252
viviparity” has been recorded in about 800 species of Diptera,
Dermaptera and Psocoptera, and in all aphids (Hemiptera)
(about 4,000 species) (Hagan 1951 ; Meier et al. 1999 ;
Bermingham and Wilkinson 2009 ). Considering the enormous overall number of arthropod species, these fi gures are
perhaps not so surprising. Phylum Bryozoa is much less
numerous (about 6,000 described Recent species), but the
number of species with placental analogues is of the same
order of magnitude. Placental analogues have evolved in all
bryozoan classes, including 87 known species of the freshwater class Phylactolaemata and about 850 species of Recent
stenolaemates (order Cyclostomata). My calculations, based
inter alia on the assumption that all species of Bugula ,
Watersipora , Adeonidae and Epistomiidae are matrotrophic,
indicate about 175 species for cheilostomes. Ten ctenostome
species also exhibit EEN (reviewed in Ostrovsky et al. 2008a ,
b ; 2009a ). Thus, at least a thousand bryozoan species are
placentotrophs, making this phylum the leader among all
aquatic invertebrates. Based on the distributional pattern of
EEN throughout the phylum, as well as the independent origin of embryonic incubation, matrotrophy apparently evolved
at least 22 times in Bryozoa.
3.4
Causes, Stages and Consequences
of Transition to Endotrophy
in Cheilostomata and Ctenostomata
Oviparity, external fertilization and planktotrophy are considered to be primitive characters (Jägersten 1972 ;
Strathmann 1978a , b , 1985 , 1993 ; McHugh and Rouse
1998 ). Among Bryozoa spermcasting, zygote spawning and
planktotrophy are attributes of reproductive pattern I, which
is thus thought to be the most ancient. It is logical to suggest
that the other patterns evolved on this basis, but the precise
causes of their origin remain open to debate.
Evolution of the lecithotrophic larva was most probably a
result of changes in oogenesis: an accumulation of more
nutrients in oocytes brought about a reduction in the larval
gut and numerous other changes. Thus, the new larval type
evolved during transition to the new reproductive pattern II
combining macrolecithal oogenesis and embryonic incubation. In this section I attempt to reconstruct this sequence of
events, discussing possible preconditions, causes and consequences of the origin and further evolution of new reproductive patterns in bryozoans.
3.4.1 Lecithotrophy and Brooding
The origin of lecithotrophy in Bryozoa invites a number of
intriguing questions. Why do all living bryozoans with
parental care have lecithotrophic larvae? And, by contrast,
why is there not a single example of a lecithotrophic larva in
broadcasting bryozoans? Lecithotrophic larvae develop from
macrolecithal eggs, so was the evolutionary change in oogenesis somehow connected with the origin of embryonic incubation? The origin of brooding and lecithotrophy had
dramatic consequences for phylum Bryozoa but what is the
connection between these two phenomena?
According to the mathematical model of Vance ( 1973 ),
species with numerous offspring and species with a reduced
number of young (in our case, with exo- and endotrophic
larvae) are equally successful (stable) from the evolutionary
viewpoint (see also Chia 1974 ), often coexisting in the same
biotopes. This model compares oocyte size expressed
through the amount of energy in regard to development rate
and mortality rate. According to the improved version of this
model (Christiansen and Fenchel 1979 ), the reproductive
pattern refl ects a compromise between productivity and survival rate. Prolonged existence in the water column and high
elimination rate of the larvae is compensated for by the large
number of eggs they develop from. On the other hand, a
decrease in the number of offspring should be compensated
for by a shorter free-swimming period, which in turn may be
compensated for by a higher development rate or embryonic
incubation. Data on the development rates of most endotrophic larvae of cheilostome bryozoans show that without
brooding they would spend a considerable time in the environment, which would result in higher mortality. Obviously,
the role of parental care in the survival of the young is very
important.
An overwhelming majority of invertebrates with lecithotrophic larvae brood their young (Wray 1995a ). Chia
( 1974 ) suggested that, owing to energetic constraints, small
invertebrates cannot produce enough eggs to ensure recruitment through dispersal, thus compensating for a small number of offspring by their larger size. Production of the large
oocytes is correlated with lecithotrophy and parental care
(brooding or viviparity) (see also Jablonski and Lutz 1983 ;
Valentine and Jablonski 1983 ; Olive 1985 ). Although it has
been criticized (Strathmann and Strathmann 1982 ), this
hypothesis remains rather attractive. When both the amount
of resources allocated for reproduction and the capacity of
the ovary are limited, the transition from oligolecithal to
macrolecithal oocytes (and from a planktotrophic to a lecithotrophic larva) results in larger size and a smaller quantity
of oocytes. This tendency, however, is fraught with risk. The
relationship between the productivity of an organism and the
survival rate of its offspring is a critical factor (Vance 1973 ;
Christiansen and Fenchel 1979 ; Jablonski and Lutz 1983 ).
Though a reduction in offspring number (accompanied by an
increase in nutrient reserves in each egg) in non- broadcasting
bryozoans may be in some degree compensated by the (1)
numerous reproductive zooids in a colony, (2) larval enlargement, and/or (3) shortening of larval life, these factors
3 Evolution of Reproductive Patterns in Cheilostomata
viviparity” has been recorded in about 800 species of Diptera,
Dermaptera and Psocoptera, and in all aphids (Hemiptera)
(about 4,000 species) (Hagan 1951 ; Meier et al. 1999 ;
Bermingham and Wilkinson 2009 ). Considering the enormous overall number of arthropod species, these fi gures are
perhaps not so surprising. Phylum Bryozoa is much less
numerous (about 6,000 described Recent species), but the
number of species with placental analogues is of the same
order of magnitude. Placental analogues have evolved in all
bryozoan classes, including 87 known species of the freshwater class Phylactolaemata and about 850 species of Recent
stenolaemates (order Cyclostomata). My calculations, based
inter alia on the assumption that all species of Bugula ,
Watersipora , Adeonidae and Epistomiidae are matrotrophic,
indicate about 175 species for cheilostomes. Ten ctenostome
species also exhibit EEN (reviewed in Ostrovsky et al. 2008a ,
b ; 2009a ). Thus, at least a thousand bryozoan species are
placentotrophs, making this phylum the leader among all
aquatic invertebrates. Based on the distributional pattern of
EEN throughout the phylum, as well as the independent origin of embryonic incubation, matrotrophy apparently evolved
at least 22 times in Bryozoa.
3.4
Causes, Stages and Consequences
of Transition to Endotrophy
in Cheilostomata and Ctenostomata
Oviparity, external fertilization and planktotrophy are considered to be primitive characters (Jägersten 1972 ;
Strathmann 1978a , b , 1985 , 1993 ; McHugh and Rouse
1998 ). Among Bryozoa spermcasting, zygote spawning and
planktotrophy are attributes of reproductive pattern I, which
is thus thought to be the most ancient. It is logical to suggest
that the other patterns evolved on this basis, but the precise
causes of their origin remain open to debate.
Evolution of the lecithotrophic larva was most probably a
result of changes in oogenesis: an accumulation of more
nutrients in oocytes brought about a reduction in the larval
gut and numerous other changes. Thus, the new larval type
evolved during transition to the new reproductive pattern II
combining macrolecithal oogenesis and embryonic incubation. In this section I attempt to reconstruct this sequence of
events, discussing possible preconditions, causes and consequences of the origin and further evolution of new reproductive patterns in bryozoans.
3.4.1 Lecithotrophy and Brooding
The origin of lecithotrophy in Bryozoa invites a number of
intriguing questions. Why do all living bryozoans with
parental care have lecithotrophic larvae? And, by contrast,
why is there not a single example of a lecithotrophic larva in
broadcasting bryozoans? Lecithotrophic larvae develop from
macrolecithal eggs, so was the evolutionary change in oogenesis somehow connected with the origin of embryonic incubation? The origin of brooding and lecithotrophy had
dramatic consequences for phylum Bryozoa but what is the
connection between these two phenomena?
According to the mathematical model of Vance ( 1973 ),
species with numerous offspring and species with a reduced
number of young (in our case, with exo- and endotrophic
larvae) are equally successful (stable) from the evolutionary
viewpoint (see also Chia 1974 ), often coexisting in the same
biotopes. This model compares oocyte size expressed
through the amount of energy in regard to development rate
and mortality rate. According to the improved version of this
model (Christiansen and Fenchel 1979 ), the reproductive
pattern refl ects a compromise between productivity and survival rate. Prolonged existence in the water column and high
elimination rate of the larvae is compensated for by the large
number of eggs they develop from. On the other hand, a
decrease in the number of offspring should be compensated
for by a shorter free-swimming period, which in turn may be
compensated for by a higher development rate or embryonic
incubation. Data on the development rates of most endotrophic larvae of cheilostome bryozoans show that without
brooding they would spend a considerable time in the environment, which would result in higher mortality. Obviously,
the role of parental care in the survival of the young is very
important.
An overwhelming majority of invertebrates with lecithotrophic larvae brood their young (Wray 1995a ). Chia
( 1974 ) suggested that, owing to energetic constraints, small
invertebrates cannot produce enough eggs to ensure recruitment through dispersal, thus compensating for a small number of offspring by their larger size. Production of the large
oocytes is correlated with lecithotrophy and parental care
(brooding or viviparity) (see also Jablonski and Lutz 1983 ;
Valentine and Jablonski 1983 ; Olive 1985 ). Although it has
been criticized (Strathmann and Strathmann 1982 ), this
hypothesis remains rather attractive. When both the amount
of resources allocated for reproduction and the capacity of
the ovary are limited, the transition from oligolecithal to
macrolecithal oocytes (and from a planktotrophic to a lecithotrophic larva) results in larger size and a smaller quantity
of oocytes. This tendency, however, is fraught with risk. The
relationship between the productivity of an organism and the
survival rate of its offspring is a critical factor (Vance 1973 ;
Christiansen and Fenchel 1979 ; Jablonski and Lutz 1983 ).
Though a reduction in offspring number (accompanied by an
increase in nutrient reserves in each egg) in non- broadcasting
bryozoans may be in some degree compensated by the (1)
numerous reproductive zooids in a colony, (2) larval enlargement, and/or (3) shortening of larval life, these factors
3 Evolution of Reproductive Patterns in Cheilostomata