268
the upper sea layers with phytoplankton, whereas nonfeeding larvae may disperse and settle much deeper. Most
Recent broadcasting bryozoans do not exist below 100–
200 m, two exceptions being Pyripora catenularia and
Electra arctica , which can be found as deep as 500–520 m
(Kluge 1975 ; Prenant and Bobin 1966 ; Hayward and Ryland
1998 ; also discussed in Taylor 1988a ). Conversely, bryozoan
brooders have even colonized the abyss down to 8300 m
( Bugula sp.; see Hayward 1981 ). The evolution of endotrophic larvae apparently revoked food restrictions, providing
bryozoans with a pass to deepwater biotopes. To note, the
early malacostegans that existed in the Late Jurassic probably inhabited shallow coastal zones (Taylor 1994 ).
Diversifi cation rates could to some extent be supported by
multiple origins of lecithotrophy. The transition to endotrophic larvae probably occurred in cheilostomes as many times
as brooding evolved (see above), each time potentially triggering speciation, although these events obviously have not
contributed signifi cantly to overall cheilostome diversity
(see Taylor 1988a ).
Yet another weighty factor ensuring successful
competition at the very beginning of the epibiotic phase of
the bryozoan life cycle is the enlargement of the ancestrula
– a result of larval metamorphosis. Greater energy input into
a single offspring should enhance its survival (Smith and
Fretwell 1974 ). In other words, larger offspring size should
considerably reduce mortality. One important conclusion
made during many studies is that most of the nutrient
resources accumulated in the egg are not used during embryonic and larval development, being reserved for peri- and
postmetamorphic periods (Emlet and Hoegh-Guldberg 1997 ;
Byrne and Cerra 2000 ; Byrne et al. 2003 ; Marshall and
Bolton 2007 ). In other words, parents provision their larvae
with more reserves than they need, thus increasing postmetamorphic performance (reviewed in Emlet et al. 1987 ).
Experiments on the removal of some lipids (50% of organic
mass) from the blastulae of the sea urchin Heliocidaris
erythrogramma have shown that embryos develop into anatomically correct but small (non-feeding) larvae as fast as
lecithotrophic larvae in the controls (Emlet and HoeghGuldberg 1997 ). The authors concluded that much of the
nutrient contained in oocytes is not used during embryogenesis and is later “placed at the disposal” of the juvenile. So,
the tendency towards increasing oocyte size is likely to be
associated with increasing viability of the young sea urchin;
enlargement of oocytes could increase survival rate of young
after larval settlement and metamorphosis.
Larval size infl uences pre- and post-metamorphic performance in cheilostome bryozoans. In Bugula species the
larger larvae swim and remain capable of metamorphosis
longer than smaller larvae (Wendt 2000 ; see also Wendt
1998 ). Field observations on Watersipora subtorquata
showed that larger larvae swim longer and are more selective
with respect to settlement substrata (Marshall and Keough
2003 ; see also Elkin and Marshall 2007 ). In this species the
larger a competent larva, the larger the juvenile (ancestrula),
and larger ancetrulae have better chances of survival (discussed in Marshall and Keough 2004a ). Further, larger
ancestrulae bud larger zooids and so develop into larger colonies. As shown in W . subtorquata , growth rate, size and survival rate of colonies are directly correlated with increasing
larval size (Marshall and Keough 2004a , 2008a ). Experiments
with Bugula neritina revealed a positive correlation between
larval size and downstream survival, growth rate, onset of
reproduction, fecundity and fi nal colony size (Marshall et al.
2003 ; Marshall and Keough 2004b , 2006 ; reviewed in
Marshall and Keough 2008b ; Marshall et al. 2008 ).
Hence, the evolution of larger eggs, and, consequently,
larvae could result in the success of the adults. McKinney
( 1992 , 1993 , 1995 ) showed that Recent Cheilostomata,
because of larger size and some morphological features, are
more effective energy consumers than Cyclostomata, also
expressed in faster growth rates of colonies (also discussed
in McKinney et al. 2001 ). As a result, beginning in the Late
Cretaceous, larger and faster-growing cheilostomes began to
dominate over cyclostomes in marine bottom communities
(Taylor and Larwood 1988 ). This dominance was expressed
as more-frequent fouling of cyclostome colonies by cheilostomes and more-numerous cheilostome colonies as compared to cyclostomes, in the same biotopes and in greater
number of cheilostome taxa. Thus, Cheilostomata was overall more competitive because cheilostome colonies were
larger. McKinney ( 1993 ) and Pachut and Fisherkeller ( 2010 )
also showed that cheilostome larvae are larger than those of
cyclostomes.
Chia ( 1974 ) noted that juveniles of marine invertebrates
developed from planktotrophic larvae are usually smaller
than those that develop from lecithotrophic ones. However,
in regard to Cheilostomata, we should not forget that exotrophic larvae enlarge considerably as they feed and grow. For
instance, in Membranipora serrilamella the diameter of ovulated oocytes is 50 μm, the width of the cyphonautes larvae
base by the time it becomes triangular is 220 μm and that of
the adult larvae is over 600 μm. The size of twinned ancestrulae in this species is 630–680 × 470–550 μm (Mawatari
and Itô 1972 ; Mawatari 1973a , 1975 ; Mawatari and Mawatari
1975 ) (see also Table 3.1 ).
To note, the size of ancestrulae may vary depending on
abundance of food. According to Cook ( 1964 ), Electra
monostachys ancestrulae formed in September were smaller
(180 × 100 μm) than those formed in July (240 × 200 μm).
It may be suggested that the evolution of Malacostega
went towards larger larvae and correspondingly larger ancestrulae. The size of ancestrulae of the earliest known cheilostome Pyriporopsis portlandensis (Tithonian, Late Jurassic)
was 240–230 × 200–170 μm. Two other malacostegans from
3 Evolution of Reproductive Patterns in Cheilostomata
the upper sea layers with phytoplankton, whereas nonfeeding larvae may disperse and settle much deeper. Most
Recent broadcasting bryozoans do not exist below 100–
200 m, two exceptions being Pyripora catenularia and
Electra arctica , which can be found as deep as 500–520 m
(Kluge 1975 ; Prenant and Bobin 1966 ; Hayward and Ryland
1998 ; also discussed in Taylor 1988a ). Conversely, bryozoan
brooders have even colonized the abyss down to 8300 m
( Bugula sp.; see Hayward 1981 ). The evolution of endotrophic larvae apparently revoked food restrictions, providing
bryozoans with a pass to deepwater biotopes. To note, the
early malacostegans that existed in the Late Jurassic probably inhabited shallow coastal zones (Taylor 1994 ).
Diversifi cation rates could to some extent be supported by
multiple origins of lecithotrophy. The transition to endotrophic larvae probably occurred in cheilostomes as many times
as brooding evolved (see above), each time potentially triggering speciation, although these events obviously have not
contributed signifi cantly to overall cheilostome diversity
(see Taylor 1988a ).
Yet another weighty factor ensuring successful
competition at the very beginning of the epibiotic phase of
the bryozoan life cycle is the enlargement of the ancestrula
– a result of larval metamorphosis. Greater energy input into
a single offspring should enhance its survival (Smith and
Fretwell 1974 ). In other words, larger offspring size should
considerably reduce mortality. One important conclusion
made during many studies is that most of the nutrient
resources accumulated in the egg are not used during embryonic and larval development, being reserved for peri- and
postmetamorphic periods (Emlet and Hoegh-Guldberg 1997 ;
Byrne and Cerra 2000 ; Byrne et al. 2003 ; Marshall and
Bolton 2007 ). In other words, parents provision their larvae
with more reserves than they need, thus increasing postmetamorphic performance (reviewed in Emlet et al. 1987 ).
Experiments on the removal of some lipids (50% of organic
mass) from the blastulae of the sea urchin Heliocidaris
erythrogramma have shown that embryos develop into anatomically correct but small (non-feeding) larvae as fast as
lecithotrophic larvae in the controls (Emlet and HoeghGuldberg 1997 ). The authors concluded that much of the
nutrient contained in oocytes is not used during embryogenesis and is later “placed at the disposal” of the juvenile. So,
the tendency towards increasing oocyte size is likely to be
associated with increasing viability of the young sea urchin;
enlargement of oocytes could increase survival rate of young
after larval settlement and metamorphosis.
Larval size infl uences pre- and post-metamorphic performance in cheilostome bryozoans. In Bugula species the
larger larvae swim and remain capable of metamorphosis
longer than smaller larvae (Wendt 2000 ; see also Wendt
1998 ). Field observations on Watersipora subtorquata
showed that larger larvae swim longer and are more selective
with respect to settlement substrata (Marshall and Keough
2003 ; see also Elkin and Marshall 2007 ). In this species the
larger a competent larva, the larger the juvenile (ancestrula),
and larger ancetrulae have better chances of survival (discussed in Marshall and Keough 2004a ). Further, larger
ancestrulae bud larger zooids and so develop into larger colonies. As shown in W . subtorquata , growth rate, size and survival rate of colonies are directly correlated with increasing
larval size (Marshall and Keough 2004a , 2008a ). Experiments
with Bugula neritina revealed a positive correlation between
larval size and downstream survival, growth rate, onset of
reproduction, fecundity and fi nal colony size (Marshall et al.
2003 ; Marshall and Keough 2004b , 2006 ; reviewed in
Marshall and Keough 2008b ; Marshall et al. 2008 ).
Hence, the evolution of larger eggs, and, consequently,
larvae could result in the success of the adults. McKinney
( 1992 , 1993 , 1995 ) showed that Recent Cheilostomata,
because of larger size and some morphological features, are
more effective energy consumers than Cyclostomata, also
expressed in faster growth rates of colonies (also discussed
in McKinney et al. 2001 ). As a result, beginning in the Late
Cretaceous, larger and faster-growing cheilostomes began to
dominate over cyclostomes in marine bottom communities
(Taylor and Larwood 1988 ). This dominance was expressed
as more-frequent fouling of cyclostome colonies by cheilostomes and more-numerous cheilostome colonies as compared to cyclostomes, in the same biotopes and in greater
number of cheilostome taxa. Thus, Cheilostomata was overall more competitive because cheilostome colonies were
larger. McKinney ( 1993 ) and Pachut and Fisherkeller ( 2010 )
also showed that cheilostome larvae are larger than those of
cyclostomes.
Chia ( 1974 ) noted that juveniles of marine invertebrates
developed from planktotrophic larvae are usually smaller
than those that develop from lecithotrophic ones. However,
in regard to Cheilostomata, we should not forget that exotrophic larvae enlarge considerably as they feed and grow. For
instance, in Membranipora serrilamella the diameter of ovulated oocytes is 50 μm, the width of the cyphonautes larvae
base by the time it becomes triangular is 220 μm and that of
the adult larvae is over 600 μm. The size of twinned ancestrulae in this species is 630–680 × 470–550 μm (Mawatari
and Itô 1972 ; Mawatari 1973a , 1975 ; Mawatari and Mawatari
1975 ) (see also Table 3.1 ).
To note, the size of ancestrulae may vary depending on
abundance of food. According to Cook ( 1964 ), Electra
monostachys ancestrulae formed in September were smaller
(180 × 100 μm) than those formed in July (240 × 200 μm).
It may be suggested that the evolution of Malacostega
went towards larger larvae and correspondingly larger ancestrulae. The size of ancestrulae of the earliest known cheilostome Pyriporopsis portlandensis (Tithonian, Late Jurassic)
was 240–230 × 200–170 μm. Two other malacostegans from
3 Evolution of Reproductive Patterns in Cheilostomata