181
et  al. 2018, 2019; Pfeiffer et  al. 2019). Thus, presumably one may assume, that
stem-group unionoids, or even their marine/brackish waters ancestors, had a character which eased acquiring a parasitic relation with fish after they invaded freshwaters, which led to the independent development of recent lasidia and glochidia larvae.
On the other hand, numerous previous morphological and molecular phylogenies
have consistently supported that glochidia-bearing mussels are  paraphyletic and
that glochidia are ancestral larvae for lasidia (e.g. Bogan and Hoeh 2000; Hoeh
et al. 2001; Graf and Cummings 2006; Whelan et al. 2011; Graf et al. 2015; Pfeiffer
and Graf 2015; Bolotov et al. 2016, 2017a; Lopes-Lima et al. 2017; Huang et al.
2018, 2019). Assuming that a parasitic relation requires a unique set of characters,
one may hypothesize that unionoids obtained this ability once and early in their
evolution, followed by this ability becoming ubiquitous and a selective force shaping their further evolution (e.g. Watters 2001; Barnhart et al. 2008)—then modification of the larva type appeared, which might support the latter hypothesis.
Unionoids are today typically large and heavy with limited mobility, thus their
dispersal relies on their minute larvae that attach to the fish (Schwalb and Push 2007;
Schwalb et al. 2011b; Graf 2013; Patterson 2018). If the size of an adult is a sufficient clue for a phoretic relationship similarly large mussels (about at least 8 cm in
length), members of the genus Tihkia, were discovered by the end of Triassic:
Carnian of Manda Beds of Tanzania and Ntawere Formation of Zambia (Cox 1969a;
Peecook et al. 2017), Carnian/Norian of Maleri Formation of the Hyderabad State
and the region of Tihki, Vindhya Pradesh, India (Sahni and Tewari 1958); early
Rhaetian of the Upper Silesia, Poland (Skawina and Dzik 2011). On the other hand,
Bauer (2001d) suggested the possibility that the development of parasitism, with the
cost of the high mortality of larvae that occurs during transition to the host (Jansen
et al. 2001), might have been a selective factor for a larger size (and thus fertility).
Watters (2001) suggested that ancestral unionoids probably had fewer, large larvae, similar to larvae of the extant Sphaeridae. Findings of Tevesz (1975) may support this proposition as the trigoniid Neotrigonia gemma, sectioned by him, had
“large, white, irregularshaped eggs”, but exhibited presumably external fertilization. The ancestral larvae of unionoids—presumably a kind of veliger—might have
been released either freely, or entangled in mucous packages. Larvae might be predated by fish and this could have been the start of the contact with the fish body, as
today sticklebacks (Gasterosteus aculeatus) actively feed on large, individual glochidia of Anodonta cygnea (Dartnall and Walkey 1979). Nevertheless this is a rare
observation—glochidia are usually not a part of a fish diet today (Jansen et al. 2001).
Glochidia of several extant species have adhesive larval threads, which can act as
an attachment organ (Wood 1974b). Wächtler et al. (2001) report that in European
species this feature varies in closely related species like Anodonta anatina, A. cygnea (present) and Pseudoanodonta complanata (absent), or Unio tumidus, U. pictorum (present) and U. crassus (absent). It is also absent in Margaritifera margaritifera.
They presume it is related to the lentic environments (streams, rivers). Nevertheless
threads of many glochidia are capable of forming a floating web, increasing their
chances of host infestation and limiting the possibility of larvae sinking to the sediments (Wood 1974b). Barnhart et al. (2008) and Haag (2012) found this structure in
many species within family Unionidae, but not in tribe Lampsilini—a group that
5 Evolutionary History of Bivalves as Parasites
Précédent

- 192/571

Suivant