80
male mortality rates relative to hermaphrodites within this study along with earlier field reports (Strenth,
1977) and general laboratory observations (Sassaman
& Weeks, 1993) support this prediction. Secondly, Otto
et al. (1993) predict that, unless inbreeding depression is quite high, levels of outcrossing must be high
when males are present. In support of this prediction,
although selfing was seen during these observations,
it was comparatively rare (11 %) compared to outcrossing (53% of observations). The close proximity
of males and hermaphrodites in this artificial situation
may, however, have contributed to the high level of outcrossing observed. The model also predicts that when
males are absent, large numbers of eggs will go unfertilized. The current study found that hermaphrodites
are gravid for a large percentage of their mature life
span (72%) even as male numbers dwindle in the population. In order for the predictions to be correct, many
of these selfed eggs must be infertile. Future studies
should be designed to ascertain the relative fecundity
of selfed versus outcrossed clutches.
During the instantaneous scan sampling, hermaphrodites were never seen dropping their clutches but there
is some suggestion that they may deposit clutches within the burrows they dig. On several occasions during
the initial observations, gravid hermaphrodites were
seen digging a burrow and then emerging from it without their clutch. Non-gravid hermaphrodites and males
were never seen digging burrows. During the focal animal observations, no substrate was provided so that all
interactions between the clam shrimp would be visible
and eggs could be collected. Absence of a substrate precluded observations of any digging behavior. During
the scan sample observations, males were seen entering
burrows. If hermaphrodites drop their clutch within a
burrow, it seems likely that males might enter burrows
in search of non-gravid hermaphrodites to mate guard
prior to the hermaphrodites, next clutch.
Outcrossing behavior
This study provides the first detailed documentation
of the sequence of events associated with outcrossing
for E. texana. Molting appears to be necessary before
fertilization is possible. This is a common occurrence
in crustaceans (Ward, 1983). In many crustaceans, the
correlation between molting and fertilization has limited the receptivity period of females and it has been
suggested that this may lead to male/male competition
for mates and mate guarding (Elwood et aI., 1987;
Anstensrud, 1992). The male's clasping and swimming with the hermaphrodite for extended periods suggests that precopulatory mate guarding exists within
this species. Male clasping of gravid hermaphrodites
provides further evidence for the occurrence of precopulatory mate guarding. In studies of other small
crustacean species, mate guarding carries a variety of
costs including high energy requirements, an increase
in susceptibility to predation, and loss of time that
could be spent searching for other females (Elwood &
Dick, 1990). There may be similar costs related to mate
guarding in E. texana, particularly relating to increases
in energy demands since males push hermaphrodites
around and fight off other males (see Knoll & Zucker,
1994).
The E. texana mating system provides an ideal opportunity to study the relative costs of sex and
selfing. It is well known that outcrossing incurs the
costs of sex. The maintenance of males in most
populations of this species suggests that there must
be costs to selfing as well. My observations suggest that the hermaphrodite can control the mode
of fertilization (selfing vs outcrossing). Outcrossing
was never observed between a male and a struggling
hermaphrodite. There were instances in which the
struggling hermaphrodite was much smaller than the
clasping male and could not have physically forced
him to release his grasp (pers. obs.). Instead, it
appeared that males released the hermaphrodite upon
its struggling. My results also suggest that there
are time costs incurred by hermaphrodites that self.
Once a hermaphrodite let a male clasp for more
than one minute, it was committed to outcrossing
which occurred an average of 27 minutes later. A
hermaphrodite that selfed, however, remained motionless an average of 65 minutes prior to fertilization.
Thus outcrossing occurred in less than half the time on
average than selfing.
During careful examination of mating behavior, I
failed to observe any indication of a spermatophoretype structure being transferred to the hermaphrodite.
A previous study suggested that males transfer such
a spermatophore-like package to the eleventh thoracic segment of the hermaphrodite during outcrossing (Strenth, 1977). If sperm were transferred in a
spermatophore-like form, it may have been too small
to be visible during these observations.
The unique mating system of Eulimnadia texana
and its close relatives provides the potential for answering many intriguing questions on the evolution of sex.
Toward that end, this contribution provides the first
detailed description of the behavior of these organisms
male mortality rates relative to hermaphrodites within this study along with earlier field reports (Strenth,
1977) and general laboratory observations (Sassaman
& Weeks, 1993) support this prediction. Secondly, Otto
et al. (1993) predict that, unless inbreeding depression is quite high, levels of outcrossing must be high
when males are present. In support of this prediction,
although selfing was seen during these observations,
it was comparatively rare (11 %) compared to outcrossing (53% of observations). The close proximity
of males and hermaphrodites in this artificial situation
may, however, have contributed to the high level of outcrossing observed. The model also predicts that when
males are absent, large numbers of eggs will go unfertilized. The current study found that hermaphrodites
are gravid for a large percentage of their mature life
span (72%) even as male numbers dwindle in the population. In order for the predictions to be correct, many
of these selfed eggs must be infertile. Future studies
should be designed to ascertain the relative fecundity
of selfed versus outcrossed clutches.
During the instantaneous scan sampling, hermaphrodites were never seen dropping their clutches but there
is some suggestion that they may deposit clutches within the burrows they dig. On several occasions during
the initial observations, gravid hermaphrodites were
seen digging a burrow and then emerging from it without their clutch. Non-gravid hermaphrodites and males
were never seen digging burrows. During the focal animal observations, no substrate was provided so that all
interactions between the clam shrimp would be visible
and eggs could be collected. Absence of a substrate precluded observations of any digging behavior. During
the scan sample observations, males were seen entering
burrows. If hermaphrodites drop their clutch within a
burrow, it seems likely that males might enter burrows
in search of non-gravid hermaphrodites to mate guard
prior to the hermaphrodites, next clutch.
Outcrossing behavior
This study provides the first detailed documentation
of the sequence of events associated with outcrossing
for E. texana. Molting appears to be necessary before
fertilization is possible. This is a common occurrence
in crustaceans (Ward, 1983). In many crustaceans, the
correlation between molting and fertilization has limited the receptivity period of females and it has been
suggested that this may lead to male/male competition
for mates and mate guarding (Elwood et aI., 1987;
Anstensrud, 1992). The male's clasping and swimming with the hermaphrodite for extended periods suggests that precopulatory mate guarding exists within
this species. Male clasping of gravid hermaphrodites
provides further evidence for the occurrence of precopulatory mate guarding. In studies of other small
crustacean species, mate guarding carries a variety of
costs including high energy requirements, an increase
in susceptibility to predation, and loss of time that
could be spent searching for other females (Elwood &
Dick, 1990). There may be similar costs related to mate
guarding in E. texana, particularly relating to increases
in energy demands since males push hermaphrodites
around and fight off other males (see Knoll & Zucker,
1994).
The E. texana mating system provides an ideal opportunity to study the relative costs of sex and
selfing. It is well known that outcrossing incurs the
costs of sex. The maintenance of males in most
populations of this species suggests that there must
be costs to selfing as well. My observations suggest that the hermaphrodite can control the mode
of fertilization (selfing vs outcrossing). Outcrossing
was never observed between a male and a struggling
hermaphrodite. There were instances in which the
struggling hermaphrodite was much smaller than the
clasping male and could not have physically forced
him to release his grasp (pers. obs.). Instead, it
appeared that males released the hermaphrodite upon
its struggling. My results also suggest that there
are time costs incurred by hermaphrodites that self.
Once a hermaphrodite let a male clasp for more
than one minute, it was committed to outcrossing
which occurred an average of 27 minutes later. A
hermaphrodite that selfed, however, remained motionless an average of 65 minutes prior to fertilization.
Thus outcrossing occurred in less than half the time on
average than selfing.
During careful examination of mating behavior, I
failed to observe any indication of a spermatophoretype structure being transferred to the hermaphrodite.
A previous study suggested that males transfer such
a spermatophore-like package to the eleventh thoracic segment of the hermaphrodite during outcrossing (Strenth, 1977). If sperm were transferred in a
spermatophore-like form, it may have been too small
to be visible during these observations.
The unique mating system of Eulimnadia texana
and its close relatives provides the potential for answering many intriguing questions on the evolution of sex.
Toward that end, this contribution provides the first
detailed description of the behavior of these organisms