The existence of fertile hybrids of closely related model earthworm species, Eisenia andrei and E. fetida

Lumbricid earthworms Eisenia andrei (Ea) and E. fetida (Ef) are simultaneous hermaphrodites with reciprocal insemination capable of self-fertilization while the existence of hybridization of these two species was still debatable. During the present investigation fertile hybrids of Ea and Ef were detected. Virgin specimens of Ea and Ef were laboratory crossed (Ea+Ef) and their progeny was doubly identified. 1 –identified by species-specific maternally derived haploid mitochondrial DNA sequences of the COI gene being either ‘a’ for worms hatched from Ea ova or ‘f’ for worms hatched from Ef ova. 2 –identified by the diploid maternal/paternal nuclear DNA sequences of 28s rRNA gene being either ‘AA’ for Ea, ‘FF’ for Ef, or AF/FA for their hybrids derived either from the ‘aA’ or ‘fF’ ova, respectively. Among offspring of Ea+Ef pairs in F1 generation there were mainly aAA and fFF earthworms resulted from the facilitated self-fertilization and some aAF hybrids from aA ova but none fFA hybrids from fF ova. In F2 generation resulting from aAF hybrids mated with aAA a new generations of aAA and aAF hybrids were noticed, while aAF hybrids mated with fFF gave fFF and both aAF and fFA hybrids. Hybrids intercrossed together produced plenty of cocoons but no hatchlings independently whether aAF+aAF or aAF+fFA were mated. These results indicated that Ea and Ef species, easy to maintain in laboratory and commonly used as convenient models in biomedicine and ecotoxicology, may also serve in studies on molecular basis of interspecific barriers and mechanisms of introgression and speciation. Hypothetically, their asymmetrical hybridization can be modified by some external factors.


Introduction
The composting earthworms belonging to the genus Eisenia (Annelida, Oligochaeta, Lumbricidae) are easy to maintain under laboratory conditions and exhibit a relatively simple body a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Experimental design
The aim of proof-of-concept preliminary experiments was to compare the cocoon production and hatchability in the laboratory formed intra-specific pairs Ea+Ea and Ef+Ef, and inter-specific pairs (Ea+Ef).
Since viable hatchlings were obtained in all three groups of experimental couples of earthworms, a possibility of hybridization versus self-fertilization in Ea+Ef pairs of earthworms was tested. Analyzes of species-specific sequences of the haploid mitochondrial marker COI (marked by letters 'a' or 'f') showing maternity and of the maternal/paternal nuclear marker 28S rRNA gene (marked by capital letters 'A' or 'F') were performed. Self-fertilized specimens of Ea and Ef must exhibit aAA or fFF markers, respectively, while hybrids of Ea+Ef shall possess two easily identified AF/FA sequences, thus they shall be either aAF or fFA, with the first nuclear marker consistent with maternity (Fig 1).
The DNA samples were obtained from the tail tips of earthworms from the controlled laboratory mating of Ea+Ef pairs and their progeny (potentially aAA, fFF, aAF, or fFA). Photos were also performed to investigate inheritance of pigmentation patterns.

Experimental scheme
Proof-of-concept preliminary experiments. To avoid copulation and uncontrolled sperm exchange, very small (<0.05 g b.w.) freshly hatched specimens were collected from Ea and Ef colonies of genotyped earthworms. On December 15, a culture of 15 pairs of virgin worms was started till sexual maturation and reproduction, forming the control intra-specific couples, 4 (Ea+Ea) and 4 (Ef+Ef), and 7 experimental inter-specific (Ea+Ef) pairs. Progeny of intra-specific mating served as a source of individuals of pure aAA and fFF genotypes for further studies, while several generations of progeny of inter-specific couples were used for detection of hybrid specimens.
Eight weeks into the experiment, both the control pairs and experimental worms were transferred to new soil and their cocoons produced over the two consecutive weeks (9 th and 10 th ) were counted and kept in separate boxes for monitoring their viability/hatchability during the 5 consecutive weeks. Hatchlings were used for further mating.
Progeny of (Ea+Ef) couples. Freshly hatched offspring of the (Ea+Ef) couples were associated in pairs, each in a separate box. Pairs of virgin earthworms were joined randomly since identity of such neonate earthworms is impossible to establish by the phenotype. This distribution hypothetically led to the formation of either intra-specific couples (aAA+aAA; fFF+fFF), or hybrid-hybrid couples (aAF+fFA; aAF+aAF; fFA+fFA), or back-crosses of hybrids with Ea or Ef (aAF+aAA or aAF+fFF; fFA+aAA or fFA+fFF). After several weeks, the pairs produced cocoons, some of which were viable, and the procedure was repeated. In order to prevent stress in the juvenile earthworms, which could be potentially disruptive factor affecting their maturation/reproduction, the genetic identification of each particular earthworm was established only during its adulthood, when cocoon production was well established (at body weight >0.5 g).
For the latter purpose, some adults (with special attention to those exhibiting atypical pigmentation patterns), were settled individually in boxes on wet filter papers. After overnight depuration, their 4-6 posterior segments were amputated and ethanol-fixed for DNA analysis while earthworms returned to soil for further reproduction. Tail tips of Eisenia sp. easily regenerate [30] thus-when necessary-tails were amputated again from the same specimens.
Pigmentation patterns were documented by photography with the DSL camera (Sony SLT-A58).

DNA extraction, PCR amplification and sequencing
The terminal segments of the earthworms were fixed in 80% ethanol. The tissues were hydrated in Tris-EDTA (TE) buffer (3×10 min); then, total genomic DNA was extracted using the SHERLOCK 1 extracting kit (A&A Biotechnology). The final product of purification was resuspended in 20μl of TE buffer. The extracted DNA was then stored at − 80˚C.
Two molecular markers were used: cytochrome oxidase subunit I (COI) for the mitochondrial DNA (mtDNA) and the 28S subunit of the nuclear ribosomal RNA (28S) for the nuclear DNA (ncDNA). Composting earthworms Ea and Ef are cross-fertilizing simultaneous hermaphrodites capable also to self-fertilization [29]. A few days after copulation and/or self-insemination each specimen produces cocoons containing its own eggs that are fertilized within cocoon by spermatozoa stored previously in spermatheca. Ova might be fertilized either by spermatozoa of the same specimen or cross-fertilized by spermatozoa coming from the partner.  To check the quality of the PCR products, 4 μl of reaction product were run on 1% agarose gel. PCR products were purified using Clean-Up 1 columns (A&A Biotechnology). The purified PCR products were sequenced in both directions using BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) following the manufacturer's protocol and using the primers described above. After removing the terminators with ExTerminator 1 columns (A&A Biotechnology) the sequencing reaction products were separated on an ABI PRISM13100 Avant Genetic Analyzer in commercial laboratory Genomed S.A. (Warszawa, Poland).
All the sequences were declared in the GenBank (MG030809-MG030998 for 28S and MG030999-MG031156 for COI).

Results of proof-of-concept preliminary experiments
Features of cocoons produced during the 9 th and the 10 th weeks by 15 pairs of earthworms put to boxes as pairs of neonate virgin specimens, i.e. 4(Ea+Ea), 4(Ef+Ef), and 7(Ea+Ef), were followed for further 5 weeks and the results are shown in Table 1.
Numbers of cocoons laid by one worm per week were similar in all groups of worms, being lowest in Ea+Ea (2.9+0.43), medium in Ef+Ef (3.4+0.68) and highest in boxes with inter-specific couples Ea+Ef (4.3+0.42). However, numbers of hatchlings per earthworm per week were in opposite sequence, being highest in Ea boxes (5.6+0.39), medium in Ef (2.8+1.05) and Table 1. Reproductive success of earthworms kept as intra-specific pairs of E. andrei (Ea+Ea) and E. fetida (Ef +Ef), or inter-specific pairs (Ea+Ef) measured by viability and hatchability of cocoons laid during the 9 th and 10 th weeks after joining of pairs of small virgin specimens.

Reproduction 4 pairs (Ea+Ea) 4 pairs (Ef+Ef) 7 pairs (Ea+Ef)
Pairs Fertile hybrids of Eisenia andrei and E. fetida earthworms lowest in (Ea+Ef) boxes (1.0+0.64). Moreover, hatchlings were present in each of four Ea and four Ef boxes, while 3 of 7 boxes with inter-species pairs contained only sterile cocoons, without any hatchlings. The percentages of sterile cocoons was low (7%) in (Ea+Ea), quite high (48%) in (Ef+Ef), but very high (77%) in (Ea+Ef) groups. In general, reproductive output was most efficient in E. andrei, lower in E. fetida, whilst it was very low in inter-specific pairings (Table 1). For present purposes it was important to check whether hatchlings from partners of inter-specific pair have resulted from self-fertilization or cross-fertilization (hybridization).

Earthworm genotyping and genealogy considerations
Sequence characterization. A possibility of hybridization in couples of Ea and Ef was tested by analyzing species-specific sequences of the mitochondrial marker-cytochrome oxidase subunit I (COI) ('a' or 'f') showing maternity and the maternal/paternal nuclear marker 28S subunit of the nuclear ribosomal RNA (28S) ('A' or 'F'). Ea and Ef are characterized by genotypes aAA and fFF, respectively. Hybrids of Ea+Ef possess two easily identified 28S sequences, AF/FA (Fig 2).
Labeling of investigated samples/specimens. Each investigated specimen was described by combining the mitochondrial and nuclear markers and its numerical code (reflecting its sequence in our analyses) as e.g. aAA45 (for Ea), fFF67 (for Ef), and aAF93 (for a hybrid from Ea ovum fertilized in Ea cocoon by Ef sperm), or fFA149 (for a hybrid from Ef ovum fertilized in Ef cocoon by Ea sperm). In a case of nuclear markers, either 'AF' or 'FA', on the first place there is the marker of maternal origin (i.e. the same origin as mitochondrial one). Amputation of easily regenerating posterior segments is not harmful for Eisenia specimens [30] and-if necessary-might be performed repeatedly; that was reflected by two or three numbers for several individuals, e.g. fFA149/194 or fFA143/159/191. Phylogenetic relatedness. The phylogeny analyses indicated a high level of genetic differentiation between E. andrei and E. fetida. The p-distance between these two species were 0.150 for COI and 0.017 for 28S. For COI marker, intraspecies variation has also been found. The COI sequences for E. fetida formed two distinct subclades, 'f' and 'f2' separated with p-distance = 0.114. Two COI subclades were also present for E. andrei, but p-distance was negligible (Fig 3). Fig 3 shows the maximum-likelihood phylogram for the mitochondrial COI gene of 158 specimens of Eisenia fetida/andrei combined with the nuclear 28S rRNA genes of the same coded individuals. This analysis showed three distinct clusters; one cluster of 96 individuals including exclusively mitochondrial 'a' COI sequences characteristic for E. andrei ova, but among them there are 70 'aAA' sequences typical for E. andrei and 26 sequences 'aAF' characteristic for hybrids. The two remaining clusters consisted of 9 and 53 individuals from E. fetida ova, 'f2' and 'f', respectively, the former with only E. fetida nuclear sequences ('f2FF'), and the later, with 49 'fFF' and 4 'fFA' sequences characteristic for hybrids.

Earthworm phenotypes
The photos from the inset of Fig 3 demonstrate that among investigated earthworms there were specimens of uniformly red pigmentation typical for E. andrei (e.g. aAA 45 ), specimens with striped red-yellow pigmentation patterns typical for E. fetida (e.g. fFF 47

Genealogy of hybrid descendants of E. andrei and E. fetida couples
The analysis of genotyped specimens from the controlled mating experiments allowed us to trace their genealogy. Pure aAA and fFF specimens were most common among 158 genotyped worms, thus only closest relatives of 30 hybrid earthworms (26 specimens from E. andrei ova, i.e. aAF, and 4 specimens from E. fetida ova, i.e. fFA) detected among descendants of the initial couples of E. andrei (aAA) with E. fetida (fFF) are presented in Table 2.
In family 1, among progeny of the parents aAA25+fFF26 ten F1 specimens were genotyped, including seven Ea (aAA), one Ef (fFF), and two aAF hybrids (family 1a). The first pair of virgin specimens randomly taken from this family, identified at their adulthood as [aAA45+aAF46] formed the family 1b with two aAA specimens and two aAF hybrids among the F2 offspring. The second pair of virgin specimens randomly taken from the family 1a represented pure Ef and Ea genotypes [AA62+fFF61]. Therefore, although they belonged formally to F1 generation, we have marked them as parental generation and ten of their genotyped offspring as F1. Among them we identified five aAA and five aAF (family 1c). From them three backcrossed pairs of aAF hybrids with pure aAA earthworms were randomly generated. All of them giving viable F2 generation; i: [aAF76/137+aAA83/138] with two hybrids and five aAA offspring (family 1d); ii: [aAF77/142+ aAA87] giving two aAA offspring (family 1e) and iii: [aAA97/153 and aAF98/ 152] giving one hybrid and four aAA among F2 offspring (family 1f in Table 2).
In family 3, among F1 progeny of the aAA31+fFF32 couple there were two aAA specimens, one fFF, and four aAF hybrids (family 3a). Then, three pairs of F1 offspring were intercrossed.  Table 2).
In family 4, the couple aAA33+fFF34 gave two fFF and three aAF hybrids among F1 offspring (family 4a). Intercrossing of F1 specimens [aAF68+fFF67] gave three Ef (fFF) specimens and two fFA hybrids in the F2 generation (family 4b). Each of the F2 hybrids demonstrating Ef specific COI mitochondrial markers, fFA149/194 and fFA162 (underlined in family 4b), was paired with the hybrid demonstrating COI characteristic for Ea, i.e. aAF86/141 (underlined in support is given. All sequences were deposited in the GenBank (MG030809-MG030998 for 28S and MG030999-MG031156 for COI). Inset: Phenotypes of Eisenia andrei (aAA45), E. fetida (fFF47) and the hybrid (aAF46). Regeneration blastemas of the posterior segments pointed out by the yellow arrows.

Table 2. Relationships between descendants of E. andrei (aAA) and E. fetida (fFF) couples with special attention to aAF (bold) and fFA (italic bold) hybrids genotyped for mitochondrial COI and nuclear 28S markers
In general, among 26 aAF hybrids detected during present investigations, the majority (18 specimens) derived from inter-specific couples (aAA+fFF), while 8 of the remaining aAF hybrids were offspring of the backcrosses, either (aAF+aAA) or (aAF+fFF). In contrast, no fFA hybrids were observed in F1 generations from aAA+fFF inter-specific pairs. All four of fFA hybrids derived in F2 generation from the back-crosses of the F1 aAF hybrids with fFF specimens.
Despite cocoon production, offspring was absent in hybrid-hybrid pairs, both aAF+aAF or aAF+fFA. So far, the back-crosses of fFA with fFF specimens have not given viable offspring in F3 generation (Table 2).
Speculations on possibilities of self-fertilization, cross-fertilization, and hybridization in mating pairs used in present studies. As illustrated on Fig 4A, the progeny of an interspecific couple Ea+Ef may contain aAA and fFF earthworms resulted from self-fertilization of 'aA' or 'fF' ova by spermatozoa of the same specimens. Inter-specific hybrids aAF resulted from cross-fertilization of 'aA' ova by 'F' spermatozoa of fFF parent, and vice versa, fFA hybrids from 'fF' ova fertilized by 'A' spermatozoa. Among progeny of inter-specific couples investigated in the present work we did find the aAA, fFF, and aAF earthworms while fFA hybrids were absent (exemplified by family 1a from Table 2).
The aAF hybrids obtained in the present studies gave vital and fertile offspring in backcrosses both with aAA and fFF specimens (Fig 4B and 4C, respectively).
The hybrids shall produce two types of spermatozoa, with either 'A' or 'F' nuclear markers, and also two types of ova, being either compatible with respect of mitochondrial and nuclear markers, i.e. 'aA' or 'fF', or incompatible with respect of these markers, like 'aF'or 'fA'. There are findings showing that the latter possibility is less probable in several species [33,34], thus for simplicity this was excluded from further speculations (shadowed on Fig 4B and 4C). Fig 4B showed that the two hybrids (aAF95 and aAF96/103), being the offspring of (aAF46 + aAA45) couple, may be resulted not only from self-fertilization of 'aA' ovum of the aAF46 hybrid with the 'F' spermatozoon of the same earthworm, but also by cross-fertilization of 'aA' ovum of the aAA parent with the 'F' spermatozoon of the aAF46 hybrid. The aAA109 and aAA110 earthworms might be descended either from self-or cross-fertilization of the 'aA' ova of the aAF hybrid, or self-or cross-fertilization of the 'aA' ova of the aAA earthworm by the 'A' spermatozoa of one of these partners.
Similar possibilities exist in the case of the aAF back-cross with the fFF partner ( Fig 4C). Both fFA149 and fFA162 hybrid offspring had to result from fertilization of 'fF' ova of fFF67 partner by the 'A' spermatozoa of the aAF68. However, aAF167 hybrid might have resulted from fertilization of the 'aA' ova either by the 'F' spermatozoon of the same aAF68 hybrid specimen or by the cross-fertilization of the 'aA' ovum of the hybrid with the 'F' spermatozoon of the FF67 partner. Also fFF148, fFF150, and fFF151 offspring might have resulted from either self-or cross-fertilization of 'fF' ova of the fFF67worm (Fig 4C).
In conclusion, in the case of aAF hybrids being the F1 offspring of Ea+Ef pair, the 'aA' ovum of aAA partner had to be fertilized by the 'F' spermatozoon from the fFF partner.

Fig 4. Speculations on inheritance of mitochondrial ('a', 'f') and nuclear ('A', 'F') markers by self-and cross-fertilization of 'aA' and 'fF'
However, analysis of COI and 28S genes performed in the present investigations was not sufficient for precise identification of parents of aAF hybrids in the progeny of backcrosses of aAF with aAA or fFF worms (Fig 4). In such a case studies of highly polymorphic microsatellite loci [35][36][37] shall give conclusive explanation. E.g. this method was sensitive enough to analyze multiple paternity in Hormogaster elisae, when one mother produced descendants from more than one partner [38,39].

Discussion
The better reproductive performance of E. andrei than E. fetida demonstrated in earthworms from the present experiment was also described by other scientists who investigated these species derived from other localities [18,19,40]. Earthworm reproductive performance changes in annual cycle, e.g. maximum mating activity of the field population of E. fetida was achieved in spring [41]. The drastic differences in percentages of sterile cocoons between E. andrei and E. fetida (7% and 48%, respectively) from present experiments might reflect different reproduction dynamics of these species that, hypothetically, might be affected by the actual length of their laboratory 'domestication'. Bimodal reproduction with spring and autumn maxima of cocoon laying/hatching and high sterility rate of inter-seasonally laid cocoons was described in another species, Eisenia lucens [42].
Reproductive performance of the laboratory-formed inter-specific Ea+Ef pairs were investigated in present studies on earthworms from France, by [18] on earthworms from South Africa and France, and by [19] on earthworms from Spain and Brasil. The results of all these studies revealed extensive cocoon production by Ea+Ef pairs. However, only the inter-specific pairs of earthworms from France produced viable cocoons and hatchlings. Future experiments shall address comparative studies on hybridization of earthworms cultured in various laboratories and/or from various natural populations.

Presence of hybrids between E.andrei and E. fetida
The earthworms from Eisenia sp. investigated during present studies formed three distinct clusters on phylogenetic tree based of haploid mitochondrial COI genes of maternal origin, one including exclusively 'a' sequences of E. andrei, and two including 'f' and 'f2' sequences of E. fetida. These results fit perfectly to data obtained by the international group of experts EBI ('Eisenia Barcoding Initiative', EBI) who analyzed mitochondrial COI sequences of the coded fixed specimens provided from 28 laboratories from 15 countries [16]. This analysis revealed three distinct haplotype clusters; one including exclusively E. andrei sequences, and two with E.fetida sequences, the latter being hypothetically two different cryptic species, E. fetida 1 and E. fetida 2. Two very divergent haplotypes of E. fetida COI sequences were detected also during other studies [13,14,30], thus this aspect is worth of further investigation. Remarkably, specimens of the molecular E. fetida clusters from EBI studies based on COI genes were always identified morphologically as E. fetida. However, this was not true the other way round, i.e., some specimens of the molecular E. andrei cluster were identified morphologically as E. fetida by the EBI group [16]. However, our results suggest that at least some of E. fetida-like specimens among E. andrei cluster might be rather interspecific hybrids 'aAF'.
Recently it turned out that the COI cluster of E. andrei also contains six distinct lineages, irrespective of the striped or uniform red coloration [17]. Hypothetically, such a discordance between coloration and molecular delimitation based on COI genes might be at least partly connected with hybridization in the natural field populations of E. andrei with E. fetida.
The novelty of the present studies is that each earthworm from three clusters of phylogenetic tree based on COI gene sequences of maternal origin is characterized also by its diploid maternal/paternal 28S genes, being either 'AA' or 'FF' for E. andrei and E. fetida, respectively, or 'AF/FA' for hybrids. It turned out that within the E. andrei COI-clade there are 70 aAA earthworms of E. andrei species and 26 of interspecific aAF hybrids. One of the two E. fetida COI-classed contains exclusively nine fFF specimens, but within the second COI-clade there are 49 fFF E. fetida specimens and four fFA hybrids. For the first time these results give a formal proof of the existence of viable interspecific hybrids between E. andrei and E. fetida. As reviewed by [19], the existence of such hybrids was suggested by some investigators but negated by others.
Sheppard [22] found hatchlings in crosses between E. andrei and E. fetida and stated that the offspring could be the result of hybridization, self-insemination or 'facilitated self-fertilization' connected with sperm mixing. The results of our present experiments and molecular analyses fully confirmed his suppositions as we have proved that among F1 offspring of aAA and fFF pairs there are both the earthworms of the pure species (aAA and fFF) derived from self-fertilization and the hybrids aAF derived from cross-fertilization. Self-fertilization of individually cultured virgin specimens of E. andrei and E.fetida was recorded as being quite common in earthworms from Spain [29] but such phenomenon is very seldom in Eisenia sp. from France (in progress). For this reason we assume that appearance of aAA or fFF offspring of (Ea +Ef) pairs is a result of self-fertilization facilitated by the partner from closely related species by mixing sperm of the both copulating earthworms. Other studies will be addressed to quantify incidences of facilitated self-fertilization and hybridization in (Ea+Ef) pairs.

Hybrid fertility and evolutionary implications
The next question addressed in the present studies was whether interspecific hybrids are fertile. To answer this question, the virgin hybrid worms were mated with other hybrids or backcrossed with each of the parental species. The results obtained so far evidenced clearly that at least aAF hybrids are fertile and gave progeny with the proper partner (Fig 5).
In present experiments, among offspring of Ea+Ef pairs there were aAA and fFF worms and aAF hybrids but none fFA hybrids. The latter appeared among offspring of backcrosses of aAF hybrids with fFF earthworms. The aAF hybrids gave also progeny (aAA and aAF) in backcrosses with aAA earthworms. Despite of normal cocoon production, offspring was absent so far in hybrid-hybrid pairs and in the back-crosses of fFA with fFF specimens. Among hybrids, there are earthworms of various pigmentation patterns resulted from introgression of genes from distinctively striped E. fetida to uniformly pigmented E. andrei (Fig 5, see photos on Fig 3).
The results obtained so far suggest that EaxEf hybridization is asymmetrical. The aAF hybrids from Ea ova are relatively frequent, and they can breed with both Ea and Ef specimens, giving next generations of aAF hybrids and their progeny or serving as donor of spermatozoa to fFF partners resulting in fFA hybrids. However, the latter type of hybrids were very sparse and without offspring till the end of present investigations (Fig 5). Asymmetrical hybridization is known in other species, e.g. like that between Operophtera brumata and O. bruceata (Lepidoptera: Geometridae) from the United States [43].
Hypothetically, the more common aAF hybrids descended from the aA ova of aAA earthworms. This might be connected with better reproductive performance of E. andrei than E. fetida, recorded not only in earthworms from France used in present experiments but also in earthworms from Spain and Brasil investigated by other scientists [18,19,40]. However, dynamic of reproduction can be altered with moisture/temperature/food source [40; 41, 42]. Therefore it is possible that various factors like seasonal cyclic changes and environmental contaminants might have different effects on the two species, including influence on their hybridization abilities. Such phenomenon is known in other species, e.g. in the temporally varying hybridization of the North Atlantic eels Anguilla anguilla and A. rostrata [44] and in temperature-dependent hybridization of Drosophila melanogaster and D. stimulans [45].
All the E. fetida parents of all hybrid earthworms from present experiments, both aAF and fFA, have derived from the 'f' lineage of E. fetida/andrei phylogenetic tree, with no contribution from the 'f2' lineage. Studies on the reproductive performance of laboratory pairing of fFF and f2FF earthworms might also be revealing as to whether they are two cryptic sibling species. It is an object of research in progress.

Conclusions
Laboratory mating of virgin earthworms Ea+Ef gave both self-fertilized specimens of aAA and fFF genotypes, and some aAF hybrids derived from aA ova of Ea. These hybrids mated with the parental species gave new generation of aAA and aAF hybrids and so far sterile fFA hybrids from fF ova of the Ef partner. Pairs of two hybrids produced plenty sterile cocoons. Hypothetically, such asymmetric hybridization may be modulated by several external factors, a phenomenon worth of further inquiry.
The existence of Ea and Ef hybridization make these common species easily maintained in laboratory the attractive models for studies on mechanisms of interspecies gene flow during introgressive hybridization [46,47], phenomenon described in other lumbricids, Allolobophora sp. [48] and Lumbricus sp. [49].