Page 9 - Scani_et_al1995

EVOLUTION OF HYBRID STICK-INSECTS 67
Downloaded by [31.185.101.124] at 08:47 23 March 2016 likely short-life paradigms do not apply to Bacillus. Further- jectural hypotheses, since, on the issue, abundant ex-
more, in several other hybrid complexes a reticulate
evolutionary pattern is becoming apparent: in the unisexual
ostracod Cyprinotus (Turgeon & Hebert, 1993), the
hybridogenetic/gynogenetic italies and Poecilia complexes
(Haskins et al, I960; Quattro et al., 1992a, b), the
gynogenetic Ambystoma group (Hedges et al., 1992) and
the parthenogenetic lizards of the genera Lacerta and
Cnemidophorus (Vyias et al, 1990; Moritz et al, 1992).
The observation that the occurrence of similar interac-
tions between unisexuals and bisexual relatives are being
discovered from very different animal groups, such as
crustaceans, insects and vertebrates, argues for much more
complex breeding behaviour and evolutionary relation-
ships than is currently accepted for hybrids.
STABILIZED HEMICLONES AND SHIFT TO CLONAL
SPECIES
Bacillus hybridogenetic strains are the only known in-
vertebrate hemiclonals, the two additional instances
being found among vertebrates, namely the already men-
tioned Poeciliopsis and the Rana esculenta (= ridi-
bunda-lessonae) complexes (Schultz, 1961, 1989; Uzzell
et al, 1980; Dawley & Bogart, 1989). Hybridogens
usually co-occur with only one parental species, the one
supplying sperm to replace the genome discarded during
hybrid gametogenesis, i.e., R. lessonae in the L-E system
of R. esculenta and P. lucida in the P. monacha-lucida
strains. Bacillus hybridogens experience two slightly dif-
ferent conditions: the southeastern B. rossius-g.grandit
is syntopic only with B. g. grandit, thus conforming to
the general situation, while the northwestern B.
rossius-g. benazzii co-occurs with both parental taxa.
However, its different ecological and reproductive con-
ditions do not affect the breeding system, since the syn-
topic B. rossius are all females and the rare
hybridogenetically originated B. rossius-g. benazzii
males are sterile; therefore, in the northwestern
hybridogenetic strains, only females interact with B. g.
benazzii males, in a quite comparable way as the
southeastern hybridogens do with B. g. grandii males.
The above-outlined reproductive conditions clearly
suggest that stabilized hemiclonal systems can be attained
only if reproductive isolation from the maternal ancestor
is achieved. This seems a critical point for hemiclonal
hybrid persistence. At the very beginning of their
production, such hybrids must co-occur with both
parental species, their syntopism being necessary for the
hybridogenetic strain origin. Given no pre-copula
barriers nor mating preferences, hybridogens will mate
with both parental species. Depending on their relative
frequency, a proportion of «hybridogen x maternal
parent» matings will occur and produce a few pure mater-
nal species individuals; they would re-enter the maternal
ancestor gene pool, thus representing a destabilizing fac-
tor for the hemiclonal system. These are not mere con-
perimental evidence is available for all three
hybridogenetic systems (Berger, 1967, 1968; Leslie &
Vrijenhoek, 1978, 1980; Graf & Polls Pelaz, 1989;
Vrijenhoek, 1989, 1993; Schmidt, 1993; Tinti et al.,
1994). Stick-insect and frog data are consistent in in-
dicating that hemiclonal genomes are still well coadapted
with haplosets derived from nowaday bisexuals of the
maternal species, i.e., viable and fertile synthetic
specimens of B. rossius and R. ridibunda can be ob-
tained from B. rossius-grandii x B. rossius and R.
esculenta x R. ridibunda crosses, respectively. On the
other hand, when P. monacha-lucida x P. monacha
crosses are analyzed, a strong reduction of viability and
almost complete sterility of offspring result, mainly
related to a genetic «deterioration» of the hemiclonal
genome. Owing to the different outcomes attained when
the hemiclonal system breaks down through the produc-
tion of homospecific hybridogenetically-derived
specimens, we could conclude that, at present, the
isolation from the maternal ancestor appears to be more
important for stick - insect and frog hybridogenetic
balance than for the fish hemiclonal system.
Hybridogenetic complexes are maintained by a very
dynamic system of necessary interactions - with
avoidance of the destabilizing ones - between parental
taxa and hybridogens. The Bacillus hemiclonal system
seems to be particularly well buffered owing to the ex-
traordinary ability of producing host species an-
drogenetics from hybridogens, certainly strengthening
the homeostasis of the system.
Acertain degree of plasticity in the cytological properties
of the hemiclonal gametogenesis allows further
microevolutionary considerations on the dynamics of
hybridogenetic strains. An increased production of clonal
daughters - actually noticed as a very rare event among the
northwestern hybridogenetic strains - and their release
from sperm need, would allow the spreading of the new
clonal females into a larger area. On the basis of com-
parative reproductive cytology and chromosomal repatter-
ning distribution, this evolutionary pathway has been
found to be very likely for the clonal allodiploid par-
thenogen B. whitei, which, from its hybridization focus in
the centre of the southeastern range of hybridogens, has
spread into the whole Iblean area and the Catania plain to
the north (Manaresi et al, 1992a, b; Mantovani & Scali,
1992; Mantovani et al, 1992; Scali et al, 1992; Tinti &
Scali, 1992a; Tinti et al, 1994). It could also well be that
similar microevolutionary steps were experienced by B.
lynceorum: the initial allodiploid hybrid could have
switched to parthenogenesis following the third genome
addition, that allowed its spreading from the initial very
limited area to the presentday range (Manaresi et al, 1993).
Several gynogenetic or parthenogenetic vertebrate hybrids,
that include both diploid and polyploid taxa, may have had
similar evolutionary histories, of which at present only the
general traits can be traced (Uzzell, 1963, 1964; Schultz,
1967, 1969; Uzzell & Goldblatt, 1967; Darevsky &