of drumming behavior by males is seasonal and is reliably correlated with
elevated levels of both testosterone and 11-ketotestosterone (11-ketoT),
a nonaromatizable testosterone found only in teleosts (Connaughton and
Taylor 1994, 1995a). However, hormonal manipulation studies have not
been carried out to determine whether either or both of these androgens
is involved in activating the vocal behavior in this species.
A role for 11-ketoT in activating courtship vocalizations is also postulated in another sonic fish, the plainfin midshipman fish, Porichthys notatus.
Males of this species produce long-duration advertisement calls, called
hums, during the breeding season (Bass 1992; Brantley and Bass 1994). In
this species, males come in two morphs, type I and type II, that show alternative reproductive strategies. About 90% of males are type I with a large
body size; they build a nest and produce hums. About 10% of males are
type II with a smaller body size resembling females; they neither build nests
nor hum. Instead, type II males sneak into type I males’ nests along with
gravid females that are attracted to hums and sneak or satellite spawn to
inseminate eggs as they are laid (Brantley and Bass 1994). By comparing
the circulating levels of androgen in courting type I males and noncourting
type II males, it was deduced that 11-ketoT plays a role in activating
courtship vocalizations in this species. In type I males, 11-ketoT is the predominant plasma androgen, whereas testosterone is the predominant
androgen in type II males (Brantley et al. 1993a). In addition, females that
do not show courtship behavior have hormonal profiles that resemble type
II males, including elevated levels of T (a feature shared with females of
other teleost fishes). Thus, 11-ketoT, but not T, probably activates courtship
vocalizations in midshipman fish. Similar correlations are found between
elevated levels of 11-ketoT and courtship displays in six other species of
fishes with courting and noncourting male morphs (reviewed by Brantley
et al. 1993b). Although artificial manipulation of 11-ketoT to determine the
causal relation between the hormone and vocal production has not been
carried out, these reliable correlations offer support for an activational role
of 11-ketoT in the expression of courtship vocalizations in fishes.
Both weakfish and plainfin midshipman fish produce vocalizations by
contracting and relaxing sonic muscles attached to the lateral walls of
the swim bladder (Cohen and Winn 1967; Demski et al. 1973; Bass and
Marchaterre 1989). Corresponding to vocal differences between male and
female weakfish, and between the two male morphs of midshipman fish, the
muscle mass of vocalizing males is larger than that of nonvocalizing individuals even when scaled to body mass (Bass and Marchaterre 1989;
Connaughton and Taylor 1995b). In male weakfish, sonic muscle mass
changes seasonally in an androgen-dependent manner (Connaughton et al.
1997). Endogenous fluctuations in T and 11-ketoT correlate well with the
hypertrophy and atrophy of the sonic muscle (Connaughton and Taylor
1994; Connaughton et al. 1997). Exogenous T implants increase muscle mass
within 3 weeks in nonbreeding males, and continuous implants of T
6. Hormonal Control of Communication
297
elevated levels of both testosterone and 11-ketotestosterone (11-ketoT),
a nonaromatizable testosterone found only in teleosts (Connaughton and
Taylor 1994, 1995a). However, hormonal manipulation studies have not
been carried out to determine whether either or both of these androgens
is involved in activating the vocal behavior in this species.
A role for 11-ketoT in activating courtship vocalizations is also postulated in another sonic fish, the plainfin midshipman fish, Porichthys notatus.
Males of this species produce long-duration advertisement calls, called
hums, during the breeding season (Bass 1992; Brantley and Bass 1994). In
this species, males come in two morphs, type I and type II, that show alternative reproductive strategies. About 90% of males are type I with a large
body size; they build a nest and produce hums. About 10% of males are
type II with a smaller body size resembling females; they neither build nests
nor hum. Instead, type II males sneak into type I males’ nests along with
gravid females that are attracted to hums and sneak or satellite spawn to
inseminate eggs as they are laid (Brantley and Bass 1994). By comparing
the circulating levels of androgen in courting type I males and noncourting
type II males, it was deduced that 11-ketoT plays a role in activating
courtship vocalizations in this species. In type I males, 11-ketoT is the predominant plasma androgen, whereas testosterone is the predominant
androgen in type II males (Brantley et al. 1993a). In addition, females that
do not show courtship behavior have hormonal profiles that resemble type
II males, including elevated levels of T (a feature shared with females of
other teleost fishes). Thus, 11-ketoT, but not T, probably activates courtship
vocalizations in midshipman fish. Similar correlations are found between
elevated levels of 11-ketoT and courtship displays in six other species of
fishes with courting and noncourting male morphs (reviewed by Brantley
et al. 1993b). Although artificial manipulation of 11-ketoT to determine the
causal relation between the hormone and vocal production has not been
carried out, these reliable correlations offer support for an activational role
of 11-ketoT in the expression of courtship vocalizations in fishes.
Both weakfish and plainfin midshipman fish produce vocalizations by
contracting and relaxing sonic muscles attached to the lateral walls of
the swim bladder (Cohen and Winn 1967; Demski et al. 1973; Bass and
Marchaterre 1989). Corresponding to vocal differences between male and
female weakfish, and between the two male morphs of midshipman fish, the
muscle mass of vocalizing males is larger than that of nonvocalizing individuals even when scaled to body mass (Bass and Marchaterre 1989;
Connaughton and Taylor 1995b). In male weakfish, sonic muscle mass
changes seasonally in an androgen-dependent manner (Connaughton et al.
1997). Endogenous fluctuations in T and 11-ketoT correlate well with the
hypertrophy and atrophy of the sonic muscle (Connaughton and Taylor
1994; Connaughton et al. 1997). Exogenous T implants increase muscle mass
within 3 weeks in nonbreeding males, and continuous implants of T
6. Hormonal Control of Communication
297