Journal of the Royal Society of Western Australia, 86(3), September 2003 
Figure 3. Scatter plots showing differences between the three 
call types of Litoria adelaidensis in (A) call duration, (B) pulse 
number and (C) pulse rate. 
Discussion 
The slender treefrog Litoria adelaidensis has a repertoire 
of acoustic signals that includes both pulsed and 
unpulsed calls. Four call types were recognised; call 
types 1, 2 and 3 were pulsed and call type 4 was 
unpulsed. All call types had two main frequency peaks. 
The presence of two frequency peaks has been shown to 
be an important factor in female mate choice in other 
hylid species (e.g. Hyla versicolor ; Gerhardt & Huber 2002) 
and consequently may influence female mate choice in L . 
adelaidensis. This awaits further study. 
Only dominant frequency varied consistently with 
male size across all pulsed call types. A negative 
correlation between dominant frequency and body length 
is usual in frogs (Gerhardt 1994) and is probably related 
Canonical axis 1 
Figure 4. Distribution of the call types (1: solid circle, 2: open 
circle, and 3: solid square) of male Litoria adelaidensis along two 
canonical discriminant axes derived from a DFA of call 
properties on call type. The correlations between canonical axis 
1 and call duration, pulse number, pulse rate, pulse duration, 
low frequency peak and dominant frequency were -0.45, -0.50, 
-0.39, 0.06, -0.10, and -0.11 respectively. The correlations between 
canonical axis 2 and call duration, pulse number, pulse rate, 
pulse duration, low frequency peak and dominant frequency 
were 0.20, -0.67, and -0.53, 0.20, 0.13, and -0.01 respectively. 
to the size of the male's vocal chords (McAlister 1961). 
The durations of call types 1 and 2 also varied inversely 
and significantly with body size, as larger males 
produced shorter calls. Accordingly, the relationships 
between body size and both frequency and call duration 
suggest that size and possibly age (frogs generally 
increase in size with age) may influence male calling 
tactics in this species. However, the relationship between 
size and frequency does not hold for all species (e.g. 
Lardner & bin Lakim 2002) 
The pulsed calls (call types 1, 2 and 3) differed in 
temporal structure (in particular, the number of pulses 
and presence or absence of more than one note). Similar 
variation in temporal structure of different calls in the 
repertoire of several species has been reported (e.g. Jehle 
& Arak 1998; Narins et al. 2000; Brenowitz et al. 2001). 
The occurrence of significant variation in the temporal 
but not spectral structure of the calls of L. adelaidensis 
may indicate different selection regimes and/or 
differential responses to selection. For example, there 
may have been stabilizing selection on the frequency 
components of the calls, but directional selection on the 
temporal structure (cf Gerhardt 1994). 
Call type 1 may function to attract females (cf 
Littlejohn 1957; Gerhardt 1994) as this call type was 
produced more often than any other. We have no basis to 
suggest a function for call types 2 and 3, but the series of 
pulsed calls described here may represent a graded set of 
signals. Graded signals have been documented in other 
species such as Uperoleia (Robertson 1984), Rana (Jehle & 
Arak 1998) and Hyla (Schwartz 2001) and typically 
appear to represent an escalation in response to 
decreasing distance to an opponent or female (Gerhardt 
& Huber 2002). For example, there is a negative 
94 
