269
the Late Cretaceous had ancestrulae of the following size –
Spinicharixa pitti (?Aptian): 160 × 140 μm; Herpetopora
laxata (Campanian–Late Maastrichtian): 220–200 × 120–
110 μm (Taylor 1986a , b , 1988b ). Recent malacostegans
have much larger ancestrulae (Table 3.1 ) and, correspondingly, larvae, indicating a distinct evolutionary trend.
The trend towards increasing ancestrular size probably
also characterized brooding bryozoans. The size of the
ancestrulae in eight species of the earliest-known (Albian)
cheilostome genus Wilbertopora with brooding varies in the
range 260–230 × 190–150 μm (Cheetham et al. 2006 ). So,
there is no signifi cant difference in the size of ancestrulae
(and, apparently, larvae and oocytes) in the fi rst Malacostegina
(broadcasters) and the fi rst Calloporidae (brooders). This
means that the transition to a new pattern of oogenesis, a new
larval type and the origination of brooding did not result in
any signifi cant increase in ancestrular size. In the course of
further evolution oocyte size gradually increased, a trend that
in many marine invertebrates is accompanied by brooding
(Wray 1995a ).
Enlargement of oocytes inevitably affected the sizes of
larvae and ancestrulae. Since the transition to lecithotrophic
larvae seems to have required only a relatively small increase
in the amount of nutrients in oocytes (see the examples of
Tendra zostericola and Triticella fl ava ), the accumulation of
extra reserves and consequent enlargement of ancestrulae
could have been an important factor infl uencing the success
of Cheilostomata. We may also speculate that accumulation
of additional reserves in oocytes would accelerate the formation of the ancestrula and the budding of daughter zooids,
also improving the survival chances of the young colony.
Based on data in the literature, Pachut and Fisherkeller
( 2010 ) calculated the average diameter of the ancestrula in
Recent brooding cheilostomes to be 220 μm. This is slightly
more than the average size of ancestrulae in Wilbertopora
(207.5 μm) (see Cheetham et al. 2006 ). However, in order to
fi nd out whether a trend can be identifi ed using ancestrular
size, much more data are required for both Recent and fossil
bryozoans.
3.5
Evolution of Sexual Reproduction
in Bryozoa
Lecithotrophy, embryonic incubation and internal fertilization are characteristic of all three classes of phylum Bryozoa.
Loss of planktotrophy and the acquisition of parental care
occurred repeatedly within each of the two gymnolaemate
orders Ctenostomata and Cheilostomata. The fact that all
bryozoans with planktotrophic larvae have internal fertilization indicates that bryozoans acquired this fertilization mode
early in their evolutionary history or it was inherited from an
ancestor. Later transition to early intra-ovarian fertilization
occurred independently in Phylactolaemata, Cyclostomata
and Cheilostomata. Moreover, if different groups of brooding cheilostomes evolved independently from different malacostegan ancestors, this transition might have occurred
several times within this order alone. Early intraovarian fertilization presumably did not happened in Ctenostomata
since sperm has so far been found only in growing and late
oocytes.
As for the loss of planktotrophy and the evolution of
parental care, Phylactolaemata either inherited a non-feeding
short-lived larva from their marine ancestor or evolved it
independently. The recent fi nding of a cyphonautes larva in a
freshwater ctenostome of the genus Hislopia (Wood 2008 ;
Nielsen and Worsaae 2010 ) demonstrates that planktotrophic
bryozoan larvae can exist in fresh water. The reproductive
pattern of phylactolaemates combines primitive and
advanced characters – numerous small oocytes (20–40
according to Wood ( 1983 ) and up to 42 in Lophopus crystallinus , 25 μm in diameter; see Marcus 1934 ), placental brooding (which phylactolaemates evolved independently),
intraovarian fertilization and putative nurse cells (in
Lophopus ). A very similar combination of characters is
found in the “protoctenostome” Labiostomella gisleni (Silén
1944 ). In both Phylactolaemata and L . gisleni numerous
small oocytes are formed in a maternal zooid but only one of
them develops into a larva in the brood sac with extraembryonic nutrition. The larva L . gisleni is unknown but we may be
fairly sure that it is endotrophic.
In Phylactolaemata brooding could have originated either
in the early phylactolaemates or in their marine ancestor.
Oocyte transfer into the brood sac, bypassing the environment, which is characteristic of Phylactolaemata (see Brien
1953 ), has not been found in any marine bryozoan. The
brood sacs of phylactolaemates are formed on the oral side
of the zooid, while in gymnolaemates they are formed on the
anal side (Jebram 1973 ). Thus, these structures, although
both being invaginations of the body wall, are not homologous. This means that Phylactolaemata evolved brooding
independently. Invagination of the cystid wall, which is triggered by the adhesion of the released oocyte in Ctenostomata,
could be triggered by the ovary, which always closely adjoins
the brood sac in Phylactolaemata. At the same time, this
invagination could have originally appeared in connection
with external brooding, which later was substituted by the
internal mode.
In summary, the reproductive features of Phylactolaemata
generally correspond to pattern III as described for the
ctenostomes L . gisleni and Nolella dilatata . Although a
fertile phylactolaemate zooid broods one embryo at a time,
the number of oocytes in the ovary remains large. This pattern might have evolved on the basis of pattern II (as
described for ctenostomes, see Sect. 3.4.4 ) in connection
with the acquisition of the placental analogue. Importantly,
3.5 Evolution of Sexual Reproduction in Bryozoa
the Late Cretaceous had ancestrulae of the following size –
Spinicharixa pitti (?Aptian): 160 × 140 μm; Herpetopora
laxata (Campanian–Late Maastrichtian): 220–200 × 120–
110 μm (Taylor 1986a , b , 1988b ). Recent malacostegans
have much larger ancestrulae (Table 3.1 ) and, correspondingly, larvae, indicating a distinct evolutionary trend.
The trend towards increasing ancestrular size probably
also characterized brooding bryozoans. The size of the
ancestrulae in eight species of the earliest-known (Albian)
cheilostome genus Wilbertopora with brooding varies in the
range 260–230 × 190–150 μm (Cheetham et al. 2006 ). So,
there is no signifi cant difference in the size of ancestrulae
(and, apparently, larvae and oocytes) in the fi rst Malacostegina
(broadcasters) and the fi rst Calloporidae (brooders). This
means that the transition to a new pattern of oogenesis, a new
larval type and the origination of brooding did not result in
any signifi cant increase in ancestrular size. In the course of
further evolution oocyte size gradually increased, a trend that
in many marine invertebrates is accompanied by brooding
(Wray 1995a ).
Enlargement of oocytes inevitably affected the sizes of
larvae and ancestrulae. Since the transition to lecithotrophic
larvae seems to have required only a relatively small increase
in the amount of nutrients in oocytes (see the examples of
Tendra zostericola and Triticella fl ava ), the accumulation of
extra reserves and consequent enlargement of ancestrulae
could have been an important factor infl uencing the success
of Cheilostomata. We may also speculate that accumulation
of additional reserves in oocytes would accelerate the formation of the ancestrula and the budding of daughter zooids,
also improving the survival chances of the young colony.
Based on data in the literature, Pachut and Fisherkeller
( 2010 ) calculated the average diameter of the ancestrula in
Recent brooding cheilostomes to be 220 μm. This is slightly
more than the average size of ancestrulae in Wilbertopora
(207.5 μm) (see Cheetham et al. 2006 ). However, in order to
fi nd out whether a trend can be identifi ed using ancestrular
size, much more data are required for both Recent and fossil
bryozoans.
3.5
Evolution of Sexual Reproduction
in Bryozoa
Lecithotrophy, embryonic incubation and internal fertilization are characteristic of all three classes of phylum Bryozoa.
Loss of planktotrophy and the acquisition of parental care
occurred repeatedly within each of the two gymnolaemate
orders Ctenostomata and Cheilostomata. The fact that all
bryozoans with planktotrophic larvae have internal fertilization indicates that bryozoans acquired this fertilization mode
early in their evolutionary history or it was inherited from an
ancestor. Later transition to early intra-ovarian fertilization
occurred independently in Phylactolaemata, Cyclostomata
and Cheilostomata. Moreover, if different groups of brooding cheilostomes evolved independently from different malacostegan ancestors, this transition might have occurred
several times within this order alone. Early intraovarian fertilization presumably did not happened in Ctenostomata
since sperm has so far been found only in growing and late
oocytes.
As for the loss of planktotrophy and the evolution of
parental care, Phylactolaemata either inherited a non-feeding
short-lived larva from their marine ancestor or evolved it
independently. The recent fi nding of a cyphonautes larva in a
freshwater ctenostome of the genus Hislopia (Wood 2008 ;
Nielsen and Worsaae 2010 ) demonstrates that planktotrophic
bryozoan larvae can exist in fresh water. The reproductive
pattern of phylactolaemates combines primitive and
advanced characters – numerous small oocytes (20–40
according to Wood ( 1983 ) and up to 42 in Lophopus crystallinus , 25 μm in diameter; see Marcus 1934 ), placental brooding (which phylactolaemates evolved independently),
intraovarian fertilization and putative nurse cells (in
Lophopus ). A very similar combination of characters is
found in the “protoctenostome” Labiostomella gisleni (Silén
1944 ). In both Phylactolaemata and L . gisleni numerous
small oocytes are formed in a maternal zooid but only one of
them develops into a larva in the brood sac with extraembryonic nutrition. The larva L . gisleni is unknown but we may be
fairly sure that it is endotrophic.
In Phylactolaemata brooding could have originated either
in the early phylactolaemates or in their marine ancestor.
Oocyte transfer into the brood sac, bypassing the environment, which is characteristic of Phylactolaemata (see Brien
1953 ), has not been found in any marine bryozoan. The
brood sacs of phylactolaemates are formed on the oral side
of the zooid, while in gymnolaemates they are formed on the
anal side (Jebram 1973 ). Thus, these structures, although
both being invaginations of the body wall, are not homologous. This means that Phylactolaemata evolved brooding
independently. Invagination of the cystid wall, which is triggered by the adhesion of the released oocyte in Ctenostomata,
could be triggered by the ovary, which always closely adjoins
the brood sac in Phylactolaemata. At the same time, this
invagination could have originally appeared in connection
with external brooding, which later was substituted by the
internal mode.
In summary, the reproductive features of Phylactolaemata
generally correspond to pattern III as described for the
ctenostomes L . gisleni and Nolella dilatata . Although a
fertile phylactolaemate zooid broods one embryo at a time,
the number of oocytes in the ovary remains large. This pattern might have evolved on the basis of pattern II (as
described for ctenostomes, see Sect. 3.4.4 ) in connection
with the acquisition of the placental analogue. Importantly,
3.5 Evolution of Sexual Reproduction in Bryozoa