 • 2007
those, one only had a penis fragment (Table 1). In 59 ofthe immature, 2) intermediate (or “sperm producing”), 3) hy- i
60 individuals lacking a penis, the vas deferens ended pertrophied (or “egg-laying”), and 4) “old” (Appendix). The
abruptly in the atrial mesentery (fig. 1). In 20 of these 59 reproductive state in the dissected slugs varied with season
'
individuals, there was a small bulb at the end of the vas (fig. 3). In the spring most (10/12) individuals were in the j
deferens as described by Pilsbry and Vanatta (1896). The intermediatestate (fig. 3) with an albumen gland ofsmall to
!
apical termination ofthe vas deferens in the individual with intermediatesize. Incontrast, in thefallandwintertherewas 7
a fragment ofa penis was on the penis fragment. This speci- a clear division into two groups ofindividuals: those in the ^•
men was in the immature reproductive state (see below). immaturestatewithaverysmallalbumen glandand thosein !i
None ofthe 17 Mills College juveniles hatched in the labo- the hypertrophied state with a very large albumen gland.
|!
ratory had a penis. Slugs collected in October and kept in captivity were also i’
UCSC atiimals foundto be moreoften in the hypertrophied stateasthedate
A total of 16 individuals were dissected (Table 2). All of of dissection moved from November to february. -j.
the individuals derived from Tomales Bay populations and There was no relationship between season or reproduc- i'
the individuals collected from UCSf and Central Marin had tive state andthe presence ofa penis, four ofthe individuals :|
a fully developed penis with an apical retractor muscle and with a peniswere in the immature reproductive state, one in
a long vas deferens that terminated in the penis (Eig. 2B). the intermediate state, and two others in the hypertrophied
None of the 10 individuals derived from the Staten Island state (Appendix; fig. 3). In addition, there was no relation- j|
population had a penis (fig. 2A), including two individuals ship between body weight and the presence of a penis; the
that spent their entire lives in isolation. Of these 10 indi- second smallest and third largest animals dissected had a
viduals, 5 had a small stub where the penis would be in a penis (fig. 4). Each reproductive state included individuals
euphallate individual (fig. 2A) whereas the remaining 5 with a broad range ofweights although the smallest animals
lacked any trace ofa penis (fig. 1). All 10 individuals lacked were in the immature state and the largest in the hypertro-
a penial retractor muscle and 9/10 had a blind termination phied state (fig. 4).
ofthe vas deferens. In one individual the vas deferens ended The reproductive tract was not distinguishable in juve-
in a bulb as was seen in many ofthe Mills College animals niles for two weeks after hatching. After two weeks, the
(Table 1). The vas deferens connected to the base of the organs of the reproductive system were in the immature .
penial stub in one individual. state with a hair-like hermaphroditic duct, leading from a '
small, white, smooth, lobed ovotestis and a small albumen
Reproductive development gland. After a short distance this duct divided into a free
oviduct and a vas deferens. These ducts were straight and
Mills College aiiii?iah hair-like in younger specimens but began to show more
Measurements and appearance ofthe individual organs coiling in older specimens. In olderjuveniles the vagina and
of the reproductive tract for the 67 dissected adult Mills vas deferens terminated jointly in the atrium (as in fig. 1)
Collegeslugsaregiven in Westfall 1952). On thebasis ofthe buttherewasno indication ofapeniseven at 19weeksofage
(
color and appearance of the organs and the length of the when other reproductive organs were well developed. Juve-
albumen gland, four reproductive states were identified: 1) nile slugs of 15-19 weeks of age weighed between 3 and
Table 1. Summary ofmorphological data on adult Arioliinax biittoni from Mills College in Oakland, California
Number in
Number Weight Number in Number in hypertrophied
Phally status ofslugs (mean ± SD) immature state* intermediate state* state* Number with state unclear
Penis present 6 32.12 1 129g 3 0 2 1 (state questionable)
Penis fragment 1 37.5 g 1 0 0
No penis but bulb on 20 32.419.08 g 2 9 8 1 (state questionable)
end ofvas deferens
No penis and no bulb 39 29.3 ± 5.62 g 18 13 8
Unknown 1 24.5 g (insufficient information
Weight (mean ± SD) 28.42 ± 6.61 g 30.49± 6.27g 33.31 ± 9.52g 32.53 ± 9.91 g
Total 67 24 22 18 3
*SeeAppendix for definition ofreproductive states.
PHALLY POLYMORPHISM IN ARIOLIMAX BUTTONI 125
UCSC animals
The spermatheca was reddish in most individuals from
Staten Island. All Tomales Bay animals had a small sper-
matheca and in 3 ofthe 4, it was pale in color. While all of
the isolated, dissected animals from Staten Island (3/8), laid
Albumengland
one or more clutches ofeggs, none ofthe isolated Tomales
Bay animals laid eggs (see below).
Egg-laying
Results on egg-laying and hatching in the Mills College
animals have been published elsewhere (Westfall 1960). In
UCSC animals, egg-laying by animals hatched in the fall-
winter of 2003-2004 from Staten Island and Tomales Bay
Ovotestis
Hermaphroditicductwithspermatozoa parents, began in October 2004 and continued until late
Ovispermductwitheggs February 2005. Egg-laying then paused for this group of
slugs until mid-August 2005 when it resumed and continued
until late February 2006. However, only in one box offour
individuals did egg production occur in both seasons (2004-
2005 and 2005-2006).
Three individuals from Staten Island parents, isolated
Vasdeferens
since hatching, laid eggs (only two were dissected and listed
Spermatheca in Table 2). One individual, from a clutch of 13 November
2003, produced three clutches for a total ot 89 eggs: 32 eggs
Freeoviduct
on 13 December 2004 (Clutch lA), 13 eggs on 20 December-
2004 (Clutch IB), and 45 eggs on 9 February 2005 (Ckrtch
Vagina
1C). This individual was found dead on 4 August 2005. All
Genitalorifice three clutches developed normally and produced a total of
45 viable hatchlings. Clutch lA produced 21 hatchlings be-
Figure 1. Drawing ofa dissected aphallate individual otAriolimax tween 31 lanuaryand 7 Febrrrary 2005. Clutch IB produced
biittoni which had been observed to lay eggs before dissection;
hypertrophied reproductive state (see Appendix) with sperm 6 hatchlings from three eggsbetween 3 and 9 February 2005:
present in thehermaphroditic duct and theoviduct distended with one egg produced a single hatchling, one produced twins,
eggs. The hermaphroditic duct proceeds from the ovotestis (=her- and oneproduced triplets. The production ofmore than one
maphroditic gland = gonad) carrying both eggs and sperm. After hatchling fror-n an egg has also been seen in other species of
thealbumengland,thetermovispermductisused foraconvoluted Ariolimax (Leonard et ai, unpubl. data; B. Miller, pers.
conjoinedvasdeferensandoviduct.Afterthevasdeferensbranches coirrm.). Clutch IC produced 18 hatchlings 24-28 March
off(tothepenisineuphallatespecimens,seeFigure2B) toendnear 2005.
the atrium here, the ovisperm duct continuesasthe free oviduct to The second individual (from a clutch found 28 Novem-
the point ofattachment ofthe spermatheca (=bursa copulatrix = ber 2003) laid a total of 120 eggs, 101 which hatched: a) 28
gametolytic gland) with the vagina. The free oviduct is usually a eggs found 12 October 2005, 16 ofwhich hatched; b) 17 eggs
wide, convoluted duct while the vagina is a short, straight tube found 18 November 2005, 18 of which hatched, including
from the spermatheca to the atrium, and is divided into two por- one set of twins; c) 56 eggs found 8 December 2005, 50 ot
tions by a small, thick, annular muscle (intrinsic muscle of the
which hatched; and d) 19 eggs found 17 February 2006, 17
vagina).
ofwhich hatched. The third individual, from a clutch of 13
November 2003 [a clutch mate of the first individual
(above)], was found with a clutch of80 eggs on 8 December
6.3 g (Westfall 1960). The juvenile slugs all showed the 2005, which produced 66 hatchlings, and a clutch of56 eggs
immature reproductive state (Appendix), except that the on 19 lanuary2006which produced 55 hatchlings. Thelatter-
very first juvenile examined, dissected at four weeks of age two individuals sui-vived until frozen on 22 June 2006, and
(representative weight, 0.7 g, Westfall 1960), was found to were found to be aphallate when dissected (Table 2).
have a relatively large, loosely lobed, ovotestis. The reason A box of four individuals, two from Staten Island par-
for the different appearance of this one individual is not ents andtwo fror-n Tomales Bayparents produced a clutch of
known. 17 eggs on 14 October 2004. The oldest individual in thebox
126 AMERICAN MALACOLOGICAL BULLETIN 23 • 1/2 • 2007
Table 2. Summary ofdissections ofAriolimax biittoni held at UCSC at Santa Cruz, California
Population Isolate/Group-held/Collected? Penis Vas deferens Spermatheca Egg laying
Central Marin Collected as adult Very large bulbous Connects to penis Never laid eggs in
lab
UCSF Collected as adult Very thick Connects to penis Sac-like Never laid eggs in
lab
Staten Island Collected as adult No trace ofpenis Blind termination Reddish, flaccid Never laid eggs in
lab
Staten Island Reared in isolation throughout Small penial stub Blind termination Reddish Laid fertile eggs
life
Staten Island Reared in isolation throughout Small penial stub Blind termination Reddish, sac-like Laid fertile eggs
life
Staten Island Laboratory reared, paired at 6 Splayed penial stub Blind termination Dark red 3 clutches offertile
months ofage eggs from pair
Staten Island Laboratory reared, paired at 6 No trace ofpenis Ends in genital pore 3 clutches offertile
months ofage eggs from pair
Staten Island Laboratory reared, paired at 6 No trace ofpenis Connects to base of Red, flaccid Fertile eggs laid by
months ofage penial stub near pair
atrium
Staten Island Laboratoiy reared, grouped at No trace ofpenis; Blind termination Large Several clutches of
6 months ofage 1 cm ofvagina fertile eggs laid
everted by group
Staten Island Laboratory reared, grouped at Small penial bulb Blind termination Dark reddish Several clutches of
6 months ofage fertile eggs laid
by group
Staten Island Laboratory reared, grouped at No trace ofpenis Ends in bulb Several clutches of
6 months ofage fertile eggs laid
by group
Staten Island Laboratory reared, grouped at No trace ofpenis; Blind termination Large, flaccid, red Several clutches of
6 months ofage 0.5 cm ofgenital fertile eggs laid
pore everted by group
Tomales Bay Reared in isolation throughout Penis present Connects to penis Flabby, small Never laid eggs
life
Tomales Bay Reared in isolation throughout Penis present Connects to penis Reddish, moderately Never laid eggs
life small
Tomales Bay Reared in isolation throughout Penis present Connects to penis Small, pale Never laid eggs
life
Tomales Bay Laboratory reared, paired at 6 Penis present Connects to penis Small, creamy, firm Never laid eggs
months ot age
came from a clutch of 13 November 2004, so the parent was maculate) slug,wasfoundlayingeggs.Atotalof54eggswere
no more than 11 months old at the time ofegg-laying. This found in the box at that time, although it is not clear that
clutch produced 3 hatchlings. On 11 November 2004, a they were all laid by the spotted mother. Lorty-one
clutch of 29 eggs was found in a box containing a pair of hatchlings were produced from this clutch. The spotted slug
individuals, both from Tomales Bay parents, one from a washatchedfrom aclutch found 16 January2004andsowas
clutch found 30 December 2003 and the other from a clutch approx. 10 months and eight days in age (from the egg).
found 2 January 2004. This means that the individual laying Nine more eggs were found in this box on 24 November
the eggs could not have been older than 10.5 months from 2004 and this clutch produced 2 hatchlings.
the date the egg was laid. Eourteen of the 29 eggs in this
clutch hatched. A second clutch of 12 eggs found in this box Sexual Behavior
on 13 December 2004 produced 6 hatchlings. On 24 No- Little is known about sexual behavior in Ariolimax hiit-
vember 2004 the Staten Island (spotted) individual ofa pair toni. To date, we have never observed a complete sexual
of one Staten Island (spotted) and one Tomales Bay (im- interaction from courtship to the termination ofcopulation
PHALLY POLYMORPHISM IN ARIOLIMAX BUTTONI 127
Figure 2. Photographs of dissections of
aphallate (A) and euphallate (B) individu-
als ofAriaUmax buttoiii. Individual A was
laboratory reared from Staten Island par-
ents. Individual B was laboratory reared
from Tomales Bay parents. AVD, ascend-
ingvasdeferens; DVD,descendingvasdef-
erens; P,penis; PR, penisstub;V,vagina;
location of atrium.
100 Figure 3. Relationship between reproduc-
90 hyperwith penis tivestate (Appendix),season,andalbumen
hypertrophied gland length in 66 individuals Ariolimax
BE 80 intermwith penis buttoni.
70 0intermediate
•immaturewith penis CD
60
0immature E
50
40
30 ^
o
0
20
o
10 ooo x> > °»
0 8 0 8 0 t
0
3-Feb-51 25-Mar-51 14-May-51 3-lul-51 22-Aug-51 ll-Oct-51 30-Nov-51 19-lan-52 9-Mar-52
in this species. However, UCSC animals maintained in the lateral. Ifthe interactions of5-6 September 2002 are counted
laboratoryin groups have occasionallybeen found in copula. as onecopulation, then 7/11 copulationsobserved havebeen
Ariolimaxbuttoni, likeA. cohimbianiis, and unlike other spe- simultaneously reciprocal. In three cases, the termination of
ciesofAriolimax, hasboth spotted (maculate) and unspotted the copulation was observed and in all of these cases, the
(immaculate) individuals, and three ofthe copulating pairs copulation was unilateral when first observed. Ofthese, one
observed in this study involved both a maculate and an terminated 62 min after the copulation was noticed, one
immaculate individual (Fig. 5; Table 3). Copulations be- terminated 98 min after first observed and one (5-6 Septem-
tween maculate and immaculate individuals are also seen in ber 2002) terminated 7.5 h after observation resumed on the
A. cohimbianus (Cody 2006). second day; 22 h and 53 min after the pair was first seen
In the laboratory, cop>ulations have been observed from copulating. Reciprocal copulations may become unilateral
Februarythrough earlySeptember. Copulation wasobserved when one penis is withdrawn considerably earlier than the
on 11 or 12 (see below) occasions involving animals from 5 other (Leonard et al. 2002). The available data (Table 3)
different populations (Table 3). In one case, a pair of slugs show that copulations in Ariolimax buttoni are often very
from Central Marin was found copulating unilaterally at long. In 3/12 observations copulation lasted more than 7 h,
11:07 pm on 5 September 2002 (by JSP), observed until and in one case (5-6 September 2002), the interaction may
12:30 am 6 September 2002, and then left alone overnight. have lasted almost 23 h. In the unilateral copulation on 5
What was apparently the same pair of slugs, in the same February 2001, between two immaculate individuals col-
position, was found copulating when the box was next lected from the UCSF campus, when the pair were first seen,
checked at 2:30 pm on 6 September 2002 and the pair finally one slug had intromission as a male and the second slug
separated at 10 pm that evening. We cannot say with con- (acting as female) had its penis completely everted and rest-
fidencewhetherthis representsoneortwo copulations. Ifwe ing against the body ofits partner. The end ot the penis was
count it as two copulations, ofthe 12 copulations that have greatly enlarged in the form ofa broad bulb. As the second
been observed, 7 involved simultaneously reciprocal copu- slug withdrew its penis, the bulb gradually deflated and the
lation between the members of a pair whereas 5 were uni- penis involuted from the tip. In the observations of 5-6
128 AMERICAN MALACOLOGICAL BULLETIN 23 - 112 - 2007
100 hyperwithpenis Figure 4. Relationship between reproduc-
90 hypertrophied tive state (Appendix),bodyweight, and al-
intermwithpenis bumen gland length in 66 individuals of
•g 80 ointermediate Ariolimax buttoni.
^E
70 •immaturewithpenis
oimmature
? 60
50
-Ri-
V 30
E
I 20
<
10 6>8 0° ® O 0
10 20 30 40 50 60
BodyWeight (gm)
Table 3. Copulations ofAriolimax buttoni from different populations in California.
1
Population Date Time first observed Observations ended Type ofintromission
UCSF 2/5/2001 16:05 17:43 Unilateral
Mendocino 4/10/2001 17:45 17:46 Simultaneously reciprocal
Mendocino 5/3/2001 17:30 20:29 Simultaneously reciprocal
Mendocino 5/17/2001 21:05 22:23 Simultaneously reciprocal
Central Marin 711712002 13:15 15:17 Simultaneously reciprocal
Central Marin 9/5/2002 23:07 24:30 Unilateral, spotted male
Central Marin 9/6/2002 14:30 22:00 Unilateral, spotted male
UCSF 4/8/2003 18:31 26:01 Simultaneously reciprocal
UCSF 5/5/2003 11:52 23:35 Simultaneously reciprocal
Tomales Bay 5/6/2003 16:04 18:28 Unilateral
Tomales Bay 6/19/2003 15:45 16:47 Unilateral
Tomales Bay 8/8/2003 13:44 18:30 Simultaneously reciprocal
September 2002 one slug was seen to contact the penis with studyhadapenis. In allcasestheaphallateslugswerelacking
its mouth on several occasions but there was no evidence of not only a penis but all penial musculature, and the vas
chewing on the penis. deferens was connected to tissue at or near the atrium (Lig.
1), terminating in a bulb in many cases, which is not what
one would expect from apophallation (see discussion by
DISCUSSION Roth 2004). Further evidence that aphallyin this population
is not derived from apophallation during copulation comes
Phally Polymorphism from the observation that none of a series of sexually im-
Since Heath’s (1916) description of apophallation in mature laboratory-reared slugs from Mills College, which
Ariolimax califomicus, the tendency has been to explain ab- were up to 4 months old, showed any signs ofdevelopment
sence ofa penis in this genus in thisway (Mead 1942, 1943). ofa penis. This led to the hypothesis that aphally occurred
However, Heath himselfexpressed doubt ofthis interpreta- naturally in some individuals ofA. buttoni, and that an in-
tion (cited in Mead 1943, see above). The current studywas nate phally polymorphism rather than apophallation was
stimulated bythe observation that all 27 individuals ofAn'o- largely responsible for the presence or absence ofa penis in
Iwiaxcollected on the Mills College campus for an anatomi- this and perhaps other species ofAriolimax.
cal study lacked a penis (Pauli 1951). This is a higher inci- This hypothesis was confirmed by the observation that
dence of aphally than would be expected from the 5% in individuals derived from the Staten Island population,
apophally rate reported by Heath (1916) for A. califoniiais. aphally was found even in two individuals that were reared
Only 7 of67 slugs collected from MillsCollege in thecurrent in isolation from the egg to the age of 30 months and had
PHALLY POLYMORPHISM IN ARIOLIMAX BUTTONI 129
both laid eggs. Aphally in these individuals could not be due alsderived from populations nearTomales Baywere euphal-
to either apophallation or sexual immaturity. Moreover, the late (Table 2).
anatomy of these aphallic isolates was consistent both with The results presented here, then, demonstrate that
that of other Staten Island individuals that were group- phally polymorphism is present in Ariolimax buttoni, show-
housed as adults and with that ofthe Mills College animals. ing that only 7/67 Mills College and 4/16 UCSC, including
The aphallates are characterized by either the complete ab- 0/10 Staten Island, slugs had a penis, and leave open the
sence ofa penis (Fig. 1) or reduction ofthe penis to a small question of whether apophallation occurs in this species.
stub (Fig. 2A); in both cases, the penial retractor muscle is Aphallate individuals have also been found in a population
absent andthevas deferens endsblindlyin a mesenteryor in oiAriolimax (Meadarion) brachyphallus from Hillsborough,
the neighborhood of the atrium. The description of dissec- San Mateo City, CA (Pearse, impubl. data). Wright (1938)
tion of a freshly apophallate A. dolichophallus by Mead stated that 415 individuals of Ariolimax californicus he ex-
(1942) suggests, however, that the distinctions between amined did not have a penis, or other “male parts” whereas
aphallyas described in Fig. 1 and 2, and a healed apophallate 248 individuals were found to have normal, “or regenerat-
individual, are subtle. ing” penes. Wright also stated that the aphallic individuals
The anatomy ofAriolimax biittoiii, as described by Pils- werethe result ofapophallation but provided no evidence to
bry and Vanatta (1896), and the aphallate individuals from support this, nor did he provide details of the anatomy or
Mills College described here (Fig. 1, also Pauli 1951) and the even information as to the source of these specimens.
Staten Island animals (Fig. 2A), is very consistent with the Wright’s observation suggests that an anatomical study of
descriptions and illustrations ofaphallic individuals in other thatspeciesisneededto determinewhetheraphallyaswell as
stylommatophorans (Watson 1934, Riedel 1955, Tompa apophallation occurs. It is possible that phally polymor-
1984, Pokryszko 1987, see discussion in Roth 2004). Phally phism will be found to be more widespread in Ariolimax.
polymorphism, in which a hermaphroditic population or The adaptive significance ofphallypolymorphism is not
species consists ofa mixture ofindividuals with normal pe- entirely clear. Local Mate Competition theory (Charnov
nes (euphallic individuals) and individualswith either mark- 1982) predicts that allocation to male function in hermaph-
edly reduced penes (hemiphallic individuals) or no penis at rodites should be reduced where few sexual partners are
all (aphallic individuals) is widespread in the Pulmonata and available, suggesting that aphally should be more common
has evolved many times (see discussion in Duncan 1975, where self-fertilization is common or population densities
Tompa 1984, Pokryszko 1990, Lace 1992, Schrag and Read are low. Baur et al. (1993) hypothesized that aphally evolves
1992, Viard et al. 1997, Doums et nl. 1998, Backeljau et al. in populations that typically self-fertilize. In both cases,
2001). The results presented in this study offer a clear dem- aphally is predicted to be associated with a capacity for uni-
onstration of phally polymorphism in A. Inittoni. parental reproduction and the role ofenvironmental factors
Aphally seems to be common in at least some popula- in influencing the ratio of aphallic to euphallic offspring is
tions ofAriolimax buttoni. Pilsbry and Vanatta (1896, dis- hypothesized to be an adaptation to colonizing new habitats
cussion in Pilsbry 1948) examined a large collection of in- (Schrag and Read 1992, but see Baur et al. 1993)
dividuals lacking a penis from Oakland, and Heath (Mead
1942, cited above) dissected 400 aphallate specimens from Uniparental Reproduction
Hog Island in Tomales Bay. Westfall (1952) also reported Uniparental reproduction, either by self-fertilization, as
that of85 specimens collected at Mills College in the spring has been widely assumed, and well-documented in some
of 1952 (subsequent to the work described here), only 9 had cases, or by apomixis, as suggested by some authors (Mc-
a penis, ofwhich 3 were in the hypertrophied state and 6 in Cracken and Selander 1980, Foltz et al. 1982a, Hoffmann
the immature reproductive state, in the terminology used 1983) is widespread, although not universal in stylommato-
here. It is characteristic of phally polymorphisms that the phorans (Tompa 1984, Heller 2001). Within a genus, some
percentage ofa given morph varies widely from one area to species may be obligate outcrossers whereas others readily
another and from season to season (Lace 1992). Baur and self-fertilize (Foltz et al. 1982b, Reise 2002). Mead (1942)
Chen (1993) found that thefrequencyofaphallic individuals rearedAriolimaxin isolation foras longas two years without
in populations of Chrondrina avenacea (Bruguiere) varied the production of any eggs but considered the question of
from 0.9% to 89.2% in thevicinityofBasel, Switzerland. The uniparental reproduction in Ariolimax still open. Here we
tendency for the frequency of phally morphs to vary with report thatA. buttoni can produce viable offspringwith high
environmental conditions {e.g., Watson 1934, Schrag and rates ofhatching success without cross-fertilization. Unipa-
Read 1992, Baur et al. 1993) may explain why Mead (1942, rental reproduction has also been observed in A. dolicho-
1943) found only euphallate individuals in the vicinity of phallus but with low hatching success (Miller and Sinervo
Hog Island whereas Heath had found only aphallate indi- 2007). Westfall (1952) reported that in Mills College ani-
viduals at Hog Island earlier. In this study, all four individu- mals, the development of the ovotestis did not differ be-
130 AMERICAN MALACOLOGICAL BULLETIN 23 • 1/2 • 2007
tween euphallic and aphallic individuals; that is the qualita- and Mills College animals held in the laboratory (Westfall
tive degree of sperm vs. egg production did not appear to 1952, 1960) laid eggs in the fall and winter as suggested for
depend on phally status. The production ofsperm obseiwed Ariolimax by Mead (1942). Mills College animals with red-
in aphallate individuals in histological studies (Westfall, dish-brown ovotestes, a spermatheca filled with reddish
1952), suggests that self-fei'tilization could occur in A. Init- fluid, and a yellow albumen gland larger than the ovotestis
toni. Similar results have been reported in the genus Zoni- (the hypertrophied reproductive state), believed to represent
toides Lehman, 1862 (Watson 1934). Resolution ofthe ques- the egg-laying stage, were first seen in a dissection of 22
tion of whether these offspring are the result of selfmg or October 1951 and then seen regularly until February 1952
apomixis will rec]uire genetic analysis. (Fig. 3). Egg-laying and juvenile growth from individuals in
the Mills College portion of the study have been reported
Sexual Behavior elsewhere (Westfall 1960). Westfall (1960) reported a mini-
Of the observations reported in Table 3, about half of mum time to hatching of 23 days and a maximum of 2
the pairs had reciprocal intromissions and the other halthad months for Mills College slugs, which brackets the 47-54
unilateral intromissions, in which one slug was acting as a days found for UCSC animals (Leonard et al. unpublished).
female and the other as a male. Copulation lasted more than In A. dolichophallus, hatching time ranged from 51-55 days
7 h after being first observed in 3 of 12 observations, and in and in A. californicus from 46-81 days (Leonard etal. 2002).
one case (5-6 September 2002) the interaction may have These hatching times all reflect laboratory conditions and
lasted almost 23 h after detection (see above; Table 3). Thus, will probably depend strongly on temperature, making it
copulation in Ariolimax biittoni often lasts more than the likely that hatching times in the field will be somewhat
two hours typical ofsimultaneously reciprocal intromissions longer. In the spring and summer, animals were found to be
ofA. dolicliophaUus and much longer than the brief, unilat- in the immature and intermediate reproductive states,
eral intromissions ofA. califorincus (Leonard etal. 2002). In whereas in fall andwinterall three stateswere found (Fig. 3).
longreciprocal copulations such as those ofA. dolicJiophallus This is consistent with data from Pauli (1951) who noticed
(Leonard et al. 2002) it is not unusual for one individual to that the reproductive system was more often in the imma-
withdraw the penis long before the other (Leonard, unpub- ture state from February to April, whereas hypertrophy of
lished observation). Therefore, it seems likely that copula- the reproductive tract was predominant in A. buttoni col-
tions between euphallate A. buttoni (Figs. 5A-C) are nor- lected from September to the middle of February. In field
mally long and simultaneously reciprocal. The unilateral observations, juveniles of A. buttoni weighing less than 1
intromissions observed here may have been the end of si- gram appeared in lanuaiy (Pearson et al. 2006).
multaneously reciprocal copulations or they may have in- Since laboratory data demonstrate that Ariolimax but-
volved copulationwith apartnerthatlacked apenis (below). toni can live more than one year, with some individuals
laying eggs at less than one year of age while some clutch
Life History mates reared underthesameconditionsdo notlayeggsuntil
The life histories of Ariolimax spp. are poorly known. over two years ofage (see below), it seems likely that not all
The observations reported here provide the most detailed individuals will reproduce in their first year, perhaps ac-
picture available on sexual development and the reproduc- counting for the occurrence ofindividuals in the immature
tive cycle for anyspecies ot'Ariolimax. We found that, in the and intermediate reproductive states throughout the year in
laboratory,Ariolimaxbuttonimaylivemorethan 30 months. this study. Protiindrous development ofthe ovotestis is typi-
This is consistentwith previousreportsthat individualsofA. cal of stylommatophorans (Tompa 1984) and has been re-
californicus and A. dolicliophaUus, collected as adults, sur- ported for A. californicus (Gottfried and Dorfman 1970),
vived more than 18 months in the laboratory (Leonard etal. makingitseem probable thatdevelopment inA. buttonimay
2002). The anatomical and histological data obtained in the involve a progression from the immature reproductive state
Mills College study indicate that, in Oakland populations, through the intermediate male state to the hypertrophied
individuals may live more than a year in the field, since two female state, and then the old reproductive state, although
classes of individual were found in fall and winter dissec- the intermediate state may be heterogeneous. The interme-
tions; spring and summer specimens showed a gradual in- diate reproductive state is the state characterized by massive
creasein albumen glandlength, suggestingthe maturation of sperm production (Westfall 1952; Appendix) andonewould
a single cohort (Fig. 3). This fitswellwith fielddata (Pearson expect it to be associated with copulation. Our observations
etal. 2006) for identifiable individuals in a population ofA. suggest that in A. buttoni may become sexually mature and
buttoni in Orinda, Contra Costa Cty, CA that showed a life begin copulation as early as the summer of their first year
span ofapprox. 2 years. and layeggs that same fall. The first specimen found to have
mm
The data also provide a clearpictureofthe phenologyof an albumen gland greater than 20 in length in the Mills
Ariolimax buttoni. Both UCSC (Leonard et al. unpublished) College study was dissected in April 1951 (Fig. 3). The first